Vehicle behavior evaluation device, vehicle behavior evaluation method, and computer program for vehicle behavior evaluation

Abstract

To provide a vehicle behavior evaluation device capable of appropriately evaluating behavior of a vehicle.SOLUTION: A vehicle behavior evaluation device includes a calculation unit 22 and an evaluation value setting unit 23. The calculation unit calculates at least one of a variation degree of a distribution of a behavior index value including at least one of acceleration / deceleration of a vehicle 2, an amount of change per unit time in the acceleration / deceleration of the vehicle 2, and an amount of change in a travel direction of the vehicle 2 and an amount of change per unit time in the travel direction of the vehicle 2, and the maximum value of the behavior index value based on vehicle behavior information that represents behavior of the vehicle 2 when the vehicle 2 travels in a predetermined road section. The evaluation value setting unit sets a smaller evaluation value with respect to the vehicle behavior information as a value calculated of the variation degree of the distribution of the behavior index value and the maximum value of the behavior index value is larger.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present invention relates to a vehicle behavior evaluation device, a vehicle behavior evaluation method, and a computer program for vehicle behavior evaluation that evaluate the behavior of a vehicle.

Background Art

[0002] It has been proposed to learn a model for determining control information of a vehicle in automatic driving control based on the behavior of the vehicle during manual driving by a driver (see Patent Document 1).

[0003] The driving support method disclosed in Patent Document 1 learns the driving characteristics in the driver's manual driving and reflects the learning result in the driving characteristics of the automatic driving control. At that time, this driving support method detects the driving characteristics of the area where the autonomous vehicle travels, adjusts the learning result according to the detected driving characteristics of the area, and executes the automatic driving control based on the adjusted learning result.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] In order to train a model so that it can output appropriate control information, it is preferable that vehicle behavior information representing the behavior of the vehicle when appropriate driving is performed be used for training the model. However, even when multiple vehicles are traveling on the same road section, the behavior of each vehicle differs depending on the driver or the surrounding situation of each vehicle during that travel. Therefore, the behavior of the vehicle represented by the vehicle behavior information may not necessarily be suitable for use in training the model. On the other hand, if we try to evaluate individual vehicle behavior information manually, it will require an extremely large amount of man-hours. Therefore, it is required to automatically evaluate the degree of suitability of individual vehicle behavior information for model training.

[0006] Therefore, an object of the present invention is to provide a vehicle behavior evaluation device capable of appropriately evaluating the behavior of a vehicle.

Means for Solving the Problems

[0007] According to one embodiment, a vehicle behavior evaluation device is provided. This vehicle behavior evaluation device calculates at least one of the degree of variation in the distribution of behavior index values including at least one of the vehicle's acceleration / deceleration, the amount of change in the vehicle's acceleration / deceleration per unit time, the amount of change in the vehicle's traveling direction, and the amount of change in the vehicle's traveling direction per unit time based on vehicle behavior information representing the behavior of the vehicle when the vehicle travels on a predetermined road section, and a calculation unit that calculates at least one of the maximum value of the behavior index value, and an evaluation value setting unit that sets a smaller evaluation value for the vehicle behavior information as the calculated value of the degree of variation in the distribution of the behavior index value and the maximum value of the behavior index value is larger.

[0008] According to another embodiment, a vehicle behavior evaluation method is provided. This vehicle behavior evaluation method is based on vehicle behavior information representing the behavior of a vehicle when the vehicle travels through a predetermined road section, and calculates at least one of the acceleration and deceleration of the vehicle, the change amount of the acceleration and deceleration of the vehicle per unit time, the change amount of the traveling direction of the vehicle, and the change amount of the traveling direction of the vehicle per unit time. The degree of variation of the distribution of the behavior index value and at least one of the maximum values of the behavior index value are calculated. The larger the calculated value of the degree of variation of the distribution of the behavior index value and the maximum value of the behavior index value, the smaller the evaluation value is set for the vehicle behavior information.

