Vehicle control system, vehicle control method, and vehicle control program

Abstract

This vehicle control system addresses discomfort caused by individual drivers' perceptions of driving and provides support for creating an optimal driving environment. [Solution] The vehicle control system 1 includes an abnormality input unit 20 that inputs abnormality information emitted by the driver of the vehicle, a scene identification unit 61 that identifies abnormality driving scenes related to the abnormality, a data extraction unit 62, an analysis unit 63, a suitability value calculation unit 64, and an output unit 65. The data extraction unit 62 extracts predetermined driving data related to abnormality driving scenes from among multiple driving data of the vehicle obtained from the driving data input unit 30. The analysis unit 63 acquires driving data by the driver of the vehicle from the feature recording unit 40 as individual feature quantities, compares these feature quantities with the extracted driving data, and identifies parameters that should be corrected. The suitability value calculation unit 64 calculates suitability values ​​to fit the above parameters to the driver's feature quantities. The output unit 65 outputs a control signal according to the suitability value.

Description

Technical Field

[0001] The present disclosure relates to a vehicle control system, a vehicle control method, and a vehicle control program.

Background Art

[0002] Conventionally, there has been proposed an information processing apparatus that evaluates and analyzes the driving behavior of an automobile using a learning model generated in advance by a machine learning algorithm based on various types of information such as various pieces of driving information including position information and acceleration of the automobile (see, for example, Patent Document 1). The information processing apparatus described in Patent Document 1 acquires the aforementioned various types of information with a mobile terminal installed in the automobile, applies the learning model to calculate the driving risk of the driver of the automobile and a driving diagnosis result as scores, and accumulates score information in a database. This information processing apparatus can perform processing such as issuing a warning when the vehicle approaches a road where dangerous driving has occurred, using the calculated scores.

Prior Art Literature

Patent Literature

[0003]

Patent Document 1

Summary of the Invention

Problem to be Solved by the Invention

[0004] In recent years, as in the information processing apparatus described above, vehicle control systems that evaluate the driving behavior of an automobile driver using a learning model generated in advance based on various types of information of the automobile and execute vehicle control of the automobile based on the evaluation result have been studied. In the field of this type of vehicle control system, utilization of information such as comments made by the driver to provide an optimal driving environment specialized for the individual driver has also been studied, but this has not yet been sufficiently realized.

[0005] Conventional vehicle control systems acquire sensor information from on-board sensors into an ECU (Electronic Control Unit) and use it to control the vehicle's behavior. ECU is an abbreviation for Electronic Control Unit. However, in this case, it is not possible to respond to subtle abnormalities that cannot be detected by the ECU or on-board sensors, or to discomfort caused by individual driving sensations.

[0006] In view of the above, this disclosure aims to provide a vehicle control system that can respond to discomfort caused by the driver's individual driving sensations and abnormalities that cannot be detected by on-board sensors, based on information about discomfort reported by the driver, and to support the provision of an optimal driving environment. It also aims to provide a vehicle control method and a vehicle control program that can provide the aforementioned support. [Means for solving the problem]

[0007] According to one aspect of this disclosure, the vehicle control system is An discomfort input unit (20-23) inputs discomfort information, which is information about discomfort emitted by the driver of the vehicle, A scene identification unit (61) identifies the driving scene of the vehicle related to the anomaly based on the information about the anomaly, A data extraction unit (62) extracts predetermined driving data related to a driving scene identified by a scene identification unit from among multiple driving data of the vehicle input from a driving data input unit (30), An analysis unit (63) compares the feature quantities stored in the feature recording unit (40), which records driving data from the driver's past driving as individual feature quantities, with predetermined driving data extracted by the data extraction unit, and identifies parameters that need to be corrected. The analysis unit calculates fitted values ​​(64) to fit the parameters identified by the analysis unit to the features, The system includes an output unit (65) that outputs a control signal corresponding to the conformance value calculated by the conformance value calculation unit to an in-vehicle device whose parameters can be changed, thereby correcting the parameters.

[0008] This results in a vehicle control system that, based on information about discomfort reported by the driver, can respond to discomfort stemming from the driver's individual driving sensations and abnormalities that cannot be detected by on-board sensors, thereby supporting the provision of an optimal driving environment.

[0009] In another aspect of this disclosure, a vehicle control method performed by a vehicle control unit (60), This involves inputting information about unusual sensations reported by the driver of the vehicle, and Based on the information about the anomaly, identify the anomaly driving scene, which is a driving scene of the vehicle related to the anomaly. The system extracts predetermined driving data related to unusual driving scenes from among multiple driving data of the vehicle input from the driving data input unit (30), The feature recording unit (40), which records driving data from the driver's past driving experiences as individual feature quantities, compares the feature quantities stored in the unit with the extracted predetermined driving data to identify the parameters that need to be corrected. This involves calculating fitted values ​​to match the identified parameters to the features, This includes outputting a control signal corresponding to the calculated suitability value to an in-vehicle device whose parameters can be changed, thereby correcting the parameters.

[0010] This results in a vehicle control method that, based on information about discomfort reported by the driver, can respond to discomfort stemming from the driver's individual driving sensations and abnormalities that cannot be detected by on-board sensors, thereby supporting the provision of an optimal driving environment.

[0011] In another aspect of this disclosure, a vehicle control program executed by the vehicle control unit (60) is, The process involves inputting information about discomfort, which is information about discomfort reported by the driver of the vehicle, Based on the information about the anomaly, a process is performed to identify the anomaly driving scene, which is a driving scene of the vehicle related to the anomaly. A process to extract predetermined driving data related to an unusual driving scene from among multiple driving data of the vehicle input from the driving data input unit (30), a process of comparing a feature amount accumulated in a feature recording unit (40) that records travel data from a driver's past driving as an individual's feature amount with extracted predetermined travel data, and identifying a parameter to be corrected; a process of calculating an adaptation value for adapting the identified parameter to the feature amount; a process of outputting a control signal corresponding to the calculated adaptation value to an in-vehicle device capable of changing a parameter, and causing the parameter to be corrected.

