JP4746000B2 - Fuel saving driving support method and fuel saving driving support system for construction machinery - Google Patents

Description

  The present invention relates to a fuel-saving driving support method and a fuel-saving driving support system for a construction machine such as a hydraulic excavator.

  A construction machine such as a hydraulic excavator performs a desired work by driving a hydraulic actuator of a work machine by an engine-driven hydraulic pump. As in the case of an automobile or the like, fuel-saving operation is required. The In order to support fuel-saving driving, it is necessary to evaluate the operating status of the work machine.

  Regarding the evaluation of the driving situation for fuel-saving driving of this type of construction machine, there is a proposed example in which guidance output of average fuel consumption is made on a monitor (for example, see Patent Document 1). Further, there is a proposal example in which various sensing parameters such as hydraulic pressure are made into a self-organizing map to calculate the neuron distance between each parameter and specify the work content of the work implement (for example, see Patent Document 2). ). Furthermore, there is an attempt to grasp the work content of the work machine by analyzing the lever operation or the like.

  For construction machines such as hydraulic excavators, the situation around the construction machine is photographed with a camera, and a peripheral image is displayed on the monitor according to the operation of the machine, so that a field of view can be secured and the working range can be maintained. There is a proposed example in which an object that has entered the camera is detected from a camera image and a warning is issued (see, for example, Patent Document 3).

JP 2005-163470 A JP 2005-25351 A JP-A-10-72851

  In construction machinery, the load varies depending on the work content of the work equipment, and even in the same work content, the way of consuming fuel varies depending on the operating conditions. Therefore, it is actually necessary to recognize the work content and estimate the fuel consumption by work content. Accurate driving evaluation in line with However, the one described in Patent Document 1 simply calculates the average fuel consumption per hour, does not take into account the fuel consumption by work content of the work implement, and accurate driving evaluation in line with the substance. It will not be. Moreover, in the thing of patent document 1, although an operation mode can be selected, the selection operation must be performed by the operator himself. Here, if the operator inputs and sets the work content, the work content is specified accurately. However, it is not realistic for the operator to input the work mode one by one during the work.

  In addition, the work content recognition method of the work machine disclosed in Patent Document 2 is an indirect recognition method, and does not look at the actual work machine movement, so the work content is not recognized correctly. . As a result, the fuel consumption evaluation for each work content becomes inaccurate and does not provide appropriate support for fuel-saving driving.

  On the other hand, as shown in Patent Document 3 and the like, a technique for mounting a camera on a construction machine such as a hydraulic excavator is well known, but it is only intended to ensure safety in the rear and surroundings.

  The present invention has been made in view of the above, and is capable of accurately recognizing the work content of the work machine and accurately evaluating the fuel-saving operation for each work content. An object is to provide a support method and a fuel-saving driving support system.

  In order to solve the above-described problems and achieve the object, a construction machine fuel-saving driving support method according to the present invention includes a step of sequentially photographing a work state of a work machine with a camera mounted on the construction machine, and a sequential photographing. Pattern matching is sequentially performed on the captured camera image and a template image stored in advance according to the work status of the work machine, and the work content of the work machine is recognized from the movement of the work machine based on the pattern matching result obtained in time series. Work, the process of accumulating the number of operations and the amount of fuel used for each recognized work content, and the fuel consumption for each work content calculated based on the accumulated number of operations and the amount of fuel used at a predetermined timing And a step of displaying the quality of the fuel usage state for each work content as compared with a target fuel consumption preset for each content.

  In the fuel-saving driving support method for a construction machine according to the present invention, in the above-described invention, in the step of recognizing, the camera image of each part of the work implement photographed and the work status of the work implement are used. A template image for each part stored in advance is used.

  Further, the fuel-saving driving support method for a construction machine according to the present invention is characterized in that, in the above-mentioned invention, the accumulation time further accumulates work time for each work content.

  In addition, a fuel-saving driving support system for a construction machine according to the present invention includes a camera that is mounted on a construction machine and sequentially captures a work state of a work machine, and a template that stores in advance a template image corresponding to the work state of the work machine. An image memory, a camera image sequentially captured by the camera, and a template image stored in advance in the template image memory are sequentially subjected to pattern matching, and from the movement of the working machine according to a pattern matching result obtained in time series, the working machine Work content recognizing means for recognizing work contents, accumulation means for accumulating the number of operations and the amount of fuel used for each work content sequentially recognized by the work content recognition means, and the work content at a predetermined timing. Fuel consumption calculation means for calculating the fuel consumption for each work based on the number of operations accumulated in the table and the amount of fuel used A display means for comparing the fuel consumption for each work content calculated by the fuel consumption calculation means with a target fuel consumption preset for each work content, and for displaying the quality of the fuel usage state for each work content. Features.