[0009] According to still another embodiment, a computer program for vehicle behavior evaluation is provided. This computer program for vehicle behavior evaluation is based on vehicle behavior information representing the behavior of a vehicle when the vehicle travels through a predetermined road section, and calculates at least one of the acceleration and deceleration of the vehicle, the change amount of the acceleration and deceleration of the vehicle per unit time, the change amount of the traveling direction of the vehicle, and the change amount of the traveling direction of the vehicle per unit time. The degree of variation of the distribution of the behavior index value and at least one of the maximum values of the behavior index value are calculated. The larger the calculated value of the degree of variation of the distribution of the behavior index value and the maximum value of the behavior index value, the smaller the evaluation value is set for the vehicle behavior information. The computer is caused to execute instructions including this.

Advantages of the Invention

[0010] The vehicle behavior evaluation device according to the present disclosure has the effect of being able to appropriately evaluate the behavior of a vehicle.

Brief Description of the Drawings

[0011]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Embodiments for Carrying Out the Invention

[0012] Hereinafter, with reference to the drawings, a vehicle behavior evaluation device, a vehicle behavior evaluation method executed by the vehicle behavior evaluation device, and a computer program for vehicle behavior evaluation will be described. This vehicle behavior evaluation device sets an evaluation value for vehicle behavior information representing the behavior of a vehicle when the vehicle travels through a predetermined road section. At that time, this vehicle behavior evaluation device calculates at least one of the degree of variation in the distribution of behavior index values and the maximum value of the behavior index values based on the vehicle behavior information. Then, this vehicle behavior evaluation device sets a smaller evaluation value for the vehicle behavior information as the degree of variation in the distribution of behavior index values is larger, or as the maximum value of the behavior index values is larger. Note that the behavior index value is a value of an index representing the driving behavior of the vehicle and includes at least one of the acceleration and deceleration of the vehicle, the amount of change in acceleration and deceleration per unit time, the amount of change in the traveling direction of the vehicle, and the amount of change in the traveling direction of the vehicle per unit time.

[0013] The vehicle behavior information is used, for example, as teacher data when learning a control model used in the automatic driving control of a vehicle. Such a control model is configured, for example, as a so-called deep neural network (DNN). Then, the control model receives as input information on a target road section (radius of curvature, lane width, speed limit, etc.) represented on a map or an image obtained by photographing the road section with an in-vehicle camera, and outputs a planned driving trajectory of the vehicle or control information of the vehicle. Therefore, the evaluation value set for the vehicle behavior information represents, for example, the degree of usefulness when using the vehicle behavior information as teacher data for the above control model. Accordingly, for example, only the vehicle behavior information with an evaluation value equal to or higher than a certain level is used as teacher data. And information such as the accelerator opening, brake pedal force, steering angle of the steering wheel, or the driving trajectory of the vehicle included in the vehicle behavior information is used during the learning of the control model.

[0014] FIG. 1 is a schematic configuration diagram of a vehicle behavior evaluation system in which a vehicle behavior evaluation device is implemented. In the present embodiment, the vehicle behavior evaluation system 1 includes at least one vehicle 2 and a server 3 which is an example of a vehicle behavior evaluation device. Each vehicle 2 is connected to the server 3 via the wireless base station 5 and the communication network 4 by accessing, for example, a wireless base station 5 connected via a communication network 4 to which the server 3 is connected and a gateway (not shown). In FIG. 1, only one vehicle 2 is shown for simplicity, but the vehicle behavior evaluation system 1 may have a plurality of vehicles 2. Similarly, in FIG. 1, only one wireless base station 5 is shown, but a plurality of wireless base stations 5 may be connected to the communication network 4.

[0015] The vehicle 2 has at least one vehicle behavior sensor, a GPS receiver, a vehicle behavior recording device, and a wireless communication terminal.

[0016] The vehicle behavior sensor is a sensor for detecting the behavior of vehicle 2, and includes, for example, at least one of a speed sensor, an acceleration sensor, a gyro sensor, a sensor for detecting the accelerator opening, a sensor for detecting the brake pedal force, and a sensor for detecting the steering angle of the steering. And each time the vehicle behavior sensor generates a behavior sensor signal representing the behavior of vehicle 2, the generated behavior sensor signal is output to the vehicle behavior recording device. The behavior sensor signal represents at least one of speed, acceleration in the traveling direction (hereinafter sometimes referred to as longitudinal G), acceleration in the direction orthogonal to the traveling direction (hereinafter sometimes referred to as lateral G), angular velocity in the yaw direction, accelerator opening, brake pedal force, and steering angle.