[0012] Accordingly, the vehicle control program can provide support for providing an optimal driving environment by responding to discomfort caused by the individual driver's driving sensation and abnormalities that cannot be detected by in-vehicle sensors, based on discomfort information issued by the driver.

[0013] Note that reference numerals in parentheses attached to each constituent element and the like indicate an example of the correspondence between the constituent element and the like and specific constituent elements and the like described in the embodiments described later. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] [Figure 1] It is a block diagram showing the configuration example of the vehicle control system of a 1st embodiment. [Figure 2] It is explanatory drawing about travel data. [Figure 3] It is explanatory drawing about the structural example which acquires a feature-value from an external recording medium. [Figure 4] It is a figure which shows an example of a keyword dictionary. [Figure 5] It is explanatory drawing about identification of the driving scene by keyword matching using a learned learning model. [Figure 6] It is a flowchart which shows the processing example of discomfort reduction by a vehicle control system. MODE FOR CARRYING OUT THE INVENTION

[0015] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. In each of the following embodiments, portions that are identical or equivalent to each other will be described with the same reference numerals. Furthermore, in cases where only a part of the constituent elements is described in an embodiment, the constituent elements described in preceding embodiments can be applied to other parts of the constituent elements. The following embodiments can be partially combined with each other, even if not explicitly stated, as long as the combination does not cause any problems.

[0016] (First Embodiment) The vehicle control system 1 according to the first embodiment will be described. The vehicle control system 1 of the present embodiment is applied to a vehicle such as a passenger car, and is configured to execute vehicle control for reducing discomfort when discomfort input from a driver of the vehicle is received. Hereinafter, for convenience of description, the vehicle to which the vehicle control system 1 of the present embodiment is applied will be referred to as "the host vehicle".

[0017] [Basic Configuration] As shown in FIG. 1, for example, the vehicle control system 1 includes an individual identification unit 10, a discomfort input unit 20, a travel data input unit 30, a feature recording unit 40, a keyword dictionary 50, and a vehicle control unit 60. The vehicle control system 1 is configured such that, for example, when information on discomfort expressed by the driver of the host vehicle (hereinafter simply referred to as "the driver") is input via the discomfort input unit 20, the vehicle control unit 60 outputs a control signal to an actuator 70 or the like of the host vehicle.

[0018] The individual identification unit 10 identifies the driver and transmits driver information to the vehicle control unit 60. The individual identification unit 10 acquires driver information by any arbitrary method, such as communicating with a mobile terminal carried by the driver, or capturing an image of the driver and analyzing the driver's image using a known image authentication technique, and transmits the driver information to the vehicle control unit 60.

[0019] The discomfort input unit 20 inputs discomfort information, i.e., discomfort information, emitted by the driver to the vehicle control unit 60. Examples of discomfort information include information on the driver's voice, movements, facial expressions, gaze, pulse, brain waves, and other physical reactions, as well as information on text input by the driver. The discomfort input unit 20 inputs, for example, at least one of the above-mentioned physical reactions as discomfort information. The discomfort input unit 20 is composed of, for example, an input device 21, a driver imaging device 22, a measurement device 23, etc., but it may be composed of only some of these, or it may be configured to have other in-vehicle devices. For example, predetermined voices, movements, facial expressions, gaze, pulse, brain waves, or changes thereof that are expected to occur when a person feels discomfort are set in advance, and the discomfort input unit 20 is configured to input discomfort information that matches these settings.

[0020] The input device 21 is, for example, a microphone or touch panel installed inside the vehicle, and inputs the driver's voice or text input results from the driver's operations as abnormality information. The driver imaging device 22 has, for example, an in-vehicle camera that images a predetermined area including the driver's face, and analyzes the image obtained by imaging the driver using known image recognition technology to acquire information on the driver's movements, facial expressions and gaze, and inputs it as abnormality information. As the driver imaging device 22, for example, a Driver Status Monitor (registered trademark) manufactured by Denso Corporation can be used, but is not limited to this. The measurement device 23 is, for example, a pulse sensor or a headset capable of measuring brain waves, and inputs biological information such as the driver's pulse and brain waves obtained by measurement as abnormality information.

[0021] The driving data input unit 30 inputs various driving data of the vehicle to the vehicle control unit 60. Examples of driving data include, for example, as shown in Figure 2, various physical quantities generated in the vehicle, such as acceleration, vehicle speed, angular velocity, and steering angle, as well as data on the amount of operation of various on-board devices by the driver, such as accelerator opening and brake opening. In the driving data shown in Figure 2, the horizontal axis represents time, and the vertical axis (not shown) represents the physical quantities or operation quantities of the vehicle that change in the driving scene. The driving data input unit 30 is composed of, for example, an on-board sensor 31 and a pedal device 32, and inputs various signals to the vehicle control unit 60, but is not limited to these, and may have other on-board devices such as a steering wheel. Examples of on-board sensors 31 include, but are not limited to, an acceleration sensor, a wheel speed sensor, an angular velocity sensor, and a steering angle sensor. The pedal device 32 has an accelerator pedal and a brake pedal, and outputs information such as accelerator opening and brake opening to the vehicle control unit 60. In the above description, a typical example was given of directly inputting driving data from the on-board sensor 31 and pedal device 32 to the vehicle control unit 60. However, the driving data input unit 30 may be configured to input driving data via an ECU that can communicate with the on-board sensor 31 and pedal device 32. In other words, the driving data input unit 30 inputs information that can capture various operations of the individual driver. The driving data input by the driving data input unit 30 is recorded, for example, on a recording medium (not shown) or a feature recording unit 40 of the vehicle control unit 60, but it may also be recorded on a recording medium outside the vehicle.