  In the fuel-saving driving support system for a construction machine according to the present invention as set forth in the invention described above, the template image memory stores in advance a template image for each part corresponding to the work status of the work implement, and recognizes the work content. The means uses a photographed camera image of each part of the working machine.

  Moreover, the fuel-saving driving support system for a construction machine according to the present invention is characterized in that, in the above invention, the accumulating means further accumulates work time for each work content.

  The fuel-saving driving support method and the fuel-saving driving support system for a construction machine according to the present invention sequentially photograph the work situation of a work machine with a camera mounted on the construction machine, and sequentially photographed camera images and the work situation of the work machine. The work contents of the work implement are recognized from the movement of the work implement based on the pattern matching results obtained in a time series by sequentially pattern matching with the template image stored in advance according to the actual image of the work implement movement. It is a direct recognition based on the results, which can increase the accuracy of the work content recognition, and accumulate the number of work and the amount of fuel used for each recognized work content to obtain the fuel efficiency calculated for each work content. Since the driving situation is evaluated in comparison with the target fuel consumption for each work content, there is an effect that the evaluation for fuel-saving driving can be accurately performed for each work content.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. The present embodiment shows an application example to a hydraulic excavator as a construction machine. The present invention is not limited to the embodiment, and various modifications can be made without departing from the spirit of the present invention.

  FIG. 1 is a schematic side view showing a configuration example of a hydraulic excavator to which the fuel-saving driving support method of the present embodiment is applied, and FIG. 2 is a schematic block diagram showing a configuration example of a fuel-saving driving support system of the hydraulic excavator. FIG. The hydraulic excavator 1 according to the present embodiment includes an upper swing body 3 including a cockpit 3a and the like on a crawler belt 2 forming a lower traveling body. A boom 4, an arm 5, and a bucket 6 are provided on a vehicle body portion of the upper swing body 3. A work machine 7 composed of each part is attached. Although not shown in FIG. 1, the boom 4, the arm 5, and the bucket 6 are operated by driving hydraulic actuators 8 (see FIG. 2) such as hydraulic cylinders for booms, arms, and buckets. A desired work such as surface finishing or surface finishing is executed.

  Here, the excavator 1 of the present embodiment includes a camera 9 on the cockpit 3a, for example. The camera 9 photographs a subject using a solid-state imaging device such as a CCD or CMOS sensor. In this embodiment, the camera 9 can photograph the working status of the work machine 7 (the boom 4, the arm 5 and the bucket 6). It is installed as follows.

  Further, the fuel-saving driving support system of the hydraulic excavator 1 according to the present embodiment includes a control controller 10, a vehicle body controller 20, and an image processing controller 30 as a control system, as shown in FIG. The vehicle body controller 20 controls most of the hydraulic excavator 1. For example, the vehicle body controller 20 controls the engine-driven hydraulic pump 22 by controlling the rotational speed of the engine 21 with the fuel injection command value, and via the operation valve 23. The hydraulic actuator 8 such as a hydraulic cylinder and a hydraulic motor is driven and controlled to perform a desired operation. The vehicle body controller 20 takes in an operation lever signal of the operation lever 24 for driving the corresponding hydraulic actuator 8 by controlling the operation valve 23 with the pilot hydraulic pressure, and a fuel injection command value corresponding to the operation amount of the operation lever 24. Is calculated.

  In addition, the image processing controller 30 is for performing processing for recognizing the work content by capturing the camera image information of the work status of the work machine 7 sequentially photographed by the camera 9, and the pattern matching processing unit 31 and the work content. A recognition unit 32. Also connected to the image processing controller 30 is a template image memory 33 that stores in advance a template image that has been photographed by the camera 9 in accordance with the work status of the work machine 7 and has undergone predetermined image processing such as image extraction and masking. Has been. The pattern matching processing unit 31 executes a pattern matching process that compares a camera image sequentially captured by the camera 9 with a template image stored in advance in the template image memory 33 to determine a template image that matches the camera image. Thus, the posture (working situation) of the work machine 7 is specified. The work content recognition unit 32 performs pattern matching processing on each of the sequentially captured camera images by the pattern matching processing unit 31, and stores the pattern matching results (sequentially determined template image information) obtained in time series in a memory. Then, by analyzing the time series and specifying the movement pattern of the work implement 7, processing for recognizing the work content (for example, excavation / loading work) corresponding to the movement pattern is performed, and the recognized work content information is obtained. Output to the controller 10.

