JP4212632B2 - Road information providing system and apparatus and road information generation method - Google Patents

Description

  The present invention relates to a method for generating traffic information representing traffic conditions, travel time, and the like, a system for providing the traffic information, and a device constituting the system, and more particularly, new traffic information such as traffic conditions and travel time. It is made into data in a form, and road information with abundant information content can be provided efficiently.

<Current VICS traffic information>
VICS (Road Traffic Information Communication System), which currently provides road traffic information services for car navigation systems, collects and edits road traffic information, and travels that express traffic information and travel time through FM multiplex broadcasting and beacons. Traffic congestion information such as time information is transmitted.
In the current VICS information, current traffic information is expressed as follows.
Traffic congestion is congested (general road: ≤ 10 km / h, expressway: ≤ 20 km / h), congested (general road: 10-20 km / h, expressway: 20-40 km / h), light (general road) : ≧ 20 km / h ・ Expressway: ≧ 40 km / h), and is displayed in three stages.
The traffic information indicating the traffic situation is the same as the VICS link (location information identifier used in VICS) in the same congestion situation.
"VICS link number + status (congested / congested / disturbed / unknown)"
Is displayed, and only a part of the link is congested,
“VICS link number + traffic jam head distance (distance from link start end) + traffic jam tail distance (distance from link start end) + status (traffic jam)”
Is displayed. In this case, when the traffic jam starts from the beginning of the link, the traffic jam head distance is displayed as 0xff. When different congestion states coexist in the link, each congestion situation is described by this method.
In addition, link travel time information showing the travel time of each link,
"VICS link number + travel time"
Is displayed.
In addition, an increase / decrease trend flag indicating four states of “increase / decrease / no change / unknown” is displayed as current information as prediction information indicating a future change trend of traffic conditions.

<Communication of road position independent of VICS link number>
In the VICS traffic information, a road is identified by a link number and the traffic information is displayed, and the receiving side of this traffic information grasps the traffic situation of the corresponding road in its own map based on the link number. However, the method in which the sender and receiver share the link number and node number to identify the position on the map requires that the link number and node number be newly established or modified each time a road is newly established or changed. Along with this, the data of each company's digital map must be updated, which requires a great social cost for maintenance.

  In order to improve such points, the inventors of the present invention, in the following Patent Document 1, etc., the transmission side arbitrarily set a plurality of nodes on the road shape, and the position of this node is represented by a data string A method has been proposed in which a “shape vector data string” is transmitted, and the receiving side performs map matching using the shape vector data string to identify a road on a digital map. Further, in order to reduce the data amount of the shape vector data string, a method of compressing data by Fourier coefficient approximation is described in Patent Document 2 below, and this data is subjected to statistical processing to data concentrated around ± 0. Japanese Patent Application No. 2001-134318 proposes a method of compressing data by variable length coding after conversion.

Further, as a technique for correcting the shape vector displayed at the relative position, the following technique is possible. When encoding is performed by relatively displaying the position of the node included in the shape vector, an accumulated error is accumulated. This accumulated error is likely to accumulate when the shape vector has a long distance and has a “smooth shape” such as the national road 246 or the national road 1. In order to prevent this, as a shape vector, as shown by a thick line in FIG. 40, the shape is extracted once bent at an intersecting road or the like and then returned to the main line. “Point” is set as the reference node to cancel the accumulated error. On the receiving side, the relative position is corrected by comparing the distance between the reference nodes of the shape vector indicated by the dotted line obtained by decoding the received data with the distance between the reference nodes of the shape vector indicated by the thick line. . The reference node selected and installed at such a position that can correct the accumulated error is hereinafter referred to as a “relative position correction reference node”.
By such a system, it is possible to transmit the road position without using a link number or a node number.

JP 2001-41757 A

JP 2002-228467 A

However, currently provided traffic information has the following problems, and flexibly responds to various requests for road information, that is, traffic information, information along the road, and traffic information along the road. I can't.
<Current traffic information issue 1>
Current traffic information has extreme resolution in information representation. Congestion information can be displayed in 10m increments with respect to its location, but the traffic information is expressed in only three states: congestion, congestion, and light.
In addition, although the traffic information related to the link travel time can be expressed in units of 10 seconds, the position resolution is only “link units”, and it is not possible to express the detailed speed distribution in the link.

As a result, the following problems may arise.
As shown in FIG. 41, when a person looks at the display of a traffic jam section (a section of 10 km / h or less) on the link A of the traffic information provision route, the traffic jam is only 500 m, so it will not take much time. I passed through the section, but the car got stuck and did not move at all, and it took 25 minutes to pass through this 500m traffic jam.
In addition, a person saw the display of "Link A travel time = 30 minutes" and thought that Link A would take time, so he avoided this and went around the detour of the traffic information provision route over 23 minutes. Link A actually takes time only in the traffic jam part nearest to the intersection (25 minutes), and it can pass in 5 minutes other than the traffic jam part, and the dotted line of the traffic information non-providing route displayed in the navigation If you use the road, you will be able to pass through in about 7 minutes.
As shown in FIG. 42, when the traffic information is located in a graph in which the vertical axis represents the number of states (traffic expression resolution) that can represent traffic information and the horizontal axis represents position (or section) resolution, the link travel is performed. Although time has a high traffic expression resolution, the position resolution is low, and the traffic jam information has a high position resolution but a low traffic expression resolution.
In the current traffic jam information and link travel time information, an intermediate resolution cannot be expressed as shown by a circle in FIG.

On the other hand, traffic information can be collected within this circle, and in the case of a probe car that collects data from actual traveling vehicles, this circle can be collected according to the purpose of information collection and the amount of data to be transmitted. The information at each level can be collected at the center (for example, if the speed is measured up to 120 km / h in units of 3 km / h every 300 m, the position resolution is 300 m and the state number resolution is 40). In addition, the original information collected by existing sensors before editing is such an intermediate level of traffic information, depending on the degree of sensor density.
Ideally, a traffic information expression method that can express everything on the graph of FIG. 42 and can arbitrarily change both the position resolution and the traffic expression resolution in accordance with the source data is desirable.

<Current traffic information issue 2>
In the current traffic information providing method, the position resolution and the traffic expression resolution are fixed, and therefore, when the amount of data is large, the transmission path capacity is exceeded as shown in FIG. In this case, data exceeding the transmission path capacity is lost, and even if the importance of this data is high, it is not transmitted to the receiving side.
Ideally, as shown in FIG. 43 (b), when the amount of data is likely to exceed the transmission path capacity, the data is not lost, but the resolution is “roughened” in order from the information with the lower importance. It is desirable to reduce the amount of data.
That is, as shown in FIG. 44 (a), when there is a margin in the transmission path, the traffic information is expressed with high position resolution and traffic expression resolution, and when the amount of information increases near the transmission path capacity, FIG. As shown in (b), the position resolution related to information on routes with low importance is reduced, the traffic expression resolution related to information on routes far from the information providing point is reduced, and the position resolution related to prediction information in the far future. It is also desirable to reduce the amount of information by reducing the traffic expression resolution, and to keep displaying information on the most recent important routes at a high resolution.

<Current traffic information issue 3>
The present traffic information expression format is not suitable for traffic prediction information.
Various traffic prediction methods have been developed, including simulation methods. In addition, with the development of future traffic information providers, the provision of traffic prediction information services is expected to flourish.
However, the current traffic information only provides data indicating “increase / decrease” as prediction information. When trying to send traffic information prediction information in the current traffic information representation format, the amount of data increases in proportion to the number of predicted time zones. On the other hand, when the traffic situation is observed, if a certain time zone is a traffic jam, there are many cases where the next time zone is also a traffic jam, and therefore, data is sent in duplicate, which is inefficient.

<Object of the present invention>
The present invention solves such problems in conventional traffic information, and can arbitrarily set position resolution and traffic expression resolution, and change position resolution and traffic expression resolution as needed according to the importance of information. A traffic information generation method capable of flexibly responding to a “prediction service” that is expected to occur in the future, a system for providing the traffic information, and an apparatus constituting the system And is intended to provide.

The road information providing device of the present invention has a sampling point setting unit that sets sampling points at equal intervals along the road from a reference point in the road section specified by the road section reference data. The traffic information state quantity that changes along the road is calculated for each sampling point based on the traffic information state quantity measured at the measurement point, and the traffic information at each sampling point is calculated. Compress the state quantity.

The traffic information providing apparatus of the present invention further includes a sampling point setting unit that sets sampling points by dividing a road section specified by road section reference data at equal intervals along the road, The traffic information state quantity that changes along the road is calculated for each sampling point based on the traffic information state quantity measured at the measurement point, and the traffic information at each sampling point is calculated. Compress the state quantity.

  The state quantity of the traffic information that changes along the road includes any one of the traveling speed of the vehicle, travel time, and traffic jam information.

  Further, the road information providing apparatus of the present invention converts the state quantity of the traffic information at each sampling point into a coefficient value of a frequency component by performing orthogonal transformation, and compresses it.