[0022] The feature recording unit 40 is a recording medium that records driving data from past driving by the driver as feature quantities. Examples of feature quantities recorded by the feature recording unit 40 include various parameters such as various physical quantities and various operation quantities that occur in the vehicle during various driving scenes such as acceleration, deceleration, stopping, turning, and reversing. The feature recording unit 40, for example, accumulates data on the amount of change over time for each of the above-mentioned parameters for each driving scene. The feature recording unit 40 also records and maintains feature quantities for each individual driver, for example, if there are multiple drivers. Since the feature quantities recorded by the feature recording unit 40 are composed of daily driving data generated by the driver's driving, they can be said to represent the distribution of the driver's individual driving preferences. Furthermore, the feature quantities recorded in the feature recording unit 40 are determined using machine learning models and statistical analysis models, similar to the scene identification unit 61 described later, to accumulate parameters with a contribution of a predetermined level or higher for each driving scene. A parameter with a contribution of a predetermined level or higher means, for example, one of several parameters whose amount of change in a particular driving scene is relatively higher than that of other parameters. Specifically, parameters whose contribution exceeds a certain threshold include, for example, "acceleration" and "accelerator opening" when the driving scene is "acceleration," and these are appropriately selected by the learning model according to the driving scene.

[0023] The feature recording unit 40 is, for example, a recording medium such as a non-volatile rewritable memory installed in the vehicle. The feature recording unit 40 may, for example, directly acquire feature quantities from the driving data input unit 30, or, as shown in Figure 3, may acquire and update feature quantities by communicating with an external recording medium 100 that records feature quantities output from the driving data input unit 30 and downloading them. The external recording medium 100 is a recording medium that is not directly installed in the vehicle, and examples include USB memory, mobile terminals such as smartphones and wearable devices, and cloud computing. USB is an abbreviation for Universal Serial Bus. When using the external recording medium 100, the driver of the vehicle can transfer and use the driver's feature quantities for support control to reduce discomfort when the driver drives another vehicle.

[0024] The keyword dictionary 50, as shown in Figure 4, for example, is a collection of keywords related to the driving and time of the vehicle, where synonymous or similar keywords are categorized and registered. The keyword dictionary 50 is used, for example, when the scene identification unit 61, described later, identifies a driving scene in which the driver felt discomfort, based on discomfort information from the discomfort input unit 20. For example, if the discomfort information input from the discomfort input unit 20 includes the word "acceleration," the keyword dictionary 50 is used for keyword matching for the word "acceleration." Keyword matching is performed, for example, on keywords included in the discomfort information, using conditions such as exact match, partial match, or similarity search. The keyword dictionary 50 is generated by any method, such as creation by the user, automatic generation using LLM, based on place names and intersection names included in map data acquired by the vehicle's in-vehicle equipment, or words obtained from character recognition in images captured by the in-vehicle camera. LLM is an abbreviation for Large language Models, which are large-scale language models constructed using deep learning technology.

[0025] The vehicle control unit 60 has a microcomputer configuration that includes a CPU, RAM, ROM, and non-volatile rewritable memory (not shown). The vehicle control unit 60 reads and executes a computer program stored in the ROM or non-volatile rewritable memory, which are non-transitional physical recording media. When this computer program is executed, the method corresponding to the computer program is executed. That is, the vehicle control unit 60 executes various control processes, such as the control process shown in Figure 6, which will be described later, according to the computer program. Note that CPU, RAM, and ROM are abbreviations for Central Processing Unit, Random Access Memory, and Read Only Memory, respectively. The vehicle control unit 60 includes, for example, a scene identification unit 61, a data extraction unit 62, an analysis unit 63, a suitable value calculation unit 64, and an output unit 65.

[0026] The scene identification unit 61 identifies driving scenes in which the driver felt something was wrong, i.e., driving scenes that felt wrong. For example, the scene identification unit 61 uses the information about the feeling of unease input from the feeling of unease input unit 20 and the keyword dictionary 50 to identify driving scenes that felt wrong. For example, if the information about the feeling of unease is a phrase entered as a voice or text, the scene identification unit 61 extracts the keywords contained in the information about the feeling of unease and identifies the driving scenes that felt wrong through keyword matching using the keyword dictionary 50. Specifically, for example, if the driver comments, "The acceleration today isn't great," the scene identification unit 61 extracts the keywords "today" and "acceleration" and uses keyword matching to identify the driving scene as acceleration and the time as today. For example, if the information about the feeling of unease is non-verbal information such as the driver's actions, facial expressions, gaze, pulse, or brain waves, the scene identification unit 61 does not use the keyword dictionary 50 and identifies the driving scene at the time the information about the feeling of unease was entered as the driving scene that felt wrong. The information about the driving scenes that felt wrong identified by the scene identification unit 61 is output to the data extraction unit 62, for example, and used in the process of identifying the parameters that caused the feeling of unease.

[0027] The scene identification unit 61 may use, for example, a learning model or statistical analysis model generated by training it using algorithms such as machine learning or deep learning, based on information about the driver's physical reactions and driving scene information as training data. Examples of learning models include well-known machine learning methods such as neural networks and support vector machines. Examples of statistical analysis models include well-known statistical methods such as hierarchical Bayesian models and Gaussian process regression models. Such trained learning models or pre-designed statistical analysis models may be stored, for example, on a recording medium (not shown) that constitutes the vehicle control unit 60, or they may be stored on a recording medium outside the vehicle and accessed and executed by an in-vehicle communication device. Furthermore, the learning models and statistical analysis models are stored in the vehicle control unit 60 and are updated as needed while in use or by being downloaded from a recording medium outside the vehicle.