[0017] The GPS receiver receives GPS signals from GPS satellites at a predetermined period, and measures the self-position of vehicle 2 based on the received GPS signals. And the GPS receiver outputs, at a predetermined period, positioning information representing the measurement result of the self-position of vehicle 2 based on the GPS signals to the vehicle behavior recording device via the in-vehicle network. Note that vehicle 2 may have a receiver compliant with a satellite positioning system other than the GPS receiver. In this case, the receiver may measure the self-position of vehicle 2.

[0018] The vehicle behavior recording device has, for example, a processor and a memory. And each time the processor of the vehicle behavior recording device acquires a behavior sensor signal from the vehicle behavior sensor, the vehicle 2 position represented by the positioning information acquired from the GPS receiver at the time closest to the acquisition time of the behavior sensor signal is associated with the behavior sensor signal. Then, the processor generates vehicle behavior information by arranging the individual behavior sensor signals associated with the position of vehicle 2 in the order of their acquisition times, and stores the generated vehicle behavior information in the memory of the vehicle behavior recording device. Therefore, the generated vehicle behavior information includes the traveling trajectory of vehicle 2 together with the behavior sensor signals arranged in the order of acquisition times. The vehicle behavior recording device may include the identification information of vehicle 2 in the vehicle behavior information.

[0019] Note that the vehicle 2 may have a camera for photographing the surrounding area of the vehicle 2. The camera may generate an image representing the surrounding area of the vehicle 2 at a predetermined cycle and output the generated image to the vehicle behavior recording device via the in-vehicle network. Further, a map representing predetermined ground features existing on or around the road, such as lane dividing lines or road signs, may be stored in the memory of the vehicle behavior device. In this case, the processor of the vehicle behavior recording device may estimate the more accurate position of the vehicle 2 by collating the image received from the camera with the map and associate the estimated position with the behavior sensor signal.

[0020] For this purpose, the processor detects a predetermined ground feature represented in the image by inputting the image to an identifier pre-trained to detect the predetermined ground feature. Then, the processor projects the detected ground feature onto the map based on the assumed position and traveling direction of the vehicle 2 with reference to parameters such as the shooting direction, focal length of the camera, and installation position with respect to the vehicle 2, and calculates the degree of coincidence between the projected ground feature and the ground feature represented in the map. The processor repeats the projection of the ground feature and the calculation of the degree of coincidence while varying the assumed position and traveling direction of the vehicle 2 variously, and may estimate the position and traveling direction of the vehicle 2 at the time when the degree of coincidence is maximized as the actual position and traveling direction of the vehicle 2 at the time of generating the image. The processor may estimate the position of the vehicle 2 at the time of acquiring each behavior sensor signal based on the position of the vehicle 2 at the time of generating each image.

[0021] Furthermore, the processor may include each image and the position and traveling direction of the vehicle 2 at the time of generating each image in the vehicle behavior information.

[0022] When a predetermined timing is reached, the vehicle behavior recording device outputs the generated vehicle behavior information to the wireless communication terminal. The predetermined timing can be, for example, the timing when the ignition switch of vehicle 2 is turned off, or the timing every time a certain period of time (for example, 30 minutes to 1 hour) has elapsed since the ignition switch of vehicle 2 was turned on. Alternatively, the collection area information representing the area to be the collection target of the vehicle behavior information may be notified in advance to vehicle 2 from server 3 via communication network 4 and radio base station 5. In this case, the vehicle behavior recording device may use the timing when vehicle 2 moves outside the area to be the collection target, with reference to the collection area information and the positioning information, as the predetermined timing.

[0023] The wireless communication terminal is a device that executes wireless communication processing conforming to a predetermined wireless communication standard. For example, by accessing radio base station 5, it is connected to server 3 via radio base station 5 and communication network 4. Then the wireless communication terminal generates an uplink wireless signal including the vehicle behavior information received from the vehicle behavior recording device. And the wireless communication terminal transmits the uplink wireless signal to radio base station 5, thereby transmitting the vehicle behavior information to server 3. Also, the wireless communication terminal receives a downlink wireless signal from radio base station 5 and passes the collection area information from server 3 included in the wireless signal to the vehicle behavior recording device.