  Moreover, the road information provision apparatus of this invention compresses the state quantity of the said traffic information by deleting the value of a high frequency component.

  The road information providing device is a probe car on-board terminal device that provides measurement information measured during traveling or a navigation terminal device mounted on a vehicle.

  The road information providing device may be a center side device that provides traffic information.

Furthermore, the traffic information providing method of the present invention includes a step of setting sampling points at equal intervals along the road from a reference point in the road section specified by the road section reference data, and along the road in the road section. The step of calculating the state quantity of the changing traffic information for each sampling point based on the state quantity of the traffic information measured at the measurement point, and the traffic information state quantity at each sampling point calculated A road information providing method comprising the step of compressing.

Further, the traffic information providing method of the present invention includes a step of dividing the road section specified by the road section reference data into equal intervals along the road to set sampling points, and a change along the road in the road section. Calculating the state quantity of traffic information to be calculated for each sampling point based on the state quantity of traffic information measured at the measurement point, and compressing the state quantity of traffic information at each sampled point calculated And a step of.

  A program for causing a computer to execute each step of the road information providing method is also included in the present invention.

  A road information utilization apparatus that reproduces a state quantity of traffic information transmitted from the road information providing apparatus on a digital map is also included in the present invention.

  In the road information providing system of the present invention, the position resolution and the traffic expression resolution can be arbitrarily set, and the information expression resolution can be changed as needed according to the importance of the road information. In addition, it is possible to flexibly deal with a “prediction service” of road information.

In the present invention, “travel time information” and “congestion information”, which are “traffic congestion indicators”, are expressed in different formats (that is, the number of position resolution and state number resolution), but traffic based on the vehicle travel speed. There is no difference in that it is information and traffic information that changes continuously along the road, and both are considered to be essentially the same.
And in this invention, the traffic information which changes continuously along a road (target road) is caught as a function of the distance (length) from the reference node defined in the start end of the target road or the target road section. . The traffic information provider provides this function or the coefficient of the function and the road segment reference data for identifying the target road to the receiver, and the receiver reproduces the function from the received information and uses the road segment reference data as the target. A road is identified, and traffic information that continuously changes along the target road is reproduced.
By doing so, the “travel time information” and the “traffic jam information” can be expressed and transmitted in a unified manner with the same idea and the same rule.

In order to functionalize the traffic information, in the embodiment of the present invention, the reference nodes of the target road (shape vector) that provides the traffic information are resampled at equal intervals and sampled in the distance direction. A value of travel speed (or travel time, traffic jam information) is obtained, and traffic information is expressed by a data string of this value.
As described later, various data can be used as the road section reference data. However, in the following embodiment, a case will be described in which a target road is specified using a shape vector indicating a road shape.

(First embodiment)
<Traffic information generation method>
In the first embodiment of the present invention, a case will be described in which the state information of traffic information (congestion information and travel time information) at each sampling point is quantized and then variable-length coded to generate provision information. .
In FIG. 2, traffic information (congestion information and travel time information) is schematically represented by a graph in which the horizontal axis represents the distance on the shape vector and the vertical axis represents the time axis. One square on the horizontal axis indicates the unit partition length of the quantization unit (distance direction quantization unit) set by sampling, and one square on the vertical axis is divided at regular time intervals. In each frame of the graph, traveling speed information corresponding to the distance from the reference node (start point) and the elapsed time from the current time is recorded. A relative position correction reference node is set as the reference node on the horizontal axis.

In FIG. 2, for the sake of convenience, this graph is divided into (a) a region where the traveling speed is ranked as a traffic jam (general road: ≦ 10 km / h / highway: ≦ 20 km / h), (General road: 10 to 20 km / h / highway: 20 to 40 km / h), (c) An area that is ranked in a quiet manner (general road: ≧ 20 km / h, highway: ≧ 40 km / h), ( d) It is divided into unknown areas.
When the traffic information is displayed as shown in FIG. 2, the actual traffic obtained from past observations of traffic conditions can be expressed as the following correlation law indicating the correlation between frames.
Correlation law A: Correlation between frames that are adjacent in the distance direction is high (if a point is a traffic jam, the adjacent points are also a traffic jam) ((1) in FIG. 2)
Correlation law B: Correlation between frames adjacent in the time direction is high (if there is a traffic jam at a certain time, the time before and after the traffic jam) ((2) in FIG. 2)
Correlation law C: Correlation with changes in the time direction is high (when starting to be crowded, all roads are crowded almost simultaneously) ((3) in FIG. 2)
Correlation law D: Congestion speed at the beginning of the bottleneck intersection (intersection where traffic starts, “Ayase Bus Stop”, etc.) and reverse propagation speed of traffic congestion on highways are almost constant (4 in FIG. 2) )
When the traffic information is variable-length encoded, the amount of data can be reduced by using such a correlation law.

FIG. 1 shows a process of performing statistical processing on traffic information data (state quantity) and converting this data into data concentrated around ± 0 when the traffic information at the current time is variable-length encoded. Yes.
First, as shown in FIG. 1 (a), the shape vector of the distance Xm is sampled at equal intervals from the reference node by the length of the unit partition length (eg, 50 to 500m), and is shown in FIG. 1 (b). Thus, the average speed of the vehicle passing through each sampling point is obtained. In FIG. 1B, the obtained velocity value is shown in the frame representing the quantization unit set by sampling. In this case, instead of the average speed, the average travel time and the traffic congestion rank of the vehicle passing through the sampling point interval may be obtained.

Next, this speed value is converted into a quantization amount using the traffic information quantization table of FIG. In this traffic information quantization table, since the user is requesting detailed information at the time of traffic jam, when the speed is less than 10 km / h, the quantization amount increases in increments of 1 km / h, and the speed is 10 to 19 km / h. In the range of h, the amount of quantization increases in increments of 2 km / h, in the range of 20 to 49 km / h, the amount of quantization increases in increments of 5 km / h, and in the range of speeds of 50 km / h or more, 10 km. The quantization amount is set to increase in increments of / h. A value quantized using this traffic information quantization table is shown in FIG.
Next, the quantized value is expressed by a difference from the statistical prediction value. Here, with respect to the quantized speed Vn of the quantization unit of interest, a difference is calculated by (Vn−Vn−1) with the quantized speed Vn−1 of the upstream quantization unit as the statistical prediction value S. . The calculation result is shown in FIG.

Thus, when the quantized speed is expressed as a difference from the statistical prediction value, the frequency of occurrence of values around ± 0 is determined by the correlation law A (the traffic conditions of adjacent quantization units are similar). Get higher.
Variable length coding is performed on the data subjected to such processing. The variable length encoding process is the same as that shown in Japanese Patent Application No. 2001-134318.
That is, the past traffic information is analyzed, and a code table for encoding the statistical prediction value difference of the traffic information as shown in FIG. 4 is created, and the values shown in FIG. Is encoded. For example, +2 is encoded as “1111000” and −2 is encoded as “1111001”. In addition, when 0 continues like 00000, it is encoded as “100”.
In addition, when the length of the unit block length is switched from a certain point, the code table includes a block length change code for indicating a change in the unit block length from the point, and a traffic information quantization table ( 3), the traffic information quantization table change code for indicating the change of the traffic information quantization table from that point, and the reference node corresponding point identification code for indicating the reference node are special codes. Is set.

In this way, traffic information is quantized, converted into statistical prediction difference values, and the occurrence frequency of values around ± 0 is increased, so that variable length coding (Huffman / arithmetic code / Shannon Fano etc.) and run length The effect of data compression by compression (run length coding) is improved. In particular, when the traffic jam information is displayed in four ranks as in the prior art, the statistical prediction value difference in many quantization units is 0, so the effect of continuous length compression becomes extremely high.
In addition, with regard to travel time information, the continuous length compression effect is enhanced by processing such as setting all the speeds above the legal speed to a constant quantization amount in the traffic information quantization table (FIG. 3).

  In addition, the position resolution of traffic information and the traffic expression resolution can be changed by changing the length of the unit block length for sampling the shape vector or switching the traffic information quantization table. According to the amount of information, the transmission path capacity of the information providing medium, the required accuracy, etc., it is easy to appropriately control the traffic expression resolution and the position resolution of the traffic information.

  Here, the upstream quantization unit speed Vn-1 is used as the statistical prediction value, but another statistical prediction value may be used. For example, when the weighted average of the speeds in the upstream three quantization units is used as the statistical prediction value S, the statistical prediction value S = aVn-1 + bVn-2 + cVn-3 (where a + b + c = 1) is used. A statistical prediction value S is calculated.

