JP5642324B2 - Vehicle approach notification device - Google Patents

Description

  The present invention relates to a vehicle approach notification device for generating a sound and notifying a pedestrian or the like of an electric mobile body having high quietness such as a hybrid vehicle, an electric vehicle, and an electric motorcycle.

  In recent years, following the development and practical use of electric bicycles and electric carts, vehicles such as electric motorcycles and electric automobiles are being electrified. Specifically, instead of a vehicle using an internal combustion engine as a power source, charging is performed by a hybrid vehicle using a gasoline engine and an electric motor as a power source, or a charger installed at a household power source or a gasoline station or power supply station. Electric vehicles powered by electric motors powered by batteries, or fuel cell vehicles that run while generating power from fuel cells that use hydrogen gas or the like as fuel, are being developed sequentially. Some of them have already been put into practical use and are beginning to spread.

  Conventional gasoline and diesel vehicles that use an internal combustion engine as a power source generate engine noise and exhaust noise emitted by the power source itself, and road noise while driving. A person on the road can recognize the approach of the vehicle by the engine sound or exhaust sound of the automobile. However, in the case of a hybrid vehicle, when driving at a low speed, it is not a driving mode by an engine but a driving mode by an electric motor, so that no engine noise or exhaust noise is generated. Since the vehicle is driven by the electric motor, any of the automobiles is an electric mobile body that is extremely quiet. However, pedestrians, bicycle drivers, etc. that exist around such moving bodies with excellent quietness, such as hybrid cars, electric cars, fuel cell cars, etc. that run with an electric motor with low noise generation and high quietness Since the approach of the electric vehicle cannot be recognized by sound, a contact accident between a highly quiet moving body and a pedestrian or the like may occur.

  For this reason, in order to solve the above-mentioned problems in which the quietness of hybrid vehicles, fuel cell vehicles, electric vehicles, etc. is harmful, other than the horn that is provided in conventional vehicles etc. and issues a warning at the driver's will Various vehicle approach notification devices have been proposed that operate independently of the driver's intention.

  In some vehicle approach notification devices, a notification sound for notifying a pedestrian or the like of the presence of the host vehicle is generated as a sound simulating a conventional engine sound. For example, Patent Document 1 has a frequency according to the motor speed, a pseudo sound signal having an amplitude according to the accelerator opening, and a frequency according to the vehicle speed detected by the vehicle speed sensor, and the accelerator opening. A pseudo sound signal having an amplitude corresponding to the frequency is generated by a computer and output from a speaker via an amplifier. Whether to use a pseudo sound based on the motor speed or a pseudo sound based on the vehicle speed is selected by a switch. Further, it is described that a pseudo sound having a frequency based on the motor speed and a frequency based on the vehicle speed may be generated. Patent Document 2 describes a technique for controlling an alarm sound by determining an emergency level of a brake operation based on information on an accelerator and a brake.

JP-A-7-32948 JP 2005-75182 A

  The notification sound generated from the conventional vehicle approach notification device changes the pitch and volume in conjunction with signals obtained from the vehicle such as a vehicle speed signal, an accelerator opening signal, and brake information. When these vehicle signals have poor time resolution of the acquisition period, that is, when the acquisition period of the vehicle signal is long, there is a problem that the change of the notification sound is not smooth and becomes stepwise. As a method for solving this, it is conceivable to use arithmetic processing such as interpolation and extrapolation. However, these methods have the following problems. Interpolation can estimate a value with a smaller error, but causes a significant delay of at least one period. On the other hand, the extrapolation does not cause a delay, but the estimation error is large, so that a sound skip phenomenon occurs. The sound skip here means that the change in volume and pitch becomes discontinuous. Various vehicle approach reporting devices have been proposed, but a vehicle approach reporting device that improves the time resolution of the acquired signal acquired by the vehicle approach reporting device and solves the problems of conventional interpolation and extrapolation is proposed. It has not been.

  The present invention has been made in order to solve the above-described problems. Even when the time resolution of the acquisition signal of the vehicle approach notification device is poor, the time resolution is improved, and there is a problem of time delay and skipping. The purpose is to solve.