[0024] Next, server 3, which is an example of a vehicle behavior evaluation device, will be described. Figure 2 is a hardware configuration diagram of server 3, which is an example of a vehicle behavior evaluation device. Server 3 has a communication interface 11, a storage device 12, a memory 13, and a processor 14. Communication interface 11, storage device 12, and memory 13 are connected to processor 14 via signal lines. Server 3 may further have an input device such as a keyboard and a mouse, and a display device such as a liquid crystal display.

[0025] The communication interface 11 is an example of a communication unit and has an interface circuit for connecting the server 3 to the communication network 4. The communication interface 11 is configured to be able to communicate with the vehicle 2 via the communication network 4 and the radio base station 5. That is, the communication interface 11 passes the vehicle behavior information received from the vehicle 2 via the radio base station 5 and the communication network 4 to the processor 14. Further, the communication interface 11 transmits the collected area information received from the processor 14 to the vehicle 2 via the communication network 4 and the radio base station 5.

[0026] The storage device 12 is an example of a storage unit and has, for example, a hard disk device or an optical recording medium and its access device. The storage device 12 stores various data and information used in the vehicle behavior evaluation process. For example, the storage device 12 stores the vehicle behavior information received from each vehicle 2. Further, the storage device 12 stores a map and information for specifying a predetermined road section that is the target of the vehicle behavior evaluation process. The information for specifying the predetermined road section includes, for example, a link ID for identifying the road section represented in the map, or position information of both ends of the road section. Furthermore, the storage device 12 may store a computer program for executing the vehicle behavior evaluation process that is executed on the processor 14.

[0027] The memory 13 is another example of a storage unit and has, for example, a non-volatile semiconductor memory and a volatile semiconductor memory. The memory 13 temporarily stores various data generated during the execution of the vehicle behavior evaluation process.

[0028] The processor 14 includes one or more CPUs (Central Processing Units) and their peripheral circuits. The processor 14 may further include other arithmetic circuits such as a logical arithmetic unit or a numerical arithmetic unit. And each time the processor 14 receives vehicle behavior information from any of the vehicles 2, it stores the received vehicle behavior information in the storage device 12. Further, the processor 14 executes vehicle behavior evaluation processing. Furthermore, the processor 14 generates collection area information based on information specifying a collection target area input via an input device, and distributes the generated collection area information to each vehicle 2 via the communication interface 11.

[0029] Figure 3 is a functional block diagram of the processor 14 related to vehicle behavior evaluation processing. The processor 14 includes a selection unit 21, a calculation unit 22, and an evaluation value setting unit 23. Each of these units included in the processor 14 is, for example, a functional module realized by a computer program operating on the processor 14. Alternatively, each of these units included in the processor 14 may be a dedicated arithmetic circuit provided in the processor 14.

[0030] The selection unit 21 selects, from the vehicle behavior information collected from each vehicle 2 and stored in the storage device 12, the vehicle behavior information when the vehicle 2 travels on a predetermined road section that is the target of the vehicle behavior evaluation processing. Note that the predetermined road section is specified, for example, by referring to information for specifying that road section, which is input from an input device or another device connected to the server 3 via a communication line and stored in the storage device 12.

[0031] The selection unit 21 reads information for specifying a predetermined road section from the storage device 12. Then, with reference to the information, the selection unit 21 determines, for each vehicle behavior information, whether any of the positions of vehicle 2 at the time of acquisition of the individual behavior sensor signals included in the vehicle behavior information is included in the predetermined road section. The selection unit 21 selects the vehicle behavior information in which the position of vehicle 2 at the time of acquisition of any of the behavior sensor signals is included in the predetermined road section. Further, the selection unit 21 further selects, from the selected individual vehicle behavior information, a set of the positions of the individual vehicles 2 included in the predetermined road section and the behavior sensor signals associated with those positions as the vehicle behavior information when vehicle 2 travels through the predetermined road section.

[0032] The selection unit 21 notifies the calculation unit 22 and the evaluation value setting unit 23 of the vehicle behavior information when vehicle 2 travels through the predetermined road section.