[0028] The scene identification unit 61 synchronizes a driving scene with comments made during that driving scene over time, as shown in Figure 5, for example, and learns comments that drivers are likely to make for each driving scene using a machine learning algorithm. In the driving scene shown in Figure 5, the horizontal axis represents time, and the vertical axis (not shown) represents the physical quantity or operation quantity of the vehicle that changes during that driving scene. For example, when the driving scene of the vehicle is "reversing," and the driver makes comments such as "stop slowly," "park," or "reverse," the scene identification unit 61 learns the combination of driving data and comments for the reversing driving scene. The scene identification unit 61 performs similar learning for various other driving scenes, such as acceleration, curves, stopping, or deceleration, and repeatedly learns comments that drivers are likely to make in various driving scenes. As a result, for example, if the driver makes the comment "I need to park calmly," the scene identification unit 61 can identify the driving scene as "reversing" based on the learning results. In other words, the scene identification unit 61 is a pre-learned model optimized to match the driver's characteristics, and its accuracy in identifying unusual driving scenes improves with daily use.

[0029] Based on the information of the unusual driving scene identified by the scene identification unit 61, the data extraction unit 62 extracts driving data related to the driver's discomfort from among multiple driving data input from the driving data input unit 30. For example, if the identified unusual driving scene is acceleration, the data extraction unit 62 extracts predetermined driving data related to the acceleration scene, such as acceleration, vehicle speed, and accelerator pedal opening, at the time of the unusual driving scene. For the sake of explanation, the data extracted by the data extraction unit 62 from among multiple driving data will be referred to as "extracted driving data." For example, the data extraction unit 62 outputs the extracted driving data to the analysis unit 63.

[0030] The analysis unit 63 compares the extracted driving data with the driver's past driving data recorded in the feature recording unit 40, i.e., the individual's characteristic quantities, and analyzes it to identify parameters that should be corrected or parameters that may be abnormal. For example, the analysis unit 63 calculates the difference between the extracted driving data and the driver's past driving data in the same driving scene as the extracted driving data, and determines whether there are any parameters that should be corrected or parameters that may be abnormal. For example, the analysis unit 63 makes the above determination based on whether the calculated difference amount is above a predetermined threshold. This makes it possible to identify the cause of any discomfort when a driver feels something is wrong while driving, even if it cannot be detected by on-board sensors or the ECU, by taking into account the driver's individual characteristic quantities.

[0031] The analysis unit 63 may identify predetermined parameters (for example, physical quantities such as acceleration or manipulated quantities such as accelerator opening) as described above, or it may identify in-vehicle equipment (for example, in-vehicle sensors or brake systems) that detect or control such parameters. For the sake of explanation, the parameters identified by the analysis unit 63 through analysis processing will be referred to as "identified parameters" below.

[0032] The analysis unit 63 identifies variables that can be changed based on the vehicle's specification information stored in a recording field (not shown) of the vehicle control unit 60, and identifies the vehicle's actuators 70 that can be used to change these variables. Here, variables refer to, for example, those related to driving data, such as specific parameters or signal outputs from on-board equipment necessary to change those specific parameters. The actuators 70 are, for example, drive mechanisms of various on-board equipment used to change physical quantities such as acceleration or manipulated quantities such as accelerator opening. The analysis results from the analysis unit 63 are output to, for example, the conformance value calculation unit 64 and used in the conformance value calculation process described later.

[0033] The fitting value calculation unit 64 calculates the necessary fitting value such that the difference between a specific parameter and a target value is less than or equal to a predetermined value, using, for example, the driver's individual characteristic quantities recorded in the feature recording unit 40 as the target value. In other words, the fitting value is a value used to adapt a specific parameter, which is considered to be the cause of the driver's discomfort, to the driver's past characteristic quantities. The fitting value calculation unit 64 calculates the amount of drive required to adapt the specific parameter to the target value when using, for example, the actuator 70 identified by the analysis unit 63, as the fitting value. Furthermore, if there are multiple variables that are effective in correcting the specific parameter, the fitting value calculation unit 64 calculates a fitting value such that the number of variables is minimized, or the impact on other in-vehicle systems is minimized, i.e., the cost of the change is minimized. The result of the fitting value calculated by the fitting value calculation unit 64 is input to, for example, the output unit 65.

[0034] The output unit 65 outputs a drive signal to the actuator 70 according to the appropriate value, for example. This ensures that the difference between a specific parameter and a target value is below a predetermined level, allowing for vehicle control tailored to the driver's individual preferences. The output unit 65 may also output a drive signal to an ECU or the like, and issue a drive command to the actuator 70 via the ECU or the like.

[0035] Furthermore, assuming that the output unit 65 outputs a drive signal corresponding to the appropriate value to the actuator 70, etc., it will stop outputting the drive instruction if it exceeds the range in which the safe driving of the vehicle can be maintained. Specifically, for various changeable parameters of the vehicle (e.g., vehicle speed, accelerator opening), a range in which the safe driving of the vehicle cannot be maintained, i.e., a dangerous range, is set in advance, and this setting data is stored in a recording medium (not shown) of the vehicle control unit 60. Then, for example, the output unit 65 calculates the value of the parameter expected by the output of the drive signal corresponding to the appropriate value, i.e., the value of the parameter after change control, and if that value is in the dangerous range, it executes control to stop outputting the drive signal. This makes it possible to maintain the safe driving of the vehicle.