<System configuration>
FIG. 5 shows a broadcast-type traffic information providing system that generates and provides traffic information. This system encodes traffic information with a traffic information measuring device 10 that measures traffic information using a sensor A (ultrasonic vehicle sensor) 21, a sensor B (AVI sensor) 22, and a sensor C (probe car) 23. A code table creating unit 50 for creating a code table, a traffic information transmitting unit 30 that encodes and transmits traffic information and information on the target section, and a receiving side device 60 such as a car navigation system that receives the transmitted information, Consists of.
The traffic information measuring device 10 includes a sensor processing unit A (11), a sensor processing unit B (12) and a sensor processing unit C (13) that process data acquired from the sensors 21, 22, and 23, and each sensor processing unit. A traffic information calculation unit 14 that generates traffic information using the data processed in 11, 12, and 13 and outputs the traffic information data and data indicating the target section is provided.
The code table creation unit 50 includes a plurality of types of traffic information quantization tables 53 used for quantization of traffic information, and a distance quantization unit parameter table 54 that defines a plurality of types of sampling point intervals (unit section lengths). The code table calculation unit 51 that creates the code table divides the past traffic situation acquired from the traffic information measuring device 10 into patterns, and for all patterns, all of the traffic information quantization table 53 and the sampling point interval Various code tables 52 corresponding to the combinations are created.

  The traffic information transmitting unit 30 determines the traffic situation based on the traffic information collecting unit 31 that collects traffic information from the traffic information measuring device 10 and the collected traffic information, and the sampling point interval (distance direction quantization unit). (Quantity unit length) is determined, the quantization unit determination unit 32 that determines the quantization table and code table to be used, and the sampling point interval and the traffic information quantization table 53 determined by the quantization unit determination unit 32 are used. The traffic information is quantized and converted into a statistical prediction difference value, and the traffic information conversion unit 33 that converts the shape vector data of the target section into the statistical prediction difference value and the quantization unit determination unit 32 Encode the traffic information using the determined code table 52, and also transmit the encoded traffic information data and the shape vector data, the encoding processing unit 34 for encoding the shape vector of the target section Information transmitter 35 and traffic information conversion A digital map database 36 referred to by the unit 33 is provided.

  The receiving side device 60 includes an information receiving unit 61 that receives information provided from the traffic information transmitting unit 30, a decoding processing unit 62 that decodes the received information and reproduces traffic information and shape vectors, and a digital map database 65. Map matching of the shape vector using the data of the map, the map matching and section determination unit 63 that determines the target section of the traffic information, and the traffic information reflection that reflects the received traffic information in the data of the target section of the link cost table 66 Unit 64, a vehicle position determination unit 68 that determines the vehicle position using the GPS antenna 69 and the gyro 70, and an information utilization unit 67 that uses the link cost table 66 for route search from the vehicle position to the destination. And a guidance device 71 that performs voice guidance based on the route search result.

The functions of the code table calculation unit 51 of the code table creation unit 50, the quantization unit determination unit 32 of the traffic information transmission unit 30, the traffic information conversion unit 33, the encoding processing unit 34, the information transmission unit 35, etc. It can be realized by causing a computer built in the device on the traffic information providing side to perform the processing specified by the program. Also, the decoding processing unit 62, the map matching and section determining unit 63 of the receiving device 60, the traffic information Functions such as the reflection unit 64, the vehicle position determination unit 68, and the information utilization unit 67 can be realized by causing a computer (CPU) built in the reception-side device 60 to perform processing specified by a program.
FIG. 7 shows a data structure of map data (a) indicating a target section of traffic information output by the traffic information measuring device 10 and traffic information data (b).

The flowchart of FIG. 6 shows the operation of each part of this system.
The code table calculation unit 51 of the code table creation unit 50 analyzes the past traffic information sent from the traffic information measuring device 10 and aggregates the traffic information in the traffic situation of the pattern L (step 1), and the distance direction A quantization unit (sampling point interval) M is set (step 2), and a traffic information quantization table N is set (step 3). Next, a statistical prediction difference value of this traffic information is calculated according to a statistical prediction value calculation formula (step 4). Next, the distribution of statistical prediction difference values is calculated (step 5), and the run-length distribution (continuous distribution of the same value) is calculated (step 6). A code table is created based on the statistical prediction difference value and the run length distribution (step 7), and the code table of case L-M-N is completed (step 8). This process is repeated until all L, M, and N cases are completed (step 9).
In this way, a large number of code tables that can correspond to various traffic condition patterns and information representation resolutions are created and held in advance.

Next, the traffic information transmitting unit 30 collects traffic information and determines a traffic information providing section (step 10). For one traffic information providing section V (step 11), a shape vector around the traffic information providing section V is generated, a reference node is set (step 12), and the shape vector is reversibly or irreversibly encoded. (Step 13). This encoding and compression method is described in detail in Japanese Patent Application No. 2001-134318.
The quantization unit determination unit 32 determines the traffic situation, and determines the sampling point interval (unit section length in the distance direction quantization unit) and the quantization level (step 14). This process will be described in detail later.

The traffic information conversion unit 33 samples the distance direction from the reference node of the shape vector with the determined unit section length, divides the traffic information provision section (step 15), and the traffic information of each distance direction quantization unit Is calculated (step 16). Next, preprocessing for enhancing the compression effect of encoding is performed (step 17). This pre-processing will be described in detail later.
The traffic information conversion unit 33 quantizes the traffic information using the traffic information quantization table 53 determined by the quantization unit determination unit 32 based on the quantization level (step 18), and statistically analyzes the quantized traffic information. Conversion into a predicted difference value (step 19).
Next, the encoding processing unit 34 performs variable length encoding compression of the quantized traffic information using the code table 52 determined by the quantization unit determining unit 32 (step 20). Further, the unit section length is corrected using the relative position correction reference node (step 21).
This process is executed for all the traffic information provision sections (step 23). The information transmission unit 35 converts the encoded data into transmission data (step 24), and transmits the data together with the code table (step 25).

FIG. 8 shows an example of the data structure of the shape vector data string information (a) and the traffic information (b) transmitted from the traffic information transmitting unit 30. In addition to these pieces of information, the traffic information transmitting unit 30 simultaneously includes a shape vector code table, a traffic information quantization table (FIG. 3), a traffic information statistical prediction difference value code table (FIG. 4), and the like ( (Or another route).
The traffic information (FIG. 8B) has a data item of “number of quantized unit sections”, but instead of this data, EOD (End indicating the end of data in a code table) of Data) code may be set as a special code to indicate the end of the distance direction quantization unit in the encoded traffic information data string.

On the other hand, as shown in the flowchart of FIG. 6, when the information receiving unit 61 receives data (step 30), the receiving side device 60 has the decoding processing unit 62 for each traffic information providing section V (step 31). Then, the shape vector is decoded, and the map matching and section determination unit 63 performs map matching on its own digital map database 65 to identify the target road section (step 32). Further, the unit section length is corrected using the relative position correction reference node (step 33).
Further, the decoding processing unit 62 decodes the traffic information with reference to the code table (step 34). The traffic information reflecting unit 64 reflects the decrypted travel time on the link cost of the own system (step 35). Such processing is executed for all traffic information provision sections (steps 36 and 37). The information utilization unit 67 uses the provided travel time to execute the required time display and route guidance (step 38).

<Determination method of quantization unit>
The quantization unit determination unit 32 of the traffic information transmission unit 30 determines the sampling point interval (unit section length of the quantization unit in the distance direction) and the quantization level in the process of FIG. Step 14) will be described.
The quantization unit determination unit 32 determines the traffic situation and determines the resolution of information representation so that the transmission data amount of the traffic information does not exceed the transmission path capacity of the traffic information transmission unit 30. As shown in FIG. 44, the resolution of the information representation is based on the position resolution and the traffic representation resolution, and the position resolution depends on the sampling point interval (unit length in the distance direction quantization unit) when sampling. The traffic expression resolution is determined by the quantization level indicating the coarseness of quantization, which is determined by the selected quantization table. The quantization unit determination unit 32 determines the sampling point interval and the quantization table as part of determining the resolution of information representation.

If the sampling point interval is small, the traffic information becomes detailed but the data amount increases. Conversely, if the sampling point interval is large, the traffic information becomes coarse but the data amount is small. Similarly, if the state quantity of traffic information is quantized using a finely divided quantization table, the traffic information can be expressed in detail, but the amount of data increases. On the other hand, if a coarse quantization table is used, traffic information is rough but the amount of data is small.
The quantization unit determination unit 32 predicts the transmission data amount of traffic information from the current traffic situation, and adjusts the information expression resolution so that the transmission data amount does not exceed the transmission path capacity. At this time, the quantization unit determining unit 32 determines the sampling point interval and the quantization table expressing the traffic information of each route in consideration of the importance of the traffic information of each route, and also performs the encoding. As a code table to be used, a code table corresponding to the sampling point interval, the quantization table, and the traffic situation pattern is determined.