[0033] The calculation unit 22 calculates at least one of the degree of variation in the distribution of the behavior index values and the maximum value of the behavior index values based on the vehicle behavior information when vehicle 2 travels through the predetermined road section notified from the selection unit 21. As described above, the behavior index values include at least one of the acceleration and deceleration of vehicle 2, the amount of change in the acceleration and deceleration per unit time, the amount of change in the traveling direction of vehicle 2, and the amount of change in the traveling direction of vehicle 2 per unit time. The calculation unit 22 calculates, for example, the standard deviation or variance of the behavior index values as the degree of variation in the behavior index values. Also, when two or more of the elements such as the above acceleration and deceleration are included in the behavior index values, the calculation unit 22 calculates the average value or the maximum value of the standard deviation or variance for each element as the degree of variation in the behavior index values. Furthermore, the calculation unit 22 may calculate the maximum value of each element for each element.

[0034] The calculation unit 22 calculates the degree of variation of the longitudinal G or the lateral G included in the vehicle behavior information as the degree of variation of the acceleration and deceleration of the vehicle 2. Further, the calculation unit 22 calculates the maximum value of the longitudinal G or the lateral G included in the vehicle behavior information as the maximum value of the acceleration and deceleration of the vehicle 2. Alternatively, the calculation unit 22 may calculate the degree of variation of the accelerator opening or the brake pedal force included in the vehicle behavior information as the degree of variation of the acceleration and deceleration of the vehicle 2. Furthermore, the calculation unit 22 may calculate the maximum value of the accelerator opening or the brake pedal force included in the vehicle behavior information as the maximum value of the acceleration and deceleration of the vehicle 2.

[0035] In addition, for each pair of two longitudinally continuous G values included in the vehicle behavior information, the calculation unit 22 may calculate the difference between the values, or the value normalized by dividing the difference between the values by the acquisition interval of the behavior sensor signal, as the amount of change in the longitudinal G per unit time, respectively. Then, the calculation unit 22 may calculate the degree of variation and the maximum value of the amount of change in the longitudinal G per unit time as the degree of variation and the maximum value of the amount of change in the acceleration and deceleration per unit time. Similarly, for the lateral G, the accelerator opening, or the brake pedal force, for each pair of two temporally continuous values included in the vehicle behavior information, the calculation unit 22 may calculate the difference between the values, or the value normalized by dividing the difference between the values by the acquisition interval of the behavior sensor signal, as the amount of change in the acceleration and deceleration per unit time, respectively. Then, the calculation unit 22 may calculate the degree of variation and the maximum value of the amount of change in the acceleration and deceleration per unit time.

[0036] Furthermore, the calculation unit 22 may calculate the degree of variation and the maximum value of the steering angle or the angular velocity in the yaw direction included in the vehicle behavior information as the degree of variation and the maximum value of the amount of change in the traveling direction of the vehicle 2.

[0037] Furthermore, for each pair of two temporally continuous values included in the vehicle behavior information regarding the steering angle or the angular velocity in the yaw direction, the calculation unit 22 may calculate, as the amount of change per unit time in the traveling direction of the vehicle 2, the difference between the values, or the value normalized by dividing the difference between the values by the acquisition interval of the behavior sensor signal. Then, the calculation unit 22 may calculate the degree of variation and the maximum value of the amount of change per unit time in the traveling direction of the vehicle 2.

[0038] The calculation unit 22 notifies at least one of the degree of variation in the distribution of the behavior index values and the maximum value of the behavior index values calculated for the vehicle behavior information to the evaluation value setting unit 23.