  The controller 10 is for evaluating the operation status of the hydraulic excavator 1 for each work content in consideration of the work content recognized by the image processing controller 20. The accumulator 11, the fuel consumption calculator 12, and the evaluation unit 13 and a monitor 14 for displaying the evaluation result is connected. The controller 10 is connected to a memory 17 having a work content table 15 and a target fuel efficiency table 16 for each work content, and an evaluation display button 18 for instructing display of evaluation information. Here, the accumulating unit 11 is for performing processing for accumulating the number of operations, the amount of fuel used, and the operation time for each operation content sequentially recognized by the operation content recognition unit 32 in the operation content table 15. Further, the fuel consumption calculation unit 12 is information on fuel consumption for each work content based on the number of operations, the amount of fuel used, and the number of operations accumulated in the work table 15 at a predetermined timing, for example, when the evaluation display button 18 is pressed. This is for performing the process of calculating. The evaluation unit 13 compares the fuel consumption for each work content calculated by the fuel consumption calculation unit 12 with the target fuel consumption set in the target fuel consumption table 16 for each work content in advance for each work content. Is to be displayed on the monitor 14.

Here, template images and pattern matching processing used in the present embodiment will be described with reference to FIGS. In the present embodiment, according to the posture (working situation) of the work implement 7, each part corresponding to the working situation indicating how the boom 4, the arm 5, and the bucket 6, which are the respective parts, are viewed from the camera 9. A template image for each part is photographed by the camera 9 and stored in the template image memory 33 in advance. Here, the displacement state of the boom 4, the arm 5, and the bucket 6 is represented by angles θ ₁ , θ ₂ , and θ ₃ in the counterclockwise direction based on the basic posture as shown in FIG. 3, for example. In this embodiment, as shown in FIGS. 4 to 6, template images are prepared in increments of 1 °. However, template images that are coarser than 1 ° or finely classified according to the required resolution. May be prepared.

First, FIG. 4 is an explanation showing an example of a template image of the boom 4 located on the base side of the work machine 7. In the template image, the hatched portion is masked to avoid the influence of the background of the boom 4 and is not used for pattern matching. Here, in the pattern matching process described later, for example, if the camera image of the boom 4 portion taken by the camera 9 matches the template image of θ ₁ = 1 °, the working angle θ ₁ of the boom 4 at that time is 1 °. It turns out that.

FIG. 5 is an explanatory diagram illustrating an example of a template image of the arm 5 supported by the boom 4 in the work machine 7. Since the posture (working situation) of the arm 5 depends on the angle θ ₁ of the boom 4 located on the base side, a template image in which the angle θ ₂ of the boom 5 is changed for each angle θ _{1 of the} boom 4 is prepared.

FIG. 6 is an explanatory diagram illustrating an example of a template image of the bucket 6 supported by the arm 5 in the work machine 7. Since the posture (working situation) of the bucket 6 depends on the angles θ ₁ and θ ₂ of the boom 4 and the arm positioned on the base side, the angle θ _{3 of the} boom 4 and the arm 5 is set for each of the angles θ ₁ and θ _2. A template image in which is changed is prepared.

In this embodiment, the template image of the arm 5 in each angle theta _1, the template image is an angle theta ₁ of the bucket 6 has been so prepared for each theta _2, respectively θ ₁ = 0, θ ₂ = A template image in the case of 0 may be used as a reference and created by calculation or enlargement / reduction processing.

Next, an example of pattern matching processing by the pattern matching processing unit 31 using the template image memory 33 in which the template image for each part is stored will be described with reference to a schematic flowchart shown in FIG. When a camera image obtained by photographing the work implement 7 with the camera 9 is input as an input image, the pattern matching processing unit 31 proceeds with the process in the order of a boom, an arm, and a bucket. First, the variable i related to the angle θ ₁ of the boom 4 is set to 0 (step S101), and it is compared whether or not the boom image included in the input image matches the template image of θ ₁ = i (step S102). . To determine whether or not they match, for example, a well-known normalized correlation may be used in the pattern matching process (the same applies to the following processes). As a result of the comparison, if the boom image included in the input image does not match the template image (step S103; No), the variable i is incremented by +1 (step S104), and the next θ ₁ = i It is compared whether or not it matches the template image (step S102). Such comparison processing is repeated until a matching result is obtained (step S103). If the matching results are obtained (step S103; Yes), the working state (posture) of the boom 4 photographed by the camera 9 is the angle θ _1. The template image is determined to be in the state (step S105). Here, the determined angle θ ₁ is defined as Θ ₁ .