[0036] The above describes the basic configuration of the vehicle control system 1. Note that the vehicle control system 1 is not limited to the configuration example shown in Figure 1, and may include other components. Furthermore, the feature recording unit 40 and keyword dictionary 50 may be stored in the vehicle control unit 60, and these configurations can be modified as appropriate. Also, while an example has been described in which the output unit 65 stops outputting the conformance value as needed, this is not the only example. The conformance value calculation unit 64 may stop outputting the calculation result to the output unit 65 according to the conformance value, thereby canceling the drive command from the output unit 65. Additionally, the vehicle control system 1 may have, for example, a safety range that can maintain the safe driving of the vehicle, pre-set instead of a danger range, and the output of the drive command may be canceled if the parameters after reflecting the conformance value are not within the safety range. Thus, the blocks that execute various processes in the vehicle control unit 60 of the vehicle control system 1 may be modified as appropriate, and the setting data and judgment processes may be modified as appropriate to obtain similar results.

[0037] [Vehicle control to reduce discomfort] Next, an example of the process used by the vehicle control system 1 to reduce discomfort will be described. When the vehicle control system 1 meets predetermined start conditions, such as the vehicle's ignition being turned on, it executes the control flow shown in Figure 6.

[0038] In step S110, for example, the vehicle control unit 60 determines whether or not abnormality information has been input based on the presence or absence of an input signal from the abnormality input unit 20. For example, if the vehicle control unit 60 determines that the result is positive in step S110, it proceeds to step S120, while if the result is negative, it returns to step S110 and repeats the determination process of step S110 at predetermined intervals.

[0039] In step S120, for example, the scene identification unit 61 identifies a driving scene in which the driver may have felt discomfort, using the keyword dictionary 50 as needed, based on the discomfort information from the discomfort input unit 20, in the manner described above. After that, the scene identification unit 61 outputs information of the identified discomfort driving scene to the data extraction unit 62. Steps S110 to S120 correspond to inputting (or processing) discomfort information emitted by the driver of the vehicle, and identifying (or processing) a discomfort driving scene of the vehicle related to the discomfort based on the discomfort information. After step S120, the vehicle control unit 60 proceeds to step S130.

[0040] In step S130, for example, the data extraction unit 62 extracts driving data that may be attributable to the unusual driving scene from among the multiple driving data input from the driving data input unit 30, based on the information of the unusual driving scene. Then, the data extraction unit 62 outputs the information of the extracted driving data to the analysis unit 63. Step S130 corresponds to extracting (or processing) predetermined driving data related to the unusual driving scene from among the multiple driving data of the vehicle input from the driving data input unit 30. After step S130, the vehicle control unit 60 proceeds to step S140.

[0041] In step S140, for example, the analysis unit 63 acquires information on the driver's feature quantities from the feature recording unit 40. Then, for example, the analysis unit 63 calculates the difference between the extracted driving data and the past driving data corresponding to the extracted driving data from the acquired feature quantities, and identifies parameters that should be corrected or parameters that may be abnormal. The analysis unit 63 also identifies variables that can be changed from the specifications of the vehicle and actuators 70 that can be used to change those variables. After that, for example, the analysis unit 63 outputs the information of the identified parameters, variables and actuators 70 to the fitting value calculation unit 64. Step S140 corresponds to comparing the feature quantities accumulated in the feature recording unit 40, which records driving data from the driver's past driving as individual feature quantities, with the extracted predetermined driving data and identifying (or processing) parameters that should be corrected. After step S140, the vehicle control unit 60 proceeds to step S150.

[0042] In step S150, for example, the fitting value calculation unit 64 calculates the necessary fitting values ​​based on the information obtained from the analysis unit 63, using the individual driver feature quantities obtained from the feature recording unit 40 as target values, so that the difference between the specific parameter and the target value is less than or equal to a predetermined value. Also, for example, if there are multiple variables that are effective in correcting the specific parameter, the fitting value calculation unit 64 corrects the fitting values ​​so that the cost of modification in correcting the specific parameter is minimized. Step S150 corresponds to calculating (or processing) fitting values ​​to fit the identified parameter to the feature quantities. After step S150, the vehicle control unit 60 proceeds to step S160.

[0043] In step S160, for example, the vehicle control unit 60 calculates the parameter values ​​assuming that a drive signal based on the conforming value calculated in step S150 is output, and determines whether the calculated values ​​are within a preset danger range. If the vehicle control unit 60 determines that the result is positive in step S160, it returns to step S110 and stops outputting the drive signal. If the vehicle control unit 60 determines that the result is negative in step S160, it proceeds to step S170.

[0044] In step S170, for example, the output unit 65 outputs a drive signal to the actuator 70 or ECU corresponding to the fitted value calculated in step S150 or the corrected fitted value. As a result, the actuator 70 identified by the analysis unit 63 is activated, and the specific parameter is corrected to a value that matches the individual driver's characteristics, thereby reducing the driver's discomfort. Step S170 corresponds to outputting a control signal corresponding to the calculated fitted value to an in-vehicle device whose parameters can be changed, causing the parameter to be corrected (or processed).

[0045] Then, the vehicle control unit 60, for example, after step S170, returns the process to step S110 and repeats the above-described series of processes until a predetermined termination condition is met, such as the ignition being turned off.

[0046] The above describes the basic content of vehicle control by the vehicle control system 1 to reduce driver discomfort. Note that the vehicle control for reducing discomfort is not limited to the above processing example, and the order of processing may be changed as appropriate within the scope of what is possible. Furthermore, for example, the vehicle control unit 60 may determine whether the difference amount calculated in step S140 is greater than or equal to a predetermined threshold. If it is greater than or equal to the threshold, it may execute the processing from step S150 onwards. If it is less than the threshold, it may execute a process to notify the driver by voice or other means. Thus, for example, when driver discomfort information is input, the vehicle control system 1 may, instead of correcting specific parameters, notify the driver of maintenance or other issues.

[0047] As described above, the vehicle control system 1 identifies parameters that need to be modified based on the driver's individual characteristics, based on the discomfort information emitted by the driver, and performs control to automatically adapt the identified parameters to the characteristics. Therefore, the vehicle control system 1 is able to reduce discomfort and optimize the system to match the individual's different sensations, such as changes in the feeling of driving the vehicle due to tire wear or the optimization of the automatic shift change function.