  A vehicle approach notification device according to the present invention includes a vehicle signal acquisition unit that acquires a vehicle signal of an electric vehicle for each acquisition cycle and outputs an acquisition signal, and a signal processing unit that performs signal processing of the acquisition signal, The signal processing unit calculates an estimated value of the vehicle signal to be acquired next time based on the inclination of the vehicle signal, and calculates the next estimated value that is the estimated value calculated this time and the current estimated value that is the previously calculated estimated value. And performing an interpolation process.

  According to the vehicle approach notification device according to the present invention, the estimated value of the vehicle signal to be acquired next time is calculated based on the inclination of the vehicle signal, and the interpolation process is performed using the next estimated value and the current estimated value. Even when the time resolution of the signal is poor, the time resolution can be improved and the problems of time delay and skipping can be solved.

It is a block diagram which shows the outline | summary of a structure of the vehicle approach notification apparatus by Embodiment 1 of this invention. It is a conceptual diagram which shows the vehicle approach notification apparatus mounted in the electrically-driven moving body. It is a figure which shows the example of the acquisition waveform of a vehicle signal waveform and a vehicle signal. It is a figure which shows the example of an interpolation process waveform. It is a flowchart of an interpolation process. It is a figure which shows the example of an extrapolation process waveform. It is a flowchart of an extrapolation process. It is a figure which shows the example of the signal processing waveform by Embodiment 1 of this invention. It is a flowchart of the signal processing by Embodiment 1 of this invention. It is a figure which shows the example of the signal processing waveform by Embodiment 2 of this invention. It is a flowchart of the signal processing by Embodiment 2 of this invention. It is a figure which shows the example of the signal processing waveform by Embodiment 3 of this invention. It is a flowchart of the signal processing by Embodiment 3 of this invention. It is a block diagram which shows the outline | summary of a structure of the vehicle approach notification apparatus by Embodiment 4 of this invention.

Embodiment 1 FIG.
FIG. 1 is a block diagram showing an outline of the configuration of a vehicle approach notification device according to Embodiment 1 of the present invention, and FIG. 2 is a conceptual diagram showing a vehicle approach notification device mounted on an electric vehicle. The vehicle approach notification device 100 is provided in an electric vehicle 200 that generates at least a part of driving force by an electric motor, such as an electric vehicle or a hybrid vehicle. The vehicle approach notification device 100 includes a notification sound control unit 10 that outputs a notification sound signal, and a sounding body 40 such as a speaker that generates notification sound outside the vehicle by the notification sound signal. The notification sound here indicates a sound reminiscent of the running state of the vehicle. The notification sound may or may not be a sound reminiscent of a conventional automobile engine sound.

  The notification sound control unit 10 acquires the vehicle signal 20 and outputs the acquisition signal of the vehicle signal 20, the signal processing unit 2 that performs signal processing of the acquisition signal, and the adjustment information from the adjustment information table 3. And an adjustment information acquisition unit 4 that calculates a pitch and volume magnification for changing the notification sound, and a notification sound generation unit 6 that generates notification sound from the notification sound data 5 based on the pitch and volume magnification. And a notification sound output unit 7 that outputs the notification sound generated by the notification sound generation unit 6 to the sounding body 40. The vehicle signal 20 is information indicating the behavior of the vehicle acquired from the vehicle. The vehicle signal 20 represents a vehicle speed signal, an accelerator opening signal, a brake signal, or a plurality of these signals. The vehicle signal 20 may be a signal acquired from a hard wire or a signal acquired from in-vehicle communication such as a CAN (Controller Area Network) bus or a LIN (Local Interconnect Network) bus. The vehicle signal 20 is acquired by the vehicle signal acquisition unit 1.

  The operation of the vehicle approach notification device 100 will be described. The vehicle signal acquisition unit 1 acquires the vehicle signal 20 every acquisition period T. The signal processing unit 2 performs signal processing for improving the time resolution of the acquisition signal of the vehicle signal 20 acquired by the vehicle signal acquisition unit 1. The signal processing unit 2 estimates the signal value of the next vehicle signal based on the inclination of the acquired vehicle signal 20, and performs an interpolation process between the estimated value and the previous estimated value, thereby obtaining the acquired signal of the vehicle signal 20. Improve time resolution. The adjustment information acquisition unit 4 calculates, from the adjustment information table 3, a pitch and a volume magnification for changing the notification sound in conjunction with the information (processed signal value) obtained from the signal processing unit 2. The adjustment information acquisition unit 4 also outputs the pitch and volume magnifications to the notification sound generation unit 6.