The flowchart of FIG. 9 shows an example of processing of the quantization unit determination unit 32.
The quantization unit determining unit 32 determines the target data size based on the transmission path capacity so that the transmission data amount does not exceed the transmission path capacity of the traffic information transmitting unit 30 (step 40). Next, the data size of the original information (FIGS. 7A and 7B) sent from the traffic information measuring device 10 in the previous cycle, and the transmission data encoded and transmitted (FIG. 8A) ) And (b)), the data size of the original information (FIGS. 7A and 7B) sent from the traffic information measuring device 10 in this cycle is determined by encoding. The degree of data size is calculated, and the expansion rate (or reduction rate) of the target data is determined based on the calculated data size (step 41).
Further, the traffic situation pattern L is determined from the current traffic situation (step 42).
Further, one traffic information providing section W around the transmission point where the traffic information transmitting unit 30 is transmitting the traffic information is extracted (steps 43 and 44), and the attribute of the map data link constituting the traffic information providing section W is extracted. (Road type / road number / number of intersections per unit length, etc.), road structure such as road width, traffic volume, traffic situation (congestion occurrence situation, etc.), and center of gravity of traffic information provision section W and transmission point The information importance of the traffic information provision section W is determined from the distance (step 45).

An increase / decrease value in a column where the information importance obtained in step 45 and the expansion rate (or reduction rate) of the target data obtained in step 41 intersect is obtained from the table of FIG. A quantization unit rank is calculated by adding an increase / decrease value to the default value of the information expression rank (quantization unit rank). Next, a sampling point interval (distance direction quantization unit) M _W and a traffic information quantization table N _W corresponding to the quantization unit rank are determined from the table of FIG. 10B (step 46). In addition, the L-M _W -N _W code table is used for encoding the traffic information of the traffic information providing section W. This process is performed for all traffic information provision sections around the transmission point (steps 47 and 48).

  By such processing, the sampling point interval and the quantization level can be dynamically changed according to the transmission data amount, the information importance of the traffic information provision section, and the like. Also, in the case of FM multiplex broadcasting, for example, when broadcasting from the Tokyo Broadcasting Station, information is provided so that information in Tokyo is fine and information on neighboring prefectures is coarse. The sampling point interval and the quantization level can be changed in accordance with the distance from the information providing point or the information providing area, such as the information around the installation point is fine, and the information becomes rougher as the distance from the point increases.

Further, the flow chart of FIG. 11 shows that when the recommended route is calculated on the center side and the recommended route and the surrounding traffic information are provided, the resolution of the traffic information on the recommended route is fine and deviates from the recommended route. Moreover, the resolution of the traffic information of the peripheral part shows the method of roughening according to the distance from a recommended route.
The quantization unit determining unit 32 collects recommended route information (step 50), determines a traffic condition pattern L from the current traffic condition (step 51), and recommends corresponding to rank 1 from the table of FIG. 10B. The route distance direction quantization unit M ₀ and the traffic information quantization table N ₀ are determined (step 52).
One traffic information providing section W around the recommended route is extracted (steps 53 and 54), the center of gravity of the traffic information providing section W is calculated, and the perpendicular distance from the center of gravity to the recommended route is calculated (step 55). The distance direction quantization unit _MW and the traffic information quantization table N _{W of the} traffic information providing section W are determined from the perpendicular distance (step 56).
This process is performed for all traffic information provision sections around the recommended route (steps 57 and 58).
Thus, the quantization unit determination unit 32 determines the resolution of information representation according to the importance of the traffic information to be provided.

<Pre-processing>
Before the traffic information is quantized, the traffic information conversion unit 33 performs a leveling process on the data before the traffic information is quantized so that the compression effect is enhanced.
The flow chart of FIG. 12 shows a pre-processing procedure for leveling data by taking a weighted average of data in adjacent N sections.
Focusing on the section p in order from the first section of the distance direction quantization unit (steps 60 and 61), the traffic of each section for a total of N sections combining the section p and the sections before and after p. Information Tp is collected (step 62). Next, the traffic information Tp of the section p is replaced with a weighted average of the traffic information of the N section calculated by the following equation (step 63).
Tp = (Σa _i × Ti) / N where Σa _i = 1
This is performed for all distance direction quantization units (step 64).
Such pre-processing processes represent the overall trend of traffic conditions that change microscopically. By performing this preprocessing, the statistical prediction difference values after quantization are concentrated in the vicinity of 0, and the compression effect in encoding is enhanced.
Further, when displaying traffic information, there is no problem for the information user even if the microscopic change in the traffic information in some sections is ignored.

  As shown in FIG. 13 (a), when the data of a part of the section is larger than the data of the section before and after the section and the difference is not less than the specified value, this is called a peak, and FIG. ) If the data in some sections is smaller than the data in the previous and subsequent sections and the difference is greater than or equal to the specified value, this is called a dip. If the dip interval is short, the information can be ignored.

The flowchart of FIG. 14 shows the pre-processing method in this case.
Focusing on the section p in order from the beginning section of the distance direction quantization unit (steps 70 and 71), traffic information Tp in each section from the section p to the N section is collected (step 71). Next, the peak and dip in the section p to p + N are searched (step 72). When the width of the peak and dip portion is less than the specified value, the peak and dip portion are replaced with the average value of the traffic information in the preceding and following sections (step 73).
In addition, the search of a peak and a dip can be performed in the following procedure, for example.
1. The average value and standard deviation of the traffic information in the sections p to p + N are calculated.
2. A deviation value is calculated for the traffic information Tp + i of each section.
3. When the deviation value is not less than the specified value or not more than the specified value, Tp + i is determined as a peak or a dip.
By performing such preprocessing, the compression effect when traffic information is encoded can be increased, and the amount of transmission data can be reduced.

<Modification>
Up to now, a fixed value (for example, 100 m unit) is set for the sampling point interval (unit length in the distance direction quantization unit) for quantizing the traffic information, and it is equally spaced in the distance direction from the reference node of the shape vector In the case of sampling at (fixed value interval), the number of divisions between the start and end of the shape vector is specified, and the distance between the start and end is divided at equal intervals by the number of divisions. You may make it set a quantization unit. In this case, the traffic information data (FIG. 8B) includes data on the reference node number on the end side, the reference node number on the start end side, and the number of divisions from the start end to the end. The receiving side that has received this calculates the unit interval length of the distance direction quantization unit by (distance between start / end reference nodes / number of divisions).

  Further, a section between component points such as nodes and interpolation points included in the shape vector can be used as a distance direction quantization unit of traffic information. In this case, the interval between the positions of each component point after compression coding of the shape vector is defined as a section in the distance direction quantization unit. Although the distance direction quantization units are not equally spaced, variable length coding is possible by expressing the travel time (or travel speed) or the like by the difference from the traffic information of the adjacent distance direction quantization units.

  In addition, when generating traffic information by the method of the embodiment of the present invention, when generating travel time information based on speed, as shown in the quantization table in FIG. It is better to perform quantization so that it becomes finer and coarser at higher speeds. Since travel time is inversely proportional to speed, even small changes can have a significant effect, especially where the speed range is small. In order to homogenize the error after conversion into travel time, it is preferable to express the velocity quantization table with a discrete value in a geometric series.

<Type of road section reference data>
So far, in order to notify the target road section, the shape vector data string is transmitted to the receiving side, and the receiving side has described the case where the target road section of traffic information is identified by referring to this shape vector data string. It is possible to use data other than the shape vector data string for the data for identifying (road section reference data). For example, as shown in FIG. 45 (a), a road segment identifier (link number) or an intersection identifier (node number) defined in a unified manner may be used.
Further, when both the providing side and the receiving side refer to the same map, the providing side can transmit latitude / longitude data to the receiving side, and the receiving side can specify a road section by this data.

Further, as shown in FIG. 45 (b), latitude / longitude data (name, road type, etc.) for reference of the positions of intermittent nodes P1, P2, P3, and P4 extracted from intersections and roads in the middle of links. Information that also holds information) may be transmitted to the receiving side to convey the target road. Here, P1 = link midpoint, P2 = intersection, P3 = link midpoint, and P4 = link midpoint. In this case, as shown in FIG. 45 (c), the receiving side first identifies each position of P1, P2, P3, and P4, and then connects each section by route search, Identify.
In addition to the shape vector data string, road segment identifiers, and intersection identifiers described above, as road segment reference data for identifying the target road, the road map is divided into tiles, and identifiers assigned to each of them are provided on the roads. The target road section of the traffic information may be specified by using the road section reference data using a kilometer post, a road name, an address, a zip code, and the like.

(Second Embodiment)
<Difference expression of prediction information>
In the second embodiment of the present invention, generation of traffic information prediction information will be described. In order to express the prediction information as a difference, as schematically shown in FIG. 15, two methods are conceivable.
The first method calculates the difference in the distance direction in the traffic information (a) in the time zone N + 1 (d)
This is a method of encoding the information of this difference (change point). This is the same as the current information encoding method described in the first embodiment.
The second method extracts the difference between the traffic information (a) in the time zone N + 1 and the traffic information (b) in the previous time zone N (c), and further calculates the difference in the distance direction of this difference. (E).

It cannot be generally said which of the first method and the second method is advantageous in reducing the amount of data.
Normal traffic jams include “bottleneck intersections (Harajuku intersection and Tomei Expressway Sagano bus stop) leading and trailing ends” and “traffic jams especially when the time difference between N and N + 1 is small. Since there are many places where the end does not move, "the second method is more advantageous overall. However, the total number of change points is greater in the second method (e) than in the first method (d) when the traffic jams before and after are connected or when the beginning and end of the traffic change. ) Is advantageous. In other words, it is considered case-by-case, and switching between the first method and the second method for each traffic information provision section is considered to be most effective depending on the time difference until the predicted time zone and the traffic change situation. .
Since the first method has been described in the first embodiment, generation of traffic information by the second method will be described in the second embodiment.