[0048] According to this embodiment, the vehicle control system 1 comprises an abnormality input unit 20 for inputting abnormality information, a scene identification unit 61 for identifying abnormal driving scenes based on the abnormality information, and a data extraction unit 62 for identifying data from driving data that is caused by the abnormality. The vehicle control system 1 also includes an analysis unit 63 that compares the feature quantities in various driving scenes of the driver held by the feature recording unit 40 with the extracted driving data and identifies parameters that need to be corrected. Furthermore, the vehicle control system 1 includes a matching value calculation unit 64 that calculates matching values ​​to adapt the specific parameters to the driver's feature quantities, and an output unit 65 that outputs signals corresponding to the matching values ​​to actuators 70, etc. As a result, when any abnormality information is emitted from the driver, the parts that need to be corrected can be immediately and accurately identified for each individual driver, and it becomes possible to support the early detection of minor abnormalities that cannot be detected by on-board sensors, ECUs, etc.

[0049] Furthermore, the vehicle control system 1 can identify subtle abnormalities that the ECU and other systems cannot detect, based on anomaly information extracted from ambiguous information such as the driver's comments and emotions, and can prompt the driver to perform preventative maintenance, thereby preventing malfunctions that could lead to recalls. In addition, the vehicle control system 1 has the function to identify abnormalities and anomalies that occurred over time, not limited to dangerous driving, thereby improving the long-term reliability of the vehicle. Moreover, the vehicle control system 1 of this embodiment has the following features.

[0050] (1) The discomfort input unit 20 inputs information based on the driver's physical reactions. The information based on physical reactions includes at least one of the following: the driver's brain waves, pulse, gaze, facial expressions, movements, and voice.

[0051] (2) The information input from the driving data input unit 30 is information related to the driver's operation of the vehicle.

[0052] (3) The discomfort input unit 20 inputs the driver's voice as discomfort information. The scene identification unit 61 identifies the vehicle's driving scene related to the discomfort by keyword matching using the keyword dictionary 50. As a result, the vehicle control system 1 can identify the discomfort driving scene with simple processing and improve the processing speed of the vehicle control to reduce discomfort.

[0053] (4) The scene identification unit 61 extracts keywords contained in the voice and identifies the driving scenes associated with the extracted keywords in the keyword dictionary 50 as driving scenes of the vehicle that are related to the sense of incongruity.

[0054] (5) The scene identification unit 61 identifies the driving scenes of the vehicle that are associated with the feeling of discomfort using a learning model obtained by a learning process that uses learning data including the driver's voice and driving data related to the voice. As a result, the vehicle control system 1 can identify driving scenes that are associated with discomfort while matching a wide variety of driving scenes of the vehicle with the individual driver's image, thereby improving the accuracy of vehicle control for reducing discomfort.

[0055] (6) The feature recording unit 40 accumulates driving data from the driver on a daily basis and records it as a feature of the driver. In recording the feature, a learning model obtained by a learning process using learning data that includes the amount of operation of in-vehicle devices such as the pedal device 32 operated by the driver and driving data output from the in-vehicle sensor 31 is used. The feature recording unit 40 then records as a feature the driving data identified by the learning model as having a contribution of a predetermined level or higher from the driving data from the driver on a daily basis. As a result, the vehicle control system 1 can personalize the target values, i.e., the ideal, for various driving scenes to match the preferences of the driver of its vehicle, and can perform processing to reduce discomfort that is tailored to the individual's unique habits.

[0056] (7) The feature recording unit 40 downloads the feature quantities from an external recording medium 100 located outside the vehicle, which stores and records the driver's daily driving data as the driver's feature quantities. As a result, even if the vehicle being driven is changed, the vehicle control system 1 can carry over the driver's individual target value data and perform processing to reduce discomfort.

[0057] (8) The analysis unit 63 identifies variables that can be changed during the operation of the vehicle and identifies the vehicle's actuators 70 that can be used to adjust the variables. The suitability value calculation unit 64 calculates the suitability value when using the actuators 70. As a result, the vehicle control system 1 can identify the parameters of the vehicle that should be changed, i.e., the on-board systems.

[0058] (9) The output unit 65 stops outputting the control signal when the parameter to be changed, assuming that a control signal corresponding to the conforming value is output, falls into a preset danger range. This allows the vehicle control system 1 to prioritize driving safety over reducing discomfort when the driver feels any discomfort, thereby ensuring the safety of its own vehicle.

[0059] (10) When there are multiple variables related to parameter modification, the fitting value calculation unit 64 calculates the fitting value that minimizes the number of variables. This simplifies the process of modifying specific parameters in the vehicle control system 1.

[0060] (Other embodiments) This disclosure is described in accordance with the embodiments, but it is understood that this disclosure is not limited to such embodiments or structures. This disclosure also includes various modifications and variations within the scope of the invention. In addition, various combinations and forms, as well as other combinations and forms including one, more, or less of those elements, fall within the scope and concept of this disclosure.

[0061] The vehicle control unit 60 and its method described in this disclosure may be implemented by a dedicated computer provided by configuring a processor and memory programmed to perform one or more functions embodied by a computer program. Alternatively, the vehicle control unit 60 and its method described in this disclosure may be implemented by a dedicated computer provided by configuring a processor by one or more dedicated hardware logic circuits. Alternatively, the vehicle control unit 60 and its method described in this disclosure may be implemented by one or more dedicated computers configured by a combination of a processor and memory programmed to perform one or more functions and a processor configured by one or more hardware logic circuits. Furthermore, the computer program may be stored as instructions executed by the computer on a computer-readable non-transitional tangible recording medium.