Subsequently, the variable j related to the angle θ ₂ of the arm 5 is set to 0 (step S106), and whether or not the arm image included in the input image matches the template image of θ ₂ = j when θ ₁ = Θ ₁ is satisfied. These are compared (step S107). If the arm image included in the input image does not match the template image as a result of the comparison (step S108; No), the variable j is incremented to +1 (step S109), and the next template with θ ₂ = j It is compared whether or not it matches the image (step S107). Such comparison processing is repeated until a matching result is obtained (step S108). If the matching results are obtained (step S108; Yes), the work status (posture) of the arm 5 photographed by the camera 9 is θ ₁ = Θ _In the case of ₁ , it is determined that the template image is at the angle θ ₂ (step S110). Here, the determined angle θ ₂ is defined as Θ ₂ .

Further, a variable k related to the angle θ ₃ of the bucket 6 is set to 0 (step S111), and the bucket image included in the input image is a template of θ ₃ = k when θ ₁ = Θ ₁ and θ ₂ = Θ _2. It is compared whether or not it matches the image (step S112). If the bucket image included in the input image does not match the template image as a result of the comparison (step S113; No), the variable k is incremented to +1 (step S114), and the next template with θ ₃ = k It is compared whether or not it matches the image (step S112). Such a comparison process is repeated until a matching result is obtained (step S113). If the matching results are obtained (step S113; Yes), the work status (posture) of the bucket 6 photographed by the camera 9 is θ ₁ = Θ It is determined that the template image is at the angle θ ₃ when ₁ and θ ₂ = Θ ₂ (step S115). Here, the determined angle θ ₃ is defined as Θ ₃ . Thereby, it becomes possible to specify the work status (posture) of the boom 4, the arm 5, and the bucket 6 captured by the camera 9 at a certain time.

  The pattern matching processing unit 31 sequentially performs such pattern matching processing on the images of the work machine 7 sequentially photographed by the camera 9, and determines the template images of the boom 4, the arm 5, and the bucket 6 each time.

  The work content recognition unit 32 stores the pattern matching results (angle information of the template images that are sequentially determined) obtained in time series from the pattern matching processing unit 31 in the memory, and stores the pattern matching results (the templates that are sequentially determined). The motion pattern of the work implement 7 is specified by analyzing the change of the work status (posture) of the boom 4, the arm 5 and the bucket 6 in time series. Here, since the boom 4, the arm 5, and the bucket 6 have their own movement patterns depending on the work contents, the work machine 7 stores the movement patterns for each work contents in advance and can recognize the work contents. When the movement pattern of the work machine 7 is specified, the work content recognition unit 32 recognizes the work content corresponding to the specified movement pattern of the work machine 7. For example, when the movement pattern of the work machine 7 is a pattern in which soil is excavated by the bucket 6, the work content is recognized as being excavation / loading work.

  Next, processing by the controller 10 will be described. First, the work content table 15 used by the accumulating unit 11 will be described. As shown in FIG. 8, the work content table 15 includes work IDs identified by work IDs such as “excavation / loading”, “slope finish”, “planar finish”, “travel”, “child split”, and “crane”. Each has a working flag column 15a, a cycle number (work number) column 15b, a fuel consumption (used fuel amount) column 15c, a work time column 15d, a fuel consumption column 15e per cycle, and a work time column 15f per cycle. .

  Also, the target fuel consumption table 16 by work content used by the evaluation unit 13 will be described. As shown in FIG. 9, the target fuel consumption table 16 by work content is specified by work IDs such as “excavation / loading”, “slope finish”, “planar finish”, “travel”, “child split”, and “crane”. Each work content has a target fuel consumption column 16a per cycle, and the value of each target fuel consumption is set in advance. Note that the numerical examples shown in FIGS. 8 and 9 do not necessarily represent realistic numerical examples.

  When the engine 21 of the excavator 1 is turned on, the accumulating unit 11 performs data collection / accumulation processing for each work content in parallel with the above-described processing by the image processing controller 20, as shown in the flowchart of FIG. Do. First, when the engine 21 is turned on, it is determined whether or not it is the first engine start of the day (step S201). If it is the first engine start (step S201; Yes), each column of the work content table 15 is determined. Is stored (step S202). If it is not the first engine start (step S201; No), the process of step S202 is jumped.