  The notification sound data 5 indicates a phoneme that is a source for generating the notification sound, and may be one or more. The phonemes are not limited to those reminiscent of conventional engine sounds, and may be anything such as sine waves, white noise, and melody sounds. The notification sound data may be data stored in a ROM (Read Only Memory) or a RAM (Random Access Memory) of an internal memory or an external memory, or may be data input in real time.

  The notification sound generation unit 6 adjusts the pitch and volume of the notification sound data 5 from the pitch and volume magnification acquired from the adjustment information acquisition unit 4, and outputs them to the notification sound output unit 7. When there are a plurality of notification sound data 5, the notification sound generation unit 6 may perform phoneme synthesis processing in addition to adjustment of pitch and volume. The notification sound output unit 7 adjusts the final sound pressure of the notification sound generated by the notification sound generation unit 6 and outputs the notification sound to the sounding body 40. The sounding body 40 indicates a speaker attached outside the vehicle for reproducing the notification sound. There may be one sounding body 40 or a plurality of sounding bodies 40.

  The signal processing unit 2 will be described in detail. The acquisition signal of the vehicle signal 20 will be described with reference to FIG. The acquisition signal of the vehicle signal 20 is a set of signal values (also referred to as acquisition values as appropriate) acquired from the vehicle signal 20 every acquisition period T. FIG. 3 is a diagram illustrating an example of a vehicle signal waveform and a vehicle signal acquisition waveform. The horizontal axis is time, and the vertical axis is signal value. FIG. 3 shows an example in which the acquisition period T of the vehicle signal 20 is 200 ms, and the time resolution of the acquisition signal in this case is 200 ms. The black circles in FIG. 3 are signal values of the vehicle signal 20 acquired by the vehicle signal acquisition unit 1. The waveform of the vehicle signal 20 is essentially smooth like the vehicle signal waveform 51, but when the time resolution of the acquisition signal in the vehicle signal acquisition unit 1 is poor, the waveform of the acquisition signal is stepwise like the acquisition waveform 52. It becomes a change. This gradual change finally appears in the notification sound. Therefore, the signal processing unit 2 performs a process of improving the time resolution of the acquired signal based on the information obtained from the signal value of the acquired vehicle signal 20 to approximate the shape (ideal value) of the vehicle signal waveform 51.

An interpolation process (also referred to as an interpolation process) when the time resolution of the acquisition signal of the vehicle signal 20 is poor will be described. FIG. 4 is a diagram showing an example of an interpolation processing waveform, and FIG. 5 is a flowchart of the interpolation processing. In FIG. 4, the horizontal axis is time, and the vertical axis is signal value. The interpolation processing waveform 53 is a waveform obtained by performing the interpolation processing by delaying the signal value of the vehicle signal 20 by the time delay 54. In the interpolation process, the processes in steps S001, S002, and S003 in FIG. 5 are performed. In step S001, the current (n-th sample) signal value x (n) and the previous (n-1 sample) signal value x (n-1) are acquired from the vehicle signal acquisition unit 1. In step S002, the change amount δ of the processing signal value is calculated, and the change amount δ is updated. If the number of expansion data that expands the number of data of the vehicle signal 20 for each acquisition period T is m, the amount of change δ can be expressed as in equation (1).
δ = (x (n) −x (n−1)) / (m + 1) (1)

For example, when the expansion data number m that expands the number of data of the vehicle signal 20 for each acquisition period T is 3, the amount of change δ can be expressed as Equation (2).
δ = (x (n) −x (n−1)) / 4 (2)