[0062] It goes without saying that, in each of the above embodiments, the elements constituting the embodiment are not necessarily essential unless explicitly stated to be particularly essential or unless they are clearly considered essential in principle. Furthermore, in each of the above embodiments, when numerical values ​​such as the number, numerical values, quantities, or ranges of the components of the embodiment are mentioned, the embodiment is not limited to those specific numbers unless explicitly stated to be particularly essential or unless it is clearly limited to a specific number in principle. Also, in each of the above embodiments, when the shape, positional relationship, etc. of the components are mentioned, the embodiment is not limited to those shapes, positional relationships, etc. unless explicitly stated or unless it is clearly limited to a specific shape, positional relationship, etc. in principle.

[0063] (Perspective of this disclosure) The above disclosure can be understood from the following perspectives, for example.

[0064] [First point of view] An discomfort input unit (20-23) inputs discomfort information, which is information about discomfort emitted by the driver of the vehicle, Based on the aforementioned anomaly information, a scene identification unit (61) identifies the driving scene of the vehicle related to the anomaly, A data extraction unit (62) extracts predetermined driving data related to the driving scene identified by the scene identification unit from among multiple driving data of the vehicle input from the driving data input unit (30), An analysis unit (63) compares the feature quantities stored in the feature recording unit (40), which records driving data from the driver's past driving as individual feature quantities, with predetermined driving data extracted by the data extraction unit, and identifies parameters that should be corrected. A fitting value calculation unit (64) calculates fitting values ​​to fit the parameters identified by the analysis unit to the feature quantities, A vehicle control system comprising: an output unit (65) that outputs a control signal corresponding to the conformance value calculated by the conformance value calculation unit to an in-vehicle device whose parameters can be changed, thereby correcting the parameters. [Second perspective] The vehicle control system according to the first aspect, wherein the aforementioned discomfort input unit inputs information based on the driver's physical response. [Third perspective] The vehicle control system according to the second view, wherein the information based on the aforementioned physical response includes at least one of the driver's brain waves, pulse, gaze, facial expressions, movements, and voice. [Fourth perspective] The vehicle control system according to any one of the first to third aspects, wherein the information input from the driving data input unit is information related to the driver's operation of the vehicle. [Fifth perspective] The aforementioned discomfort input unit inputs the driver's voice as discomfort information, The vehicle control system according to any one of the first to fourth views, wherein the scene identification unit identifies the driving scene of the vehicle related to the discrepancy by keyword matching using a keyword dictionary (50). [Sixth perspective] The vehicle control system according to the fifth aspect, wherein the scene identification unit extracts keywords contained in the sound and identifies the driving scenes associated with the extracted keywords in the keyword dictionary as the driving scenes of the vehicle related to the sense of incongruity. [Seventh perspective] The aforementioned discomfort input unit inputs the driver's voice as discomfort information, The vehicle control system according to any one of the first to sixth views, wherein the scene identification unit identifies the driving scene of the vehicle related to the discrepancy using a learning model obtained by a learning process using learning data including the driver's voice and driving data related to the voice. [Perspective 8] The vehicle control system according to any one of the first to seventh views, wherein the feature recording unit accumulates driving data of the driver on a daily basis and records it as the driver's feature quantity. [Perspective 9] The vehicle control system according to the eighth aspect, wherein the feature recording unit uses a learning model obtained by a learning process using learning data including the amount of operation of an in-vehicle device (32) operated by the driver and driving data output from the vehicle's in-vehicle sensor (31) to record driving data from the driver's daily driving that has a contribution of a predetermined amount or more as the feature quantities. [Perspective 10] The vehicle control system according to any one of the first to seventh or ninth views, wherein the feature recording unit downloads the feature quantities from an external recording medium (100) located outside the vehicle that stores and records the driver's daily driving data as the driver's feature quantities. [Perspective 11] The analysis unit identifies variables that can be changed during the operation of the vehicle, and identifies the actuator (70) of the vehicle that can be used to adjust the variables. The vehicle control system according to any one of the first to tenth aspects, wherein the conformance value calculation unit calculates the conformance value when the actuator is used. [Perspective 12] The vehicle control system according to the eleventh aspect, wherein the output unit stops outputting the control signal when the parameter that is changed, assuming that the output unit outputs the control signal corresponding to the conforming value, falls within a preset danger range. [Perspective 13] The vehicle control system according to the 11th or 12th aspect, wherein the conformance value calculation unit calculates the conformance value in which the variables are minimized when there are multiple variables related to the modification of the parameter. [Perspective 14] A vehicle control method performed by a vehicle control unit (60), This involves inputting information about unusual sensations reported by the driver of the vehicle, and Based on the aforementioned anomaly information, identify the anomaly driving scene, which is a driving scene of the vehicle related to the anomaly, Extracting predetermined driving data related to the abnormal driving scene from among multiple driving data of the vehicle input from the driving data input unit (30), The feature recording unit (40), which records driving data from the driver's past driving as individual feature quantities, compares the feature quantities stored in the feature recording unit with the extracted predetermined driving data to identify parameters that should be corrected, To calculate fitting values ​​for fitting the identified parameters to the feature quantities, A vehicle control method comprising outputting a control signal corresponding to the calculated suitable value to an in-vehicle device capable of changing the aforementioned parameters, thereby causing the parameters to be modified. [Perspective 15] A vehicle control program executed by the vehicle control unit (60), The process involves inputting information about discomfort, which is information about discomfort reported by the driver of the vehicle, Based on the aforementioned anomaly information, a process is performed to identify an anomaly driving scene, which is a driving scene of the vehicle related to the anomaly. A process to extract predetermined driving data related to the abnormal driving scene from among multiple driving data of the vehicle input from the driving data input unit (30), A feature recording unit (40) records driving data from the driver's past driving as individual feature quantities. The feature quantities stored in the feature recording unit compare these with the extracted predetermined driving data and identify parameters that need to be corrected. A process to calculate fitted values ​​for fitting the identified parameters to the feature quantities, A vehicle control program that includes a process of outputting a control signal corresponding to the calculated suitable value to an in-vehicle device whose parameters can be changed, thereby correcting the parameters. [Explanation of symbols]