[0039] The evaluation value setting unit 23 sets an evaluation value for the vehicle behavior information when the vehicle 2 notified from the selection unit 21 travels on a predetermined road section. In the present embodiment, the evaluation value setting unit 23 sets the evaluation value such that the greater the degree of variation in the distribution of the behavior index values calculated for the vehicle behavior information, or the greater the maximum value of the behavior index values, the smaller the evaluation value. At that time, the evaluation value setting unit 23 refers to a reference table showing the relationship between the calculated one of the degree of variation in the distribution of the behavior index values and the maximum value of the behavior index values and the evaluation value, thereby determining the evaluation value corresponding to the calculated degree of variation or maximum value. Note that when both the degree of variation and the maximum value are calculated, the reference table is created in advance so as to represent the relationship between both the degree of variation and the maximum value and the evaluation value. Further, when the behavior index value includes a plurality of elements and the maximum value is calculated for each element, the reference table may be created in advance so as to represent the relationship between the maximum value for each element and the evaluation value. Alternatively, the evaluation value setting unit 23 may determine the evaluation value by inputting the calculated degree of variation or maximum value into a relational expression showing the relationship between the calculated one of the degree of variation in the distribution of the behavior index values and the maximum value of the behavior index values and the evaluation value. Note that when both the degree of variation and the maximum value are calculated, the relational expression is created in advance so as to represent the relationship between both the degree of variation and the maximum value and the evaluation value. Further, when the behavior index value includes a plurality of elements and the maximum value is calculated for each element, the relational expression may be created in advance so as to represent the relationship between the maximum value for each element and the evaluation value. Such a reference table or relational expression may be stored in advance in the storage device 12.

[0040] The evaluation value setting unit 23 stores the evaluation value set for the vehicle behavior information in the storage device 12 in association with the vehicle behavior information. Alternatively, the evaluation value setting unit 23 may output the set evaluation value and the vehicle behavior information to another device via the communication interface 11. At that time, the evaluation value setting unit 23 may output, via the communication interface 11, the vehicle behavior information for which the evaluation value is equal to or greater than a predetermined threshold to another device, and may delete the vehicle behavior information for which the evaluation value is less than the predetermined threshold from the storage device 12.

[0041] Figures 4(a) and 4(b) are diagrams for explaining the relationship between the distribution of behavior index values and the evaluation values. In Figures 4(a) and 4(b), the horizontal axis represents the behavior index values, and the vertical axis represents the frequency for each behavior index value.

[0042] The distribution 400 of the behavior index values shown in Figure 4(a) is concentrated on relatively small values. As a result, the degree of variation is small, and the maximum value of the behavior index values is also relatively low. Thus, when the degree of variation in the distribution of the behavior index values is small and the maximum value of the behavior index values is small, it is assumed that the vehicle 2 did not exhibit unstable behaviors such as excessive acceleration or deceleration or suddenly changing its traveling direction. That is, it is assumed that the driver of the vehicle 2 was not driving unreasonably. Therefore, a relatively large evaluation value is set for the vehicle behavior information with such a distribution of behavior index values.

[0043] On the other hand, the distribution 410 of the behavior index values shown in Figure 4(b) is relatively dispersed from small values to large values. As a result, the degree of variation is large, and the maximum value of the behavior index values is relatively high. Thus, when the degree of variation in the distribution of the behavior index values is large or the maximum value of the behavior index values is high, it is assumed that the vehicle 2 exhibited unstable behaviors. Therefore, a relatively small evaluation value is set for the vehicle behavior information with such a distribution of behavior index values.

[0044] Figure 5 is an operation flowchart of vehicle behavior evaluation processing in the server 3. The processor 14 of the server 3 may execute the vehicle behavior evaluation processing according to the operation flowchart shown below.

[0045] The selection unit 21 of the processor 14 selects the vehicle behavior information when the vehicle 2 travels through a predetermined road section that is the target of the vehicle behavior evaluation processing from the vehicle behavior information stored in the storage device 12 (step S101).

[0046] Further, the calculation unit 22 of the processor 14 calculates at least one of the degree of variation in the distribution of the behavior index values and the maximum value of the behavior index values based on the selected vehicle behavior information (step S102).

[0047] The evaluation value setting unit 23 of the processor 14 sets an evaluation value for the vehicle behavior information such that the greater the degree of variation in the distribution of the behavior index values calculated for the selected vehicle behavior information, or the greater the maximum value of the behavior index values, the smaller the evaluation value (step S103). Then, the processor 14 ends the vehicle behavior evaluation process.