In step S003, the number of current processing signal values (n-th sample) is expanded as follows. When the number of extended data m is 3, the number of processed signal values is m + 1, that is, 4. Four processing signal values 0 (x0 (n)), processing signal value 1 (x1 (n)), processing signal value 2 (x2 (n)), and processing signal value 3 (x3 (n)) are respectively expressed by the formulas ( It can be expressed as 3) to (6).
x0 (n) = x (n-1) (3)
x1 (n) = x (n-1) + (m-2) * δ
= X (n-1) + δ (4)
x2 (n) = x (n-1) + (m-1) * δ
= X (n-1) + 2 * δ (5)
x3 (n) = x (n-1) + m * δ
= X (n-1) + 3 * δ (6)

  FIG. 4 shows an example in which the acquisition period T of the vehicle signal 20 is 200 ms and the number of extended data m is 3. By performing the interpolation process in the signal processing unit 2, the time resolution of the acquired signal is improved. The interpolation process delays the acquired value of the vehicle signal 20 by one cycle and linearly interpolates the value between the signal values, so that the error is small. However, since the signal value is delayed by one period, when the time resolution of the acquisition signal of the vehicle signal 20 is poor, for example, when the acquisition period T is 200 ms as shown in FIGS. End up. For this reason, the interpolation process acquires a signal such as an accelerator opening that requires a response as fast as possible, and is not suitable for improving the time resolution of the acquired signal.

  An extrapolation process (also called extrapolation process) when the time resolution of the acquisition signal of the vehicle signal 20 is poor will be described. FIG. 6 is a diagram illustrating an example of an extrapolation processing waveform, and FIG. 7 is a flowchart of the extrapolation processing. In FIG. 6, the horizontal axis is time, and the vertical axis is signal value. The extrapolation processing waveform 55 is a waveform obtained by performing extrapolation processing on the signal value of the vehicle signal 20. In the extrapolation process, the processes of steps S001, S002, and S004 in FIG. 7 are performed. Steps S001 and S002 of the extrapolation process are the same as the interpolation process of FIG.

In step S004, the number of current processing signal values (n-th sample) is expanded as follows. When the number of extended data m is 3, the number of processed signal values is m + 1, that is, 4. Four processing signal values 0 (x0 (n)), processing signal value 1 (x1 (n)), processing signal value 2 (x2 (n)), and processing signal value 3 (x3 (n)) are respectively expressed by the formulas ( 7) to (10).
x0 (n) = x (n) (7)
x1 (n) = x (n) + (m−2) * δ
= X (n) + δ (8)
x2 (n) = x (n) + (m−1) * δ
= X (n) + 2 * δ (9)
x3 (n) = x (n) + m * δ
= X (n) + 3 * δ (10)

  FIG. 6 is an example in which the acquisition cycle T of the vehicle signal 20 is 200 ms and the number of extended data m is 3, as in FIG. By performing extrapolation processing in the signal processing unit 2, the time resolution of the acquired signal is improved. The extrapolation process estimates the signal value after one cycle (next) from the signal value acquired one cycle before (previous) and the current signal value, and linearly interpolates between the estimated value and the current acquired value. . Therefore, unlike the interpolation process, no time delay occurs. However, since the signal value after one cycle is estimated, the error is large, and the signal value in the case where the signal value is discontinuous as shown by the broken lines 56 and 57 in FIG. A value that rises from a descent (an out-of-trend rise value) occurs. This discontinuous value appears in the notification sound as a skipping phenomenon. For this reason, the extrapolation process is not suitable for the process of improving the time resolution of the acquisition signal of the vehicle signal 20 with respect to the vehicle signal in which the vehicle signal waveform 51 of the vehicle signal 20 changes abruptly.

  A method for solving the respective disadvantages of the interpolation processing and extrapolation processing described above will be described. FIG. 8 is a diagram showing an example of signal processing waveforms according to the first embodiment of the present invention, and FIG. 9 is a flowchart of signal processing according to the first embodiment of the present invention. The signal processing performed in the present invention will be referred to as estimated value interpolation processing. In FIG. 8, the horizontal axis is time, and the vertical axis is signal value. The signal processing waveform 58 is a waveform obtained by performing the estimated value interpolation processing of the first embodiment on the signal value of the vehicle signal 20. In the estimated value interpolation process of the first embodiment, the processes of steps S101, S102, S103, and S104 in FIG. 9 are performed. Step S101 of the estimated value interpolation process is the same as step S001 of the interpolation process and the extrapolation process.