<Encoding of prediction information>
FIG. 16A shows the traffic information in the current information and the prediction information of the next time zone in each distance direction quantization unit.
First, the traffic information in the current information and prediction information is quantized using a quantization table (FIG. 16 (b)).
Next, the prediction information is expressed by a difference from the current information (FIG. 16C). At this time, according to the correlation law B, the value of the prediction information increases in data concentrated around ± 0. In FIG. 16C, the value of the current information is expressed as a difference from the statistical prediction value with the value of the adjacent distance direction quantization unit as the statistical prediction value.

Next, the prediction information is also expressed by a difference from the statistical prediction value (FIG. 16D). At this time, according to the correlation law C, most of the statistical prediction difference values of the prediction information are concentrated around ± 0.
In the process of FIG. 16C, the same result can be obtained even when the difference is obtained by subtracting the prediction information from the current information. Expression in the opposite direction in time is also possible.
The statistical prediction difference value between the current information and the prediction information thus obtained is encoded using a code table. As shown in FIG. 17A, the code table for encoding the statistical prediction difference value of the current information is the same as in the case of the first embodiment (FIG. 4). As shown in FIG. 17B, the code table for encoding the statistical prediction difference value of the prediction information is the same as the code table of the current information except that there is no special code.

<System configuration>
FIG. 18 shows a broadcast-type traffic information providing system that generates and provides traffic information including prediction information. The traffic information measuring device 10 of this system generates current information of traffic information using data processed by each sensor processing unit 11, 12, 13 and also generates prediction information using statistical information 16, A traffic information / prediction information calculation unit 15 that outputs the traffic information data and data indicating the target section is provided. Other configurations are the same as those of the first embodiment (FIG. 5).

  Further, the flowchart of FIG. 19 shows the operation of each part of this system. Compared with the process in the first embodiment (FIG. 6), the process of the code table creation unit includes a process of creating a code table (step 104 to step 08) used for encoding the prediction information. In addition, the traffic information transmitting unit is different in that the prediction information data is encoded together with the current information in the processing of step 116 to step 120. Others are the same.

  FIG. 20 shows an example of the data structure of the shape vector data string information (a) and the traffic information (b) transmitted from the traffic information transmitting unit 30. The shape vector data string information (a) is the same as that in the first embodiment (FIG. 8A). Regarding the traffic information (b), the data indicated by the star is different from that of the first embodiment (FIG. 8B), and the identification code that defines the code table of the prediction information and the effective time zone of the prediction information are shown. Information indicated, encoded data of prediction information, and the like are added. The prediction information includes a plurality of data with different effective time zones. In addition to the shape vector data string information (a) and the traffic information (b), the traffic information transmission unit 30 includes a shape vector code table, a traffic information quantization table, and a traffic information statistical prediction difference value code table. (FIG. 17 (a)), a prediction information code table (FIG. 17 (b)) and the like are transmitted simultaneously (or via another route).

<Change of prediction information resolution>
As for the prediction information, since the prediction accuracy decreases as the future is predicted, the information may be provided with lower resolution as the future time comes.
FIG. 21B schematically shows how the position resolution is lowered and the traffic expression resolution is lowered according to the future time from the original information (FIG. 21A). When lowering the position resolution, a plurality of distance direction quantization units are combined into one distance direction quantization unit, and the average value of the data of each distance direction quantization unit is used as the data of the combined distance direction quantization units.
In addition, when reducing the traffic expression resolution, data is quantized using a coarse quantization table.

FIG. 22 shows the prediction information obtained by reducing the position resolution from the original prediction information (a) and reducing the traffic expression resolution using the quantization table in which the roughness is set at a plurality of levels shown in FIG. The example which calculates | requires the statistical prediction difference value of is shown. In FIG. 22B, the position resolution is reduced to half. The traffic information values are averaged and rounded up.
In FIG. 22C, the prediction information and the current information are quantized using the quantization table of “quantization amount (current)” in FIG. In FIG. 22D, the prediction information is further quantized using the quantization table of “quantization amount (prediction 1)” in FIG. 23 to reduce the traffic expression resolution of the prediction information. In FIG. 22E, the time direction difference between the current information quantized using the quantization table of “quantization amount (prediction 1)” and the prediction information of FIG. 22D is extracted.
22F, the prediction information obtained in FIG. 22E and the current information obtained in FIG. 22C (current information quantized with the quantization table of “quantization amount (current)”). With respect to the above, the statistical prediction difference value is calculated using the value of the adjacent quantization unit as the statistical prediction value.
Here, the case of changing both the position resolution and the traffic expression resolution has been described, but only one of the resolutions may be lowered.

In this example, the position resolution of the prediction information is reduced to half (that is, the unit section length is doubled), but the unit section length can be set to an arbitrary value of 1.0 times or more. is there. (However, since the calculation is cumbersome, it is practically appropriate to set 1.5 times or 1.25 times)
In this example, the traffic expression resolution of the prediction information is also halved, but can be arbitrarily set as long as the direction becomes coarse (however, the calculation is troublesome). Since it is troublesome to calculate the last fraction in the distance direction, it is realistic to calculate “the roughest unit section length” in advance and divide it into 2 ^N based on the section length. It is done.

Decoding procedures are shown by (1) to (8) on the right side of FIG.
FIG. 24 shows the data structure of traffic information when the resolution of the prediction information is changed. For the prediction information in each effective time zone, a position resolution identification code (a code indicating that the unit section length becomes p times longer) and a quantization table number are added.
As described above, with respect to the prediction information, the resolution of the information expression can be changed according to the future time when the prediction is performed, and the transmission data amount can be reduced.

<Modification>
Here, when calculating the statistical prediction difference value, the case where the value of the adjacent quantization unit in the distance direction is used as the statistical prediction value is shown, but considering both space and time as shown in FIG. , The statistical prediction value in the black circle quantization unit,
Statistical prediction value = a (1) + b (2) + c (3) (however, a + b + c = 1)
Or = ((1) + (3)) / 2
May be set.

(Third embodiment)
In the third embodiment of the present invention, a method will be described in which traffic information represented by a function of distance from a reference node is subjected to orthogonal transformation to be decomposed into frequency components, and the traffic information is expressed by coefficients of each frequency component. .
Note that FFT (Fast Fourier Transform), DCT (Discrete Cosine Transform), wavelet transform, and the like are known as techniques for performing orthogonal transform on time series data and transforming it to frequency component coefficients. Here, the most common FFT (Fast Fourier Transform) will be described.

The Fourier transform is a transform that obtains a Fourier coefficient using a finite number of discrete values (sample values), and the complex function C is converted to the discrete value represented by the complex function f.
C (k) = (1 / n) Σf (j) · ω ^−jk (k = 0, 1, 2,..., N−1)
(Σ is an addition from j = 0 to n−1) (Equation 1)
This is called “Fourier transform”. Note that ω = exp (2πi).
C (k) is referred to as a Fourier coefficient. n is called the order.
vice versa,
f (j) = ΣC (k) · ω ^jk (j = 0, 1, 2,..., n−1)
(Σ is added from k = 0 to n−1) (Equation 2)
This is called inverse Fourier transform.
When performing a Fourier transform, the discrete values that f (j) can take are
・ Sampling interval δ = constant ・ n = 2 ^N
In this case, FFT (Fast Fourier Transform) is possible. Various FFT algorithms have been proposed.

FIG. 26 shows an experimental example in which traffic information is actually expressed by Fourier coefficients. Through this experimental example, a method for generating traffic information represented by Fourier coefficients will be described.
here,
(1) “Original traffic information data” is the state quantity of traffic information at each sampling point (FIG. 1B).
Is equivalent to the data of The number of data is set to 2 ⁵ (= 32) so that FFT is possible. In the case of FFT, it is possible to send two pieces of information simultaneously using the real part and imaginary part of the complex function, so here we set “speed information” in the real part and “congestion information” in the imaginary part. is doing. In the case of FFT, since the relative error is smaller when the value is the same, the congestion information is expressed as “congestion = 10, congestion = 20, quiet = 40” according to the numerical level of the speed information. ing.
(2) In the FFT conversion process, the data of (1) is expressed as an imaginary number (the numerical value with “i” at the end represents an imaginary coefficient), the FFT conversion process is performed, and the obtained FFT coefficient is indicated. Yes.
(3) The quantization table corresponds to the “traffic information quantization table” at the time of encoding compression, and the FFT coefficient in (2) is divided by the value of the quantization table (ie, quantized) to obtain “( 4) “Transmission data” is obtained. In this table, the FFT coefficient is “the lower frequency coefficient is described in the upper part of the row, and the higher frequency coefficient is described in the lower part”. The value of “(3) quantization table” is set small so that it can be expressed in detail, and the value of “(3) quantization table” is set so as to become coarser as the frequency increases.