[0048] As described above, this vehicle behavior evaluation device calculates at least one of the degree of variation in the distribution of the behavior index values included in the vehicle behavior information and the maximum value of the behavior index values. And this vehicle behavior evaluation device sets a smaller evaluation value for the vehicle behavior information as the degree of variation in the distribution of the behavior index values is greater or the maximum value of the behavior index values is greater. Therefore, this vehicle behavior evaluation device can set an evaluation value that appropriately evaluates the behavior of the vehicle for the vehicle behavior information.

[0049] Note that the processor 14 may learn a control model used in the automatic driving control of the vehicle as described above based on the individual vehicle behavior information for which the evaluation value has been set. That is, the processor 14 may use only the vehicle behavior information for which the evaluation value is equal to or greater than a certain value as teacher data and learn the control model according to a predetermined supervised learning algorithm such as the error backpropagation method. Then, the processor 14 may distribute the learned control model to each vehicle 2 or other devices via the communication interface 11.

[0050] A computer program for causing a computer to realize the functions of each part of the processor of the vehicle behavior evaluation device according to each of the above embodiments or modifications may be provided in a form stored in a computer-readable recording medium. Note that the computer-readable recording medium can be, for example, a magnetic recording medium, an optical recording medium, or a semiconductor memory.

[0051] As described above, those skilled in the art can make various modifications according to the implemented forms within the scope of the present invention.

Explanation of Signs

[0052] 1 Vehicle behavior evaluation system 2 Vehicle 3 Server 11 Communication interface 12 Storage device 13 Memory 14 Processor 21 Selection unit 22 Calculation unit 23 Evaluation value setting unit 4 Communication network 5 Radio base station

Claims

1. A calculation unit that represents the behavior of a vehicle when the vehicle travels on a predetermined road section, and calculates at least one of the variation degree of the distribution of behavior index values including the acceleration and deceleration of the vehicle, the change amount of the acceleration and deceleration of the vehicle per unit time, the change amount of the traveling direction of the vehicle, and the change amount of the traveling direction of the vehicle per unit time, based on vehicle behavior information including an image representing the surroundings of the vehicle generated by a camera mounted on the vehicle; An evaluation value setting unit that sets a smaller evaluation value for the vehicle behavior information as the calculated value of the variation degree of the distribution of the behavior index values and the maximum value of the behavior index values is larger; A learning unit that learns a control model that outputs control information for controlling the traveling of the vehicle when the image is input, by using the vehicle behavior information for which the evaluation value is equal to or greater than a predetermined value as teacher data; A vehicle behavior evaluation device having the above.

2. When a computer represents the behavior of a vehicle when the vehicle travels on a predetermined road section, and calculates at least one of the variation degree of the distribution of behavior index values including the acceleration and deceleration of the vehicle, the change amount of the acceleration and deceleration of the vehicle per unit time, the change amount of the traveling direction of the vehicle, and the change amount of the traveling direction of the vehicle per unit time, based on vehicle behavior information including an image representing the surroundings of the vehicle generated by a camera mounted on the vehicle; The computer sets a smaller evaluation value for the vehicle behavior information as the calculated value of the variation degree of the distribution of the behavior index values and the maximum value of the behavior index values is larger; The computer learns a control model that outputs control information for controlling the traveling of the vehicle when the image is input, by using the vehicle behavior information for which the evaluation value is equal to or greater than a predetermined value as teacher data. A vehicle behavior evaluation method including the above.

3. Represents the behavior of a vehicle when it travels on a specified road section, and based on vehicle behavior information including an image representing the surroundings of the vehicle generated by a camera mounted on the vehicle, at least one of the acceleration and deceleration of the vehicle, the change amount of the acceleration and deceleration of the vehicle per unit time, the change amount of the traveling direction of the vehicle, and the change amount of the change amount of the traveling direction of the vehicle per unit time is calculated, and the degree of variation of the distribution of the behavior index value including at least one of the above and the maximum value of the behavior index value are calculated. The larger the calculated value of either the degree of variation of the distribution of the behavior index value or the maximum value of the behavior index value, the smaller the evaluation value is set for the vehicle behavior information. By using the vehicle behavior information for which the evaluation value is equal to or greater than a predetermined value as teacher data, a control model that outputs control information for controlling the traveling of the vehicle when the image is input is learned. A computer program for vehicle behavior evaluation for causing a computer to execute the above.

Images