In step S101, the current (n-th sample) signal value x (n) (current acquisition value) and the previous (n-1 sample) signal value x (n-1) (previous time) are obtained from the vehicle signal acquisition unit 1. Acquired value). In step S102, an estimated value est_x (n + 1) (next estimated value) obtained by estimating the next (n + 1 sample) signal value is calculated as in Expression (11).
est_x (n + 1) = (x (n) −x (n−1)) + x (n)
= 2 * x (n) -x (n-1) (11)
Note that the initial value (initial value) of the signal value x (n−1) is 0 km / h if the vehicle signal 20 is a vehicle speed signal, and 0% if the vehicle signal 20 is an accelerator opening. Also in the case where the vehicle signal 20 is another vehicle signal, the initial state when the vehicle is started may be set as the initial value of the signal value x (n−1). Since the signal value x (n) and the signal value x (n−1) are acquired every certain acquisition period T, the estimated value est_x (n + 1) is the slope (x (n) −x (n) of the vehicle signal 20. -1)) / T and the gradient per cycle (x (n) -x (n-1)). The inclination per period is the same as the difference between the current acquired value (x (n)) of the acquired signal acquired this time and the previous acquired value (x (n-1)) of the acquired signal acquired last time.

In step S103, the change amount δ of the processing signal value is calculated, and the change amount δ is updated. Assuming that the number of extended data that expands the number of data of the vehicle signal 20 for each acquisition period T is m, the estimated value est_x (n + 1) that is the next estimated value and the estimated value est_x (n) that is the current estimated value calculated last time The amount of change δ can be expressed as shown in Equation (12).
δ = (est_x (n + 1) −est_x (n)) / (m + 1) (12)

For example, if the number of expansion data m that expands the number of data of the vehicle signal 20 for each acquisition period T is 3, the amount of change δ can be expressed as in Expression (13).
δ = (est_x (n + 1) −est_x (n)) / 4 (13)

In step S104, the number of processing signal values at the present time (nth sample) is expanded as follows. When the number of extended data m is 3, the number of processed signal values is m + 1, that is, 4. Four processing signal values 0 (x0 (n)), processing signal value 1 (x1 (n)), processing signal value 2 (x2 (n)), and processing signal value 3 (x3 (n)) are respectively expressed by the formulas ( 14) to (17).
x0 (n) = est_x (n) (14)
x1 (n) = est_x (n) + (m−2) * δ
= Est_x (n) + δ (15)
x2 (n) = est_x (n) + (m−1) * δ
= Est_x (n) + 2 * δ (16)
x3 (n) = est_x (n) + m * δ
= Est_x (n) + 3 * δ (17)

  FIG. 8 shows an example in which the acquisition period T of the vehicle signal 20 is 200 ms and the number of extended data m is 3, as in FIGS. 4 and 6. By performing the estimated value interpolation process in the signal processing unit 2, the time resolution of the acquired signal is improved, the delay as in the interpolation process does not occur, and the sound skip phenomenon as in the extrapolation process does not occur. . The estimated value interpolation process estimates the future signal value as in the extrapolation process, but estimates the signal value after one period (next) from the signal value acquired one period before (previous) and the current signal value. . Unlike the extrapolation process, the estimated value interpolation process performs linear interpolation using the previous estimated value est_x (n−1) with the current estimated value est_x (n). Thereby, the signal processing of the first embodiment improves the time resolution of the acquired signal, does not cause a delay as in the interpolation process, and can prevent the occurrence of a sound skip phenomenon as in the extrapolation process. . That is, the signal processing of the first embodiment can improve the time resolution and solve the problems of time delay and skipping even when the time resolution of the acquired signal is poor. For this reason, the signal processing (estimated value interpolation processing) of the first embodiment can cope with a vehicle signal 20 that requires a quick response or has a sudden change.