(4) The transmission data is obtained by quantizing the FFT coefficient of (2) with the (3) quantization table. (If the value of the quantization table is 64, the real part and imaginary part of the FFT coefficient are divided by 64 and rounded off after the decimal point). (4) Since the transmission data is quantized more coarsely as the high frequency component, the value becomes relatively smaller from the sixth line onward, and the value is concentrated around “0”. As will be described later, the transmission data (4) is transmitted after being compressed with variable length coding (in the case of FFT coefficients, since there is no regularity, processing such as statistical difference is not performed), By concentrating the values around “0”, the compression effect of variable length coding is produced. Note that (3) a quantization table is also transmitted to the receiving side as table information necessary for decoding.
(5) The inverse FFT transform process is a process for inverse FFT transforming the received Fourier coefficient (4). First, each coefficient of the real part integer and the imaginary part integer of the received data is restored based on the value of the “quantization table” (if the quantization table value is 64, the real part and imaginary part of the received data Is multiplied by 64 and rounded to the nearest whole number). The value is subjected to inverse FFT conversion processing.

(6) The restored traffic information data is obtained by rounding off the coefficient values of the real part and imaginary part of the inverse FFT coefficient obtained by the inverse FFT transform, and restoring the real part to speed information and the imaginary part to traffic jam information. .
(7) Difference between original and restored data describes the difference between “restored data” and “measured data” for reference. Although an error of about ± 4 at the maximum occurs, almost the same value is obtained as compared with the original “measurement data”. In particular, the traffic information can be accurately reproduced if an error of this level is determined in advance as “congestion = 0 to 15, congestion = 16 to 25, quiet = 26 or more”.

FIG. 27 shows a case where (3) the quantization table is finely set for the same measurement data as FIG. 26 and the data is transmitted in detail in a form close to the raw data.
The difference between (7) original-restored data at this time is much smaller than in FIG. 26, and information is accurately reproduced. However, (4) the transmission data has a wider range of high-frequency components than that of FIG. 26 (large variation around ± 0). The compression effect is lost.
In this way, when traffic information is converted into FFT coefficients and transmitted, by adjusting the value of the quantization table, “information to be transmitted can be accurately reproduced although the amount of information is large,” “information Data with a small amount but low traffic information reproduction accuracy ”can be obtained, and the information amount can be adjusted in consideration of the position resolution described in the first embodiment.

The configuration of a system that provides traffic information by converting it into FFT coefficients is the same as that shown in FIG.
The flowchart of FIG. 28 shows a processing procedure in this system. The code table creation unit obtains an FFT coefficient by performing FFT (Step 204), quantizes the FFT coefficient to calculate a quantization coefficient (Step 205), calculates a distribution of the quantization coefficient (Step 207), The run length distribution is calculated (step 207), and a code table is created based on the run length distribution (step 208).
Further, the traffic information transmission unit performs level adjustment of the traffic information set in the real part and the imaginary part (step 218), performs FFT to convert it into a Fourier coefficient (step 219), and converts the Fourier coefficient to the code table. Refer to variable length coding compression (step 220).
Also, the receiving side device refers to the code table, performs inverse Fourier transform, and decodes the traffic information (step 234).
Other procedures are the same as in the case of FIG.

FIG. 30 shows a code table used for encoding, and FIG. 29 shows a data configuration example of traffic information transmitted from the traffic information transmitting unit. Here, the traffic quantization table identification code corresponds to the identification number of (3) quantization table in FIG. The code table identification code represents the identification code of the code table of FIG.
In the case of FFT, the number of data to be transmitted (corresponding to the number of rows in the table of FIG. 27 × 2) is usually twice the number of section divisions 2 ^N between reference nodes. However, when the high frequency component is cut and transmitted, the number is different. This is identified by the EOD code in the code table. At this time, the receiving side performs decoding processing by regarding the Fourier coefficient of the high frequency component as 0.
In this way, the traffic information is sampled in the distance direction of the shape vector indicating the road, the traffic information function represented by the state quantity at each sampling point is decomposed into frequency components, and the coefficient value of each frequency is encoded. By providing the traffic information, the receiving side device can restore the traffic information.

This method can be similarly applied to prediction information of traffic information. In the case of prediction information, the difference between the state quantity of the traffic information in the time zone N and the state quantity of the traffic information (prediction information) in the time zone N + 1 that is temporally adjacent is taken, and at each sampling point The difference state quantity may be subjected to orthogonal transformation to be converted into coefficient values of each frequency component, and the obtained coefficient values may be encoded.
In addition, among the coefficient values of each frequency obtained by conversion to frequency components, the coefficient values of the high frequency components are quantized so as to have a bias in the statistical frequency of occurrence, and the quantized coefficient values of each frequency are encoded. By doing so, the amount of data can be greatly reduced.
Also, encoding may be performed by deleting the coefficient value of the high frequency component from the coefficient value of each frequency obtained by the conversion to the frequency component.

(Fourth embodiment)
In the fourth embodiment of the present invention, a special data transmission method for traffic information expressed by Fourier coefficients or the like will be described.
<Transmission method from low frequency to high frequency for each frequency layer>
In this transmission method, traffic information is transmitted by a method according to the progressive transmission method of image information.
In the progressive transmission method of image information, the sending side
(1) First, send the low frequency component all at once. (2) Next, send the higher frequency component coefficient. (3) Next, send the higher frequency component coefficient:
Repeat this. On the receiving side to view the image,
(1) First, a blurred image appears. (2) Gradually, the image becomes finer. In this case, even if the communication speed is low, the receiving side can almost determine what image it is before receiving all the data, so it can determine “whether it is necessary” at an early stage.

This data transmission method can also be applied to traffic information that is converted into frequency components and expressed by coefficient values of each frequency.
This data transmission method can be realized by hierarchizing the coefficient value of each frequency representing traffic information according to the frequency, and dividing and transmitting by the hierarchy.
FIG. 31 shows a data configuration when traffic information is divided. FIG. 31A shows basic information and FFT coefficient information of low frequency components transmitted in the initial stage. This includes “the number of traffic information divisions” that indicates how many low-frequency components to high-frequency components are divided, and also includes “the number of this information” that indicates the number of this information among them. . FIG. 31B shows FFT coefficient information of a high frequency component which is one of the divided traffic information. Here, “the number of traffic information divisions” and “the number of this information” are included, but the data items overlapping with FIG. 31A are omitted.

When this traffic information transmission method is applied to the provision of traffic information from the Internet, etc., the user may look at the summary information and skip it when he / she wants to see the traffic information around the Internet. You can see the following: Further, when the traffic information is referred to while scrolling along the highway on the Internet or the like, if it is possible to skip the outline information by looking at the summary information, the scrolling can be advanced one after another.
In addition, when looking at past traffic information accumulated in chronological order as if it were an animation, you can go forward one after another if there is no traffic jam at the point of interest. .
Further, the coefficient values of the frequencies in a plurality of roads that provide information may be transmitted in the order indicated by the arrows in FIG. FIG. 32A shows a normal transmission order for comparison.

<Complementation with multiple transmission media>
When the coefficient values of each frequency representing traffic information are hierarchized by frequency and divided into low frequency components and high frequency components as shown in FIG. 31, it is possible to send traffic information separately from multiple media by layer. Become.
For example, terrestrial digital broadcasting provides shape vector data (location information) and rough traffic information (low frequency component of FFT coefficient) for all target roads in a wide area. Information (high-frequency component of FFT coefficient) that provides detailed traffic information provided by terrestrial digital broadcasting is provided for the vicinity of the place.
As described above, the wide-area broadcasting type media can provide public summary information and leave the provision of detailed information on the area to a beacon or the like.

FIG. 33 shows a system configuration diagram in this case. The traffic information transmission unit 30 includes an information transmission unit A (135) that provides traffic information through the wide area medium A, and an information transmission unit B (235) that provides traffic information through the beacon (media B), 60 includes an information receiving unit A (161) that receives the provision information of the wide area medium A, and an information receiving unit B (261) that receives the provision information from the beacon.
The receiving side device 60 reproduces the traffic information using the traffic information received by the information receiving unit A (161) and the traffic information received by the information receiving unit B (261).
Thus, in a system that provides traffic information by complementing wide area media and beacons, wide area media can transmit traffic information over a wider range because it is not necessary to send detailed information. Further, the receiving side device can reproduce necessary traffic information based on information obtained from the wide area media and detailed information obtained from the local beacon while traveling.

It is also possible to provide detailed information from the beacon as information to the in-vehicle device that has provided the probe information through the beacon, thereby promoting the provision of probe information.
Although the relationship between the wide area media and the beacon has been described here, the media that establishes the complementary relationship may be other combinations, or may be other media such as a mobile phone instead of the beacon.