[0065] 20...Discomfort input unit, 21...Input device, 22...Driver imaging device, 23...Measurement device, 30...Driving data input unit, 31...On-board sensor, 32...Pedal device, 40...Feature recording unit, 50...Keyword dictionary, 60...Vehicle control unit, 61...Scene identification unit, 62...Data extraction unit, 63...Analysis unit, 64...Fitment value calculation unit, 65...Output unit, 70...Actuator, 100...External recording medium

Claims

1. An discomfort input unit (20-23) that inputs discomfort information, which is information about discomfort emitted by the driver of the vehicle, Based on the aforementioned anomaly information, a scene identification unit (61) identifies the driving scene of the vehicle related to the anomaly, A data extraction unit (62) extracts predetermined driving data related to the driving scene identified by the scene identification unit from among multiple driving data of the vehicle input from the driving data input unit (30), An analysis unit (63) compares the feature quantities stored in the feature recording unit (40), which records driving data from the driver's past driving as individual feature quantities, with predetermined driving data extracted by the data extraction unit, and identifies parameters that should be corrected. A fitting value calculation unit (64) calculates fitting values ​​to fit the parameters identified by the analysis unit to the feature quantities, A vehicle control system comprising: an output unit (65) that outputs a control signal corresponding to the conformance value calculated by the conformance value calculation unit to an in-vehicle device whose parameters can be changed, thereby causing the parameters to be modified.

2. The vehicle control system according to claim 1, wherein the abnormality input unit inputs information based on the driver's physical reaction.

3. The vehicle control system according to claim 2, wherein the information based on the physical response includes at least one of the driver's brain waves, pulse, gaze, facial expressions, movements, and voice.

4. The vehicle control system according to claim 1, wherein the information input from the driving data input unit is information related to the operation of the vehicle by the driver.

5. The aforementioned discomfort input unit inputs the driver's voice as discomfort information, The vehicle control system according to claim 1, wherein the scene identification unit identifies the driving scene of the vehicle related to the discrepancy by keyword matching using a keyword dictionary (50).

6. The vehicle control system according to claim 5, wherein the scene identification unit extracts keywords contained in the sound and identifies the driving scenes associated with the extracted keywords in the keyword dictionary as the driving scenes of the vehicle related to the sense of incongruity.

7. The aforementioned discomfort input unit inputs the driver's voice as discomfort information, The vehicle control system according to claim 1, wherein the scene identification unit identifies the driving scene of the vehicle related to the discrepancy using a learning model obtained by a learning process using learning data including the driver's voice and driving data related to the voice.

8. The vehicle control system according to claim 1, wherein the feature recording unit accumulates driving data of the driver on a daily basis and records it as the driver's feature quantity.

9. The vehicle control system according to claim 8, wherein the feature recording unit uses a learning model obtained by a learning process using learning data including the amount of operation of an in-vehicle device (32) operated by the driver and driving data output from the vehicle's in-vehicle sensor (31) to record driving data from the driver's daily driving that has a contribution of a predetermined amount or more as the feature quantities.

10. The vehicle control system according to claim 1, wherein the feature recording unit downloads the feature quantities from an external recording medium (100) located outside the vehicle that stores and records the driver's daily driving data as the driver's feature quantities.

11. The analysis unit identifies variables that can be changed during the operation of the vehicle, and identifies the actuator (70) of the vehicle that can be used to adjust the variables. The vehicle control system according to claim 1, wherein the conformance value calculation unit calculates the conformance value when the actuator is used.

12. The vehicle control system according to claim 11, wherein the output unit stops outputting the control signal when the parameter that is changed, assuming that the control signal corresponding to the conforming value is output, falls within a preset danger range.

13. The vehicle control system according to claim 11, wherein the conformance value calculation unit calculates the conformance value in which the variables are minimized when there are multiple variables related to the modification of the parameters.

14. A vehicle control method performed by a vehicle control unit (60), This involves inputting information about unusual sensations reported by the driver of the vehicle, and Based on the aforementioned anomaly information, identify the anomaly driving scene, which is a driving scene of the vehicle related to the anomaly, The system extracts predetermined driving data related to the abnormal driving scene from among multiple driving data of the vehicle input from the driving data input unit (30), The feature recording unit (40), which records driving data from the driver's past driving as individual feature quantities, compares the feature quantities stored in the feature recording unit with the extracted predetermined driving data to identify parameters that need to be corrected, To calculate fitting values ​​for fitting the identified parameters to the feature quantities, A vehicle control method comprising outputting a control signal corresponding to the calculated suitable value to an in-vehicle device capable of changing the aforementioned parameters, thereby causing the parameters to be modified.

15. A vehicle control program executed by the vehicle control unit (60), The process involves inputting information about discomfort, which is information about discomfort reported by the driver of the vehicle, Based on the aforementioned anomaly information, a process is performed to identify an anomaly driving scene, which is a driving scene of the vehicle related to the anomaly. A process to extract predetermined driving data related to the abnormal driving scene from among multiple driving data of the vehicle input from the driving data input unit (30), A feature recording unit (40) records driving data from the driver's past driving as individual feature quantities. The feature quantities stored in the feature recording unit compare these with the extracted predetermined driving data and identify parameters that need to be corrected. A process to calculate fitted values ​​for fitting the identified parameters to the feature quantities, A vehicle control program that includes a process of outputting a control signal corresponding to the calculated suitable value to an in-vehicle device whose parameters can be changed, thereby correcting the parameters.

Images