  The estimated value interpolation process will be described in detail in comparison with the extrapolation process estimation. The estimated value est_x (n + 1) for the next time (n + 1 sample) in the first embodiment is obtained as the signal value acquired this time (n sample) and the previous time (n-1 sample) as shown in the equation (11). It is calculated by adding the signal value x (n) acquired this time (nth sample) to the difference (x (n) −x (n−1)) from the measured signal value. In the extrapolation process, the next (n + 1 sample) signal value is not actually estimated, but when the next (n + 1 sample) signal value is estimated by applying the extrapolation processing method, the following temporary estimation is performed: The value can be calculated. This temporary estimated value is x (n) +4 * ((x (n) −x (n−1)) / 4) = x (n) + (x (n) −x (n−1)). . This temporary estimated value is the same as the estimated value est_x (n + 1) of the estimated value interpolation process.

  In the extrapolation process, since the signal value x (n + 1) acquired at the time corresponding to the temporary estimated value is set to the processing signal value 0 without using the temporary estimated value, the temporary estimated value and the signal value x (n + 1) are If the difference is large, a sudden change will occur. When the sign of the slope of the vehicle signal waveform 51 changes, as shown by the broken line graphics 56 and 57 in FIG. 6, both the current (n-th sample) and the next (n + 1 sample) change amount δ are positive. Despite being (or minus), the slope of the processing signal value 3 of this time (n-th sample) and the processing signal value 0 of the next time (n + 1-th sample) becomes minus (or plus), and the broken line diagram 56 in FIG. , 57, the discontinuity of the code occurs in the slope of the processed signal. On the other hand, in the estimated value interpolation processing, the current process signal value 0 (n-th sample) is set to the estimated value est_x (n) estimated last time (n-1 sample), so the previous (n-1 sample) The change between the processing signal value 3 of (eye) and the processing signal value 0 of the current time (n-th sample) is less than that of the extrapolation processing. Further, there is no sign discontinuity in the slope of the processed signal (signal processing waveform). Therefore, in the estimated value interpolation processing, it is possible to prevent the occurrence of a sound skip phenomenon such as extrapolation processing.

  According to the vehicle approach notification device 100 of the first embodiment, the vehicle signal 20 of the electric vehicle 200 is acquired at every acquisition cycle T, and the acquisition signal is processed by the vehicle signal acquisition unit 1 that outputs the acquisition signal. The signal processing unit 2 calculates an estimated value (estimated value est_x (n + 1)) of the vehicle signal 20 to be acquired next time based on the inclination of the vehicle signal 20, and the estimated value calculated this time Since the next estimated value (estimated value est_x (n + 1)) and the current estimated value (estimated value est_x (n)) which is the previously calculated estimated value are subjected to interpolation processing, Even when the time resolution of the acquired signal is poor, the time resolution can be improved and the problems of time delay and skipping can be solved.

Embodiment 2. FIG.
FIG. 10 is a diagram showing an example of signal processing waveforms according to the second embodiment of the present invention, and FIG. 11 is a flowchart of signal processing according to the second embodiment of the present invention. In the signal processing of the second embodiment, the estimated value error can be made smaller than that of the first embodiment by adding a coefficient when deriving the estimated value. In FIG. 10, the horizontal axis is time, and the vertical axis is signal value. The signal processing waveform 59 is a waveform obtained by performing the estimated value interpolation processing of the second embodiment on the signal value of the vehicle signal 20 when a coefficient Coef described later is set to 0.8. In the estimated value interpolation process of the second embodiment, the processes of steps S101, S102, S103, and S104 in FIG. 11 are performed. Steps S101, S103, and S104 of the estimated value interpolation process of the second embodiment are the same as the estimated value interpolation process of the first embodiment.

In step S102, an estimated value est_x (n + 1) obtained by estimating the next (n + 1 sample) signal value is calculated as in Expression (18).
est_x (n + 1)
= (X (n) -x (n-1)) * Coef + x (n)
= (1 + Coef) * x (n) -x (n-1) (18)
Here, Coef is a coefficient and is a number larger than 0 and smaller than 1.