(Fifth embodiment)
In the fifth embodiment of the present invention, a method for providing the latest traffic information using difference information from the previous traffic information will be described.
In this method, the method used to calculate the prediction information arranged in time series on the time axis of future time (that is, the difference with the previous time zone is taken, the difference with the adjacent point is taken, and the difference is taken. Method for encoding the value) to calculate the latest information of each time zone arranged in time series on the real time axis, and when calculating the latest information of the current time zone, The difference with the latest information is taken, the difference with the adjacent point is taken, and the difference value is encoded.
However, in the case of prediction information, it was possible to display and transmit the prediction information of each time zone in the traffic information data format of FIG. 20 (b). If that time does not come, data cannot be sent.

Therefore, here, as shown in FIG. 34, the data format of the traffic information in FIG. 20 (b) is based on the traffic information (FIG. 34 (a)) and the traffic information (prediction information for each time zone) ( The data is divided into the data format shown in FIG. 34 (b), and when the real time arrives, the latest information at that time is transmitted in the traffic information data format shown in FIG. 34 (b).
In these pieces of traffic information, a “traffic information division number” indicating the number of divisions and a “number of this information” indicating the number of the information in the information are described. When the division number is N-1, in the first cycle, the traffic information that is the basis from the next time to the N-1 cycle is transmitted with the shape vector data string information in the data format of FIG.
In the second cycle, the latest traffic information represented by the difference value from the information in the first cycle is transmitted in the data format of FIG.
In the third cycle, the latest traffic information represented by the difference value from the information in the second cycle is transmitted in the data format of FIG. 34 (b).
:
In the Nth cycle, the traffic information that is the basis for the next and subsequent times is transmitted together with the shape vector data string information in the data format of FIG.

By doing so, the total amount of information can be small. Further, the receiving side needs less map matching, and the total performance of the entire system is improved.
Even when the historical data of traffic information provided in the past is stored, the data amount is small.
In addition, when providing real-time traffic information on the Internet or the like, or providing a service that allows past traffic information to be animated one after another in time series, the burden of communication charges on the user can be reduced.

(Sixth embodiment)
In the sixth embodiment of the present invention, application of this method to a storage medium will be described.
Up to now, the traffic information generated by the present invention has been mainly described in the case of transmitting information by communication. However, this traffic information can be stored in a storage medium such as a hard disk, a CD, a DVD, or the like through the storage medium. It is also possible to move to other terminals.
FIG. 35 shows a system configuration diagram in this case. The traffic information conversion / recording device 330 includes an information storage unit 335 that stores encoded traffic information. The information storage unit 335 stores the traffic information encoded by the encoding processing unit 34 in the internal recording medium 331 or the like. Store in the external recording medium 332. In addition, the traffic information reference / utilization device 360 includes a decoding processing unit 362 that decodes encoded data. The decoding processing unit 362 is a traffic stored in the external recording medium 332 or the internal recording medium 361. Read and decode information. The method of utilizing the decrypted traffic information is the same as in the case of FIG.
As described above, the traffic information generated by the method of the present invention can be stored and used in the storage medium.

(Seventh embodiment)
In the seventh embodiment of the present invention, interactive traffic information provision will be described.
In this system, as shown in FIG. 36, the client specifies the range of traffic information and the amount of data (no more data is required), sends request information requesting traffic information, and the server responds to the request. Provide appropriate traffic information. The client-side device can be a car navigation system, a personal computer, a portable terminal, or the like.
FIG. 39 is a block diagram showing the configuration of this system. The client device 460 includes an input operation unit 463 for a user to input a request, a display range / data size determination unit 462 for determining a display range and a data size to be requested based on the input operation, and request information transmission for transmitting a request Unit 461, response information receiving unit 464 that receives response information, decoding processing unit 465 that decodes encoded data, traffic information utilization unit 466 that utilizes reproduced traffic information, and traffic information utilization And a digital map database 467 referred to by the section 466.

On the other hand, the server device 430 uses the request information receiving unit 431 that receives the request information, the transmission traffic information area / detail level determination unit 432 that determines the area and level of detail of the traffic information to be transmitted, and the code table data 434. A traffic information quantization / encoding unit 435 that encodes the traffic information data 433 and a response information transmission unit 436 that transmits the encoded traffic information are provided.
The flowchart of FIG. 38 shows the operation procedure of this system.
The client device 460 determines a range of traffic information necessary for processing such as display and route search and a desired data size (step 310), and transmits request information to the server 430 (step 311).

When the server device 430 waiting for the request from the client (step 300) receives the request information from the client (step 301), it determines the level of detail of the traffic information transmitted to the client from the request information (step 302). ) Quantize and encode the traffic information (step 303), and transmit the encoded traffic information and code table to the client (step 304). At this time, the server device 430 transmits the data shown in FIGS. 8 and 20 to the client.
When the client device 460 receives the response information from the server 430 (step 312), the client device 460 refers to the code table, decodes the traffic information represented by the code (step 313), and maps based on the position information (shape vector, etc.). Matching is performed, the position of the received traffic information is specified (step 314), and the traffic information is utilized.

FIG. 37 shows an example of this request information.
The “desired maximum data size” may be a communication fee or a communication time in the case of the packet fee system. The range of the request is “rectangular lower left / upper right latitude / longitude”, “center point”, “prefecture / city code”, “road designation”, “route search request start / end latitude / longitude”, “current location latitude / longitude + travel direction” You may specify using, and you may specify combining them.
When the traffic information is requested by the “start / end latitude / longitude for route search request”, the transmission traffic information area / detail level determination unit 432 is detailed about the traffic information on the recommended route, and the distance from the recommended route is long. Indeed, the level of detail of traffic information is determined so that it becomes coarser.

When the traffic information is requested by “current location latitude / longitude + traveling direction”, the traffic information on the traveling direction and the traveling road in the vicinity of the current location is detailed in detail so that the traffic information becomes coarser in the distance. To decide.
As described above, in the interactive information provision, the resolution of the information expression in the traffic information can be finely adjusted according to the request.
In addition, the prediction information around the estimated arrival time is fine according to the estimated travel time for each link, and adjustments such as providing the information by making the prediction information coarser as it deviates from the estimated arrival time are also possible.

(Eighth embodiment)
In the embodiments so far, the case where the traffic information providing device (traffic information transmitting unit) serving as the center provides traffic information to a traffic information using device such as a car navigation on-vehicle device has been described. The in-vehicle device becomes a traffic information providing device, the center that collects information from the probe car on-board device becomes the traffic information utilization device, and the on-vehicle probe car on-board device provides various measurement information such as travel speed and fuel consumption as traffic information. The traffic information generation method of the present invention can also be applied to the system provided to the center. In the eighth embodiment of the present invention, such a probe car system will be described.

  As shown in FIG. 46, this system includes a probe car in-vehicle device 90 that measures and provides data during travel, and a probe car in-vehicle device 80 that collects this data. Detection information of a code table receiving unit 94 that receives a code table used for data encoding from the probe car collection system 80, a sensor A106 that detects a speed, a sensor B107 that detects a power output, and a sensor 108 that detects fuel consumption. The sensor information collecting unit 98 that collects, the own vehicle position determining unit 93 that determines the position of the own vehicle using information received by the GPS antenna 101 and the information of the gyro 102, and the traveling locus and sensors A, B, and C of the own vehicle A travel locus measurement information storage unit 96 that stores measurement information of the measurement information, a measurement information data conversion unit 97 that generates sampling data of the measurement information, and a sample of measurement information using the received code table data 95 An encoding unit 92 for encoding Gudeta and traveling locus data, and a traveling locus transmission unit 91 which transmits the encoded data to the probe car collection system 80.

On the other hand, the probe car collection system 80 includes a travel locus receiving unit 83 that receives travel data from the probe car on-vehicle device 90, an encoded data decoding unit 82 that decodes received data using the code table data 86, and a decoding Measurement information data reverse conversion unit 87 that restores measurement information using the converted data, travel locus measurement information utilization unit 81 that utilizes the restored measurement information and travel locus data, and the current position of the probe car A code table selection unit 85 that selects a code table to be supplied to the probe car on-vehicle device 90, and a code table transmission unit 84 that transmits the selected code table to the probe car.
Here, the compression coding method shown in the third embodiment, that is, the case where the traffic information is subjected to orthogonal transformation and expressed in the coefficient of each frequency component will be described.

  The own vehicle position determination unit 93 of the probe car on-vehicle device 90 identifies the own vehicle position using the information received by the GPS antenna 101 and the information of the gyro 102. The sensor information collection unit 98 collects measurement values such as speed information detected by the sensor A106, engine load detected by the sensor B107, gasoline consumption detected by the sensor C108, and the like. The measurement information collected by the sensor information collection unit 98 is stored in the travel locus measurement information accumulation unit 96 in association with the vehicle position identified by the vehicle position determination unit 93.

  The measurement information data conversion unit 97 represents the measurement information accumulated in the travel locus measurement information accumulation unit 96 as a function of the distance from the measurement start point (reference position) on the traveling road, and generates sampling data of the measurement information. The encoding processing unit 92 performs orthogonal transformation on the sampling data, converts the measurement information into coefficient values of frequency components, and encodes the travel locus data and the converted coefficient values using the received code table data 95. To do. The encoded travel locus data and measurement information are sent to the probe car collection system 80 through the travel locus transmitter 91.