  Thereafter, it is checked whether or not work content information is input from the work content identification unit 32 of the image processing controller 20 (step S203). When the current work content is identified by the work content identification unit 32 and the identified work content information is input (step S203; Yes), the work flag is turned on (for the corresponding work content in the work content table 15). “1” is set in the working flag column 15a), and the start time is stored (step S204). Further, by integrating the fuel injection command value obtained from the vehicle body controller 30, the fuel usage amount of the engine 21 for the work content is accumulated (step S205). Then, it is checked whether or not there is a work end instruction when the engine is turned off or the evaluation display button 18 is pressed (step S206).

  If there is no work end instruction (step S206; No), it is checked whether or not the next work content information is input from the work content identification unit 32 (step S207). If the work content information is not input (step S207; No), the integration of the fuel injection command value is continued assuming that the current work content is continued (step S205).

  On the other hand, when the next work content information is input with the progress of the work by the work machine 7 (step S207; Yes), it is determined whether or not the input work content is the same as the previous work content. Determination is made (step S208). If it is not the same as the previous work content (step S208; No), the previous work content is turned off from the difference from the above-mentioned start time by turning off the working flag of the previous work content and storing the end time. The work time is calculated, the work time is added to the work time column 15d of the corresponding work content in the work content table 15, and the integrated value of the fuel injection command value is added to the corresponding work content in the work content table 15. Is added to the fuel consumption amount column 15c (step S209). Subsequently, the in-work flag of the current work content is turned on and the start time is stored (step S210). Further, by integrating the fuel injection command value obtained from the vehicle body controller 30, the fuel usage amount of the engine 21 for the current work content is accumulated (step S211). Then, it is checked whether or not there is a work end instruction by turning off the engine or pressing the evaluation display button 18 (step S212).

  If it is the same as the previous work content (step S208; Yes), the cycle number field 15b of the previous work content in the work content table 15 is incremented by 1 and added (step S213). Also, the integration of the fuel injection command value is continued on the assumption that the previous work content is continued (step S205).

  By repeating such processing every time the next work content information is input, the number of cycles, fuel consumption, and work time are sequentially accumulated in the work content table 15 for each work content.

  Finally, when there is a work end instruction by turning off the engine or pressing the evaluation display button 18 (step S206; Yes, or step S212; Yes), the current work flag is turned off, By storing the end time, the work time of the work content is calculated from the difference from the start time described above, and this work time is added to the work time column 15d of the corresponding work content in the work content table 15, and Then, the integrated value of the fuel injection command value is added to the fuel consumption amount column 15c of the corresponding work content in the work content table 15 (step S214).

  The fuel consumption calculation unit 12 and the evaluation unit 13 are activated at a timing when a specific fuel saving diagnosis display request event such as pressing the evaluation display button 18 occurs, and performs processing and control as shown in FIG. First, when there is a specific fuel saving diagnosis display request event, the fuel consumption calculation unit 12 uses information on the number of cycles and fuel consumption accumulated in the work content table 15 to calculate the fuel consumption per cycle for each work content. F [L / cycle] is calculated and stored in the fuel consumption column 15e per cycle (step S301). That is, the fuel consumption is calculated as an average fuel consumption per cycle. At this time, using the information on the number of cycles and the work time, the work time [h / cycle] per cycle is calculated for each work content and stored in the work time column 15f per cycle.

  When the calculation of the fuel consumption F for each work content is completed, the evaluation unit 13 reads information on the target fuel consumption Ft per cycle for each work content from the target fuel consumption table 16 for each work content (step S302). Then, the fuel efficiency F calculated for each work content is compared with the target fuel efficiency Ft (step S303). As a result of the comparison, if F ≧ Ft is not satisfied (if F <Ft) (step S303; No), the target fuel consumption is cleared for the work content, and thus the fuel-saving operation is performed as the fuel usage state of the work content. Displays “good” evaluation on the monitor 14 (step S304). On the other hand, if F ≧ Ft as a result of the comparison (step S303; Yes), the target fuel consumption is not cleared for the work content, so the fuel-saving operation is evaluated as “bad” as the fuel usage state of the work content. A message to that effect is displayed on the monitor 14 (step S305).

  As a result of the comparison, when F ≧ Ft (step S303; Yes), a level determination indicating the degree of deviation of the fuel consumption F from the target fuel consumption Ft is performed, and “slightly bad” “ You may make it display the step-wise fuel-saving driving evaluation like "bad" and "impossible".