  The estimated value est_x (n + 1) in Embodiment 2 is the difference between the current acquired value (x (n)) of the acquired signal acquired this time and the previous acquired value (x (n-1)) of the acquired signal acquired last time. It is obtained by an operation of adding the current acquired value (x (n)) to the correction difference multiplied by the coefficient Coef. Further, the estimated value est_x (n + 1) in the second embodiment is the slope (x (n) −x (n−1)) / T of the vehicle signal 20 or the slope per cycle (x (n) −x (n−)). This is based on a correction slope obtained by multiplying 1)) by the coefficient Coef.

  The next-time (n + 1 sample) estimated value est_x (n + 1) in the estimated value interpolation process of the second embodiment is slightly less trackable than the first embodiment in which there is no coefficient, ie, signal processing. Although the signal processing waveform 59 has a delay time slightly longer than that of the waveform 58, it is possible to reduce the estimated value error when the acquired signal value changes greatly. The vehicle approach notification device 100 according to the second embodiment is slightly less trackable than the first embodiment, but can reduce the estimated value error when the acquired signal value changes greatly. Note that the coefficient Coef may be determined from, for example, an assumed use situation of the vehicle. The smaller the value of the coefficient Coef, the smaller the error, but a time delay occurs. Therefore, the coefficient Coef is preferably calculated from the maximum allowable time delay of the assumed change in the vehicle signal and the change in the notification sound associated therewith. 0.8 is an example of a coefficient Coef having a good balance between time delay and error in a certain vehicle.

Embodiment 3 FIG.
FIG. 12 is a diagram showing an example of signal processing waveforms according to the third embodiment of the present invention, and FIG. 13 is a flowchart of signal processing according to the third embodiment of the present invention. There is a possibility that the estimated value calculated by the estimated value interpolation process of the first embodiment is larger than the range that the original vehicle signal 20 can take. Therefore, in the third embodiment, the limiter process is performed so that the estimated value does not become a value larger than the predetermined threshold value or does not become a value smaller than the predetermined threshold value. In FIG. 12, the horizontal axis is time, and the vertical axis is signal value. The signal processing waveform 60 is a waveform obtained by performing the estimated value interpolation processing of the third embodiment when the signal value threshold is set to 10 with respect to the signal value of the vehicle signal 20 and the signal value is not larger than 10. It is. In the estimated value interpolation process of the third embodiment, the processes of steps S101, S102, S103, S104, and S105 of FIG. 13 are performed. Steps S101, S102, S103, and S104 of the estimated value interpolation process of the third embodiment are the same as the estimated value interpolation process of the first embodiment.

  In step S105, limiter processing is performed. For example, a case where an upper limit is set for the signal value will be described. The threshold value of the signal value is set to the upper limit threshold value s1, and when each processed signal value calculated in step S104 is larger than the upper limit threshold value s1, each processed signal value is set to the upper limit threshold value s1. Moreover, when providing a lower limit in a signal value, it carries out as follows, for example. The threshold value of the signal value is set to the lower limit threshold value s2, and when each processed signal value calculated in step S104 is smaller than the lower limit threshold value s2, each processed signal value is set to the lower limit threshold value s2. When an upper limit and a lower limit are provided for the signal value, for example, the following is performed. The limiter process is performed by setting the threshold value on the side with the larger signal value as the upper limit threshold value s1 and the threshold value on the smaller side as the lower limit threshold value s2. In the limiter process, when each processing signal value calculated in step S104 becomes larger than the upper limit threshold s1, each processing signal value is set to the upper limit threshold s1, and each processing signal value calculated in step S104 is more than the lower limit threshold s2. When it becomes smaller, each processing signal value is set to the lower limit threshold s2. Thereby, the vehicle approach notification apparatus 100 of Embodiment 3 can prevent an abnormal value from being output after the signal processing for improving the time resolution of the acquisition signal of the vehicle signal 20.

  For example, when the vehicle signal 20 is the accelerator opening, the minimum value of the original input range is 0%, and the maximum value of the original input range is 100%. In this case, the upper limit threshold s1 is 100% and the lower limit threshold s2 is 0%. Therefore, if the processing signal value calculated in step S104 is estimated to be smaller than 0%, which is the minimum value of the original input range, the processing signal value may be processed to be 0%. Further, when the processing signal value calculated in step S104 is estimated to be larger than 100% which is the maximum value of the original input range, the processing signal value may be processed to be 100%. Similarly, when the vehicle speed signal is estimated to be smaller than 0 km / h, which is the minimum value of the original input range, the vehicle speed signal may be processed to be 0 km / h.