  In the probe car collection system 80 that has received the data, the encoded data decoding unit 82 decodes the encoded travel locus data and measurement information using the code table data 86. The measurement information data inverse transform unit 87 performs orthogonal inverse transform using the decoded coefficient value to restore the measurement information. The travel locus measurement information utilization unit 81 uses the reconstructed measurement information to create traffic information of the road on which the probe car has traveled. Thus, the traffic information generation method of the present invention can also be used to generate information to be uploaded from the probe car onboard device.

  As is clear from the above description, in the traffic information providing system of the present invention, the position resolution and the traffic expression resolution can be arbitrarily set, and the resolution of the information expression can be changed at any time according to the importance of the traffic information. it can. In addition, the traffic information “prediction service” can be flexibly dealt with.

The figure which shows the calculation method of the statistical prediction difference value of the traffic information in the 1st Embodiment of this invention. The figure which shows the correlation utilized by the production | generation of the traffic information in the 1st Embodiment of this invention The figure which shows the traffic information quantization table in the 1st Embodiment of this invention. The figure which shows the code table of the statistical prediction difference value in the 1st Embodiment of this invention. System configuration diagram according to the first embodiment of the present invention The flowchart which shows operation | movement of the system in the 1st Embodiment of this invention. Data configuration diagram of original information in the first embodiment of the present invention Data configuration diagram of shape vector data and traffic information in the first embodiment of the present invention The flowchart which shows the quantization unit determination procedure in the 1st Embodiment of this invention. The figure which shows the quantization unit determination table in the 1st Embodiment of this invention. The flowchart which shows the quantization unit determination procedure based on the distance from the recommendation path | route in the 1st Embodiment of this invention. The flowchart which shows the pre-processing process procedure in the 1st Embodiment of this invention. The figure which shows the peak and dip in the 1st Embodiment of this invention The flowchart which shows the processing procedure of the peak and dip deletion in the 1st Embodiment of this invention The figure explaining the difference expression of the prediction information in the 2nd Embodiment of this invention. The figure which shows the calculation method of the statistical prediction difference value of the prediction information in the 2nd Embodiment of this invention. The figure which shows the code table of the statistical prediction difference value and prediction information in the 2nd Embodiment of this invention. System configuration diagram according to the second embodiment of the present invention The flowchart which shows operation | movement of the system in the 2nd Embodiment of this invention. Data configuration diagram of shape vector data and traffic information in the second embodiment of the present invention The figure explaining the change of the information representation resolution of the prediction information in the 2nd Embodiment of this invention. The figure which shows the change process of the information representation resolution of the prediction information in the 2nd Embodiment of this invention. The figure which shows the quantization table of the prediction information in the 2nd Embodiment of this invention. Data configuration diagram of traffic information with changed information representation resolution of prediction information in the second exemplary embodiment of the present invention The figure which shows the example of the other statistical prediction value in the 2nd Embodiment of this invention. The figure which shows the quantization procedure of FFT utilization in the 3rd Embodiment of this invention. The figure which shows the quantization procedure of FFT utilization when the quantization table in the 3rd Embodiment of this invention is changed. The flowchart which shows operation | movement of the system in the 3rd Embodiment of this invention. Data configuration diagram of traffic information expressed in FFT in the third embodiment of the present invention The figure which shows the code table of the FFT coefficient in the 3rd Embodiment of this invention. Data configuration diagram of traffic information in the fourth embodiment of the present invention The figure which shows the transmission procedure of the traffic information in the 4th Embodiment of this invention. System configuration diagram according to the fourth embodiment of the present invention Data configuration diagram of traffic information in the fifth embodiment of the present invention System configuration diagram according to the sixth embodiment of the present invention Explanatory drawing of the interactive system in the 7th Embodiment of this invention Data structure diagram of request information in the seventh embodiment of the present invention The flowchart which shows operation | movement of the system in the 7th Embodiment of this invention. System configuration diagram according to the seventh embodiment of the present invention Explanatory drawing of the relative position correction method using the conventional reference node Explanatory drawing explaining the subject of conventional traffic information Explanatory drawing explaining the problem in the information display resolution of conventional traffic information Explanatory drawing explaining the problem at the time of transmission of conventional traffic information A diagram explaining how information display resolution should be Figure showing road section reference data The block diagram which shows the structure of the probe car information collection system in the 8th Embodiment of this invention.

Explanation of symbols

10 Traffic information measuring device
11 Sensor processing part A
12 Sensor processing part B
13 Sensor processing part C
14 Traffic information calculator
15 Traffic Information / Prediction Information Calculation Department
16 Statistics
21 Sensor A (Ultrasonic vehicle sensor)
22 Sensor B (AVI sensor)
23 Sensor C (probe car)
30 Traffic information transmitter
31 Traffic Information Collection Department
32 Quantization unit decision unit
33 Traffic information converter
34 Encoding processing section
35 Information transmitter
36 Digital map database
50 Code table generator
51 Code table calculator
52 Code table
53 Traffic information quantization table
54 Distance quantization unit parameter table
60 Receiver device
61 Information receiver
62 Decryption processor
63 Map matching and section determination section
64 Traffic Information Reflection Department
66 Link cost table
67 Information Utilization Department
68 Own vehicle position determination unit
69 GPS antenna
70 Gyro
71 Guidance device
80 probe car collection system
81 Traveling Track Measurement Information Utilization Department
82 Encoded data decoder
83 Running track receiver
84 Code table transmitter
85 Code table selection section
86 Code table data
87 Measurement information data reverse conversion section
90 Probe car in-vehicle device
91 Traveling track transmitter
92 Encoding processing section
93 Self-vehicle position determination unit
94 Code table receiver
95 Code table data
96 Running track measurement information storage unit
97 Measurement information data converter
98 Sensor information collection unit
101 GPS antenna
102 Gyro
106 Sensor A
107 Sensor B
108 Sensor C
135 Information transmitter A
161 Information receiver A
235 Information transmitter B
261 Information receiver B
330 Traffic Information Conversion / Recording Device
331 Internal recording media
332 External storage media
335 Information storage unit
360 Traffic information reference / utilization device
361 Internal storage media
362 Decryption processing unit
430 server equipment
431 Request information receiver
432 Transmission traffic information area and level of detail judgment section
433 Traffic information data
434 Code table data
435 Traffic Information Quantization / Encoding Department
436 Response information transmitter
460 client device
461 Request information transmitter
462 Display range / data size determination section
463 Input operation section
464 Response information receiver
465 Decryption processor
466 Traffic Information Utilization Department
467 Digital Map Database

Claims (11)

A sampling point setting unit for setting sampling points at equal intervals along the road from a reference point in the road section specified by the road section reference data;
The traffic information state quantity that changes along the road in the road section is calculated for each sampling point based on the traffic information state quantity measured at the measurement point ,
A road information providing device that compresses a state quantity of traffic information at each sampling point calculated. A sampling point setting unit for setting sampling points by dividing a road section specified by road section reference data at equal intervals along the road;
The traffic information state quantity that changes along the road in the road section is calculated for each sampling point based on the traffic information state quantity measured at the measurement point ,
A road information providing device that compresses a state quantity of traffic information at each sampling point calculated. The road information providing device according to claim 1 or 2,
The traffic information state device that changes along the road includes any one of a vehicle traveling speed, travel time, and traffic jam information. The road information providing device according to any one of claims 1 to 3,
A road information providing apparatus that converts a state quantity of the traffic information at each sampling point into a coefficient value of a frequency component by orthogonal transformation and compresses it. The road information providing device according to claim 4,
A road information providing apparatus that compresses a state quantity of the traffic information by deleting a value of a high frequency component. The road information providing device according to any one of claims 1 to 5,
A road information providing device, which is a probe car on-board terminal device or a navigation terminal device mounted on a vehicle, which provides the measurement information measured during travel by the road information providing device. The road information providing device according to any one of claims 1 to 5,
The road information providing apparatus, wherein the road information providing apparatus is a center side apparatus that provides traffic information. Setting sampling points at equal intervals along the road from a reference point in the road section identified by the road section reference data;
Calculating a state quantity of traffic information changing along the road in the road section for each sampling point based on a state quantity of traffic information measured at a measurement point ;
Compressing the traffic information state quantity at each calculated sampling point;
A road information providing method comprising: Dividing the road section identified by the road section reference data into equal intervals along the road and setting sampling points;
Calculating a state quantity of traffic information changing along the road in the road section for each sampling point based on a state quantity of traffic information measured at a measurement point ;
Compressing the traffic information state quantity at each calculated sampling point;
A road information providing method comprising:   The program which makes a computer perform each step of the road information provision method of Claim 8 or 9.   The road information utilization apparatus which reproduces the state quantity of the traffic information transmitted from the road information provision apparatus of any one of Claim 1 thru | or 7 on a digital map.

Images