  In addition, when a target work time per cycle for each work content is set in advance in the target fuel efficiency table 16 for each work content and F ≧ Ft is not satisfied (step S303; No), per cycle calculated for each work content. By adding a step of comparing the work time with the target work time per cycle, the quality of the fuel-saving driving for each work content may be evaluated in consideration of the work time. Even in the case of F ≧ Ft (step S303; Yes), by adding a step of comparing the work time per cycle calculated for each work content with the target work time per cycle, the work time can be improved. You may make it judge whether it is a bad factor.

  Moreover, in this Embodiment, although the camera 9 was installed on the cockpit 3a of the upper turning body 3, if it is a location which can image | photograph the working condition of the working machine 7, it may install in an appropriate location. Good. For example, the arm 5, the bucket 6, and the upper swing body 3 can be photographed by being installed on the boom 4, the state of the arm 5 and the bucket 6 with respect to the boom 4 is grasped, and the state of the boom 4 with respect to the upper swing body 3 is grasped. Then, the posture of the work machine 7 may be specified.

1 is a schematic side view showing a configuration example of a hydraulic excavator to which a fuel-saving driving support method according to an embodiment of the present invention is applied. It is a schematic block diagram which shows the structural example of the fuel-saving driving | operation assistance system of a hydraulic excavator. It is explanatory drawing about the angle which shows the displacement condition of a working machine. It is explanatory drawing which shows the example of the template image of a boom. It is explanatory drawing which shows the example of the template image of an arm. It is explanatory drawing which shows the example of the template image of a bucket. It is a schematic flowchart which shows the example of a pattern matching process. It is explanatory drawing which shows a work content table. It is explanatory drawing which shows the target fuel consumption table according to work content. It is a schematic flowchart which shows the process example by an accumulation part. It is a schematic flowchart which shows the process example by a fuel consumption calculation part and an evaluation part.

Explanation of symbols

4 Boom 5 Arm 6 Bucket 7 Working machine 9 Camera 11 Accumulation unit 12 Fuel consumption calculation unit 13 Evaluation unit 14 Monitor 15 Work content table 31 Pattern matching processing unit 32 Work content recognition unit 33 Template image memory

Claims (6)

A process of sequentially photographing the work status of the work machine with a camera mounted on the construction machine,
The work content of the work implement is determined from the movement of the work implement according to the pattern matching results obtained in time series by sequentially pattern matching the sequentially captured camera image and the template image stored in advance according to the work status of the work implement. The process of recognizing;
A process of accumulating the number of operations and the amount of fuel used for each of the recognized work details;
Comparing the fuel consumption for each work content calculated based on the accumulated number of operations and the amount of fuel used at a predetermined timing with the target fuel consumption preset for each work content, the quality of fuel usage for each work content is determined. A process of displaying;
A fuel-saving driving support method for a construction machine, comprising:   The step of recognizing uses a photographed camera image for each part of the work implement and a template image for each part stored in advance according to the work situation of the work implement. 2. A fuel-saving driving support method for a construction machine according to 1.   3. The fuel-saving driving support method for a construction machine according to claim 1, wherein in the step of accumulating, the work time is further accumulated for each work content. A camera that is mounted on a construction machine and sequentially captures the work status of the work implement;
A template image memory pre-stored with a template image corresponding to the work status of the work implement;
Recognizing the work content of the work implement from the movement of the work implement according to the pattern matching result obtained in time series by sequentially pattern matching the camera image sequentially taken by the camera and the template image stored in the template image memory in advance. Work content recognition means to
Accumulating means for accumulating the number of operations and the amount of fuel used for each work content sequentially recognized by the work content recognition means in a work content table;
A fuel consumption calculating means for calculating a fuel consumption for each work content based on the number of operations accumulated in the work content table and the amount of fuel used at a predetermined timing;
Display means for comparing the fuel consumption for each work content calculated by the fuel consumption calculation means with a target fuel consumption set in advance for each work content and displaying the quality of the fuel usage state for each work content;
A fuel-saving driving support system for construction machinery, characterized by comprising: The template image memory stores in advance a template image for each part according to the work status of the work implement,
The said work content recognition means uses the camera image for every part of the said working machine image | photographed, The fuel-saving driving | operation assistance system of the construction machine of Claim 4 characterized by the above-mentioned.   6. The fuel-saving driving support system for a construction machine according to claim 4, wherein the accumulating unit further accumulates the work time for each work content.

Images