Embodiment 4 FIG.
FIG. 14 is a block diagram showing an outline of the configuration of the vehicle approach notification device according to Embodiment 4 of the present invention. The vehicle approach notification device 100 of the fourth embodiment differs from the vehicle approach notification device 100 of the first to third embodiments in that a filter processing unit 11 is provided between the signal processing unit 2 and the adjustment information acquisition unit 4. . The filter processing unit 11 performs filter processing such as FIR (Finite Impulse Response) and IIR (Infinite Impulse Response). The vehicle approach notification device 100 according to the fourth embodiment performs a filtering process on the processed signal value after the signal processing for improving the time resolution of the acquired signal, so that the vehicle signal is smoother and more natural than the first to third embodiments. Can simulate change. However, since the digital filter has a high processing load, it must be implemented by a CPU (Central Processing Unit) having a high processing capability.

  It should be noted that the present invention can be freely combined with each other within the scope of the invention, and each embodiment can be appropriately modified or omitted.

DESCRIPTION OF SYMBOLS 1 ... Vehicle signal acquisition part, 2 ... Signal processing part, 11 ... Filter processing part, 20 ... Vehicle signal, 40 ... Sound generator, 100 ... Vehicle approach notification apparatus, 200 ... Electric vehicle, x (n), x (n -1) ... signal value, x0 (n), x1 (n), x2 (n), x3 (n) ... processed signal value, est_x (n + 1), est_x (n) ... estimated value, Coef ... coefficient, T ... Acquisition cycle, s1... Upper limit threshold, s2.

Claims (7)

A vehicle approach notification device that is provided in an electric mobile body that generates at least a part of driving force by an electric motor and emits a notification sound from a sounding body to the outside of the electric mobile body,
A vehicle signal acquisition unit that acquires a vehicle signal of the electric vehicle for each acquisition period and outputs an acquisition signal;
A signal processing unit that performs signal processing of the acquired signal,
The signal processing unit
Calculate an estimated value of the vehicle signal to be acquired next time based on the slope of the vehicle signal,
A vehicle approach notification device characterized in that an interpolation process is performed using a next estimated value that is the estimated value calculated this time and a current estimated value that is the estimated value calculated last time.   The signal processing unit calculates the estimated value of the vehicle signal by an operation of adding the current acquired value to the difference between the current acquired value of the acquired signal acquired this time and the previous acquired value of the acquired signal acquired last time. The vehicle approach notification device according to claim 1.   The vehicle approach notification device according to claim 1, wherein the signal processing unit calculates the estimated value of the vehicle signal to be acquired next time based on a corrected inclination obtained by multiplying the inclination of the vehicle signal by a coefficient.   The signal processing unit is configured to add the current acquired value to a correction difference obtained by multiplying a difference between a current acquired value of the acquired signal acquired this time and a previous acquired value of the acquired signal acquired last time by a coefficient. The vehicle approach notification device according to claim 3, wherein the estimated value is calculated. The signal processing unit
When the processing signal value calculated by executing the interpolation processing is larger than the upper threshold, the processing signal value is limited so as not to be larger than the upper threshold, or the processing signal value is smaller than the lower threshold. The vehicle approach notification device according to any one of claims 1 to 4, wherein the processing signal value is limited so as not to become smaller than the lower limit threshold when the value becomes. The signal processing unit
When the processing signal value calculated by executing the interpolation processing is larger than an upper threshold, the processing signal value is limited so as not to be larger than the upper threshold, and the processing signal value is smaller than the lower threshold. The vehicle approach notification device according to any one of claims 1 to 4, wherein the processing signal value is limited so as not to become smaller than the lower limit threshold when the value becomes.   The vehicle approach according to any one of claims 1 to 6, further comprising a filter processing unit that performs a filtering process on a processing signal obtained by processing the acquired signal by the signal processing unit. Notification device.

Images