Driving assistance systems

Abstract

This driver assistance system provides optimal driving support at the optimal timing, taking into account the driver's internal model and emotional control process model, thereby effectively improving the driver's driving skills. [Solution] The driver assistance device includes a driver achievement calculation means 35 that calculates the degree of driving achievement in a target driving scene, a driver evaluation means 36 that determines whether the degree of driving achievement meets the evaluation criteria, a driver assistance control means 37 that controls the HMI 26 to provide a positive evaluation to the driver when the degree of driving achievement meets the evaluation criteria, a driver state estimation means 38 that estimates the driver's internal state, and an evaluation criterion changing means 39 that changes the evaluation criteria based on the driver's internal state. The evaluation criterion changing means 39 relaxes the evaluation criteria when it is estimated that the driver has negative emotions.

Description

Technical Field

[0001] The present invention relates to a driving support device provided in a vehicle such as an automobile, and particularly relates to an improvement that can effectively improve the driving skills of a driver by performing driving support considering the driver's internal model (driver internal model).

Background Art

[0002] In recent years, vehicles such as automobiles have come to be equipped with driving support devices (for example, advanced driving support systems (ADAS)) that support the driving of a driver. Some of these driving support devices provide information (for example, coaching) as driving support for the driver, and various technologies have been proposed in relation to such driving support.

[0003] For example, in Patent Document 1 (Japanese Unexamined Patent Application Publication No. 2022-178816), in a vehicle control device, the driving (driving performance) of a driver is observed by dividing it into recognition, judgment, and operation, and for the driving performance determined to be insufficient, driving support (coaching) is performed at an optimal timing to improve the driving skills of the driver. Also, in Patent Document 2 (Japanese Patent No. 6428748), in a driving support system, a technique for guiding a driver of a vehicle toward a target emotional state (for example, a pleasant state) by appropriately applying sensory stimuli (applying visual and auditory stimuli, changing the accelerator sensitivity, the sensitivity of the steering wheel, etc.) has been proposed. Further, in Patent Document 3 (Japanese Patent No. 6221776), in a driving evaluation device, a technique has been proposed that can allow a driver to recognize whether the driving operation performed by the driver was appropriate by evaluating the driving operation and notifying the driver of the evaluation result.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

[0005] Thus, even in conventional driver assistance systems, technologies have been proposed to improve drivers' driving skills and emotional well-being by devising ways to optimize the timing and content of driver assistance (information provision and coaching). However, conventional technologies have not considered (modeled) the processing within the driver's brain to optimize driver assistance. In other words, conventional driver assistance systems have not considered the driver's brain state (driver internal model) and emotional control processes when providing driver assistance to the driver, in order to provide the optimal driver assistance at the optimal timing.

[0006] This invention was made in consideration of the circumstances described above, and aims to provide a driving assistance device that can effectively improve a driver's driving skills by providing optimal driving assistance at the optimal timing, taking into account the driver's internal model and the process model of emotional control. [Means for solving the problem]

[0007] To achieve the above objective, the present invention adopts the following solution. That is, as described in claim 1, a driving assistance device that assists the driving of a vehicle driver, comprising: driving scene recognition means for recognizing the driving scene of the vehicle; driving detection means for detecting the driving operations of the driver and / or the behavior of the vehicle; driving target calculation means for calculating target driving operations and / or target vehicle behavior that are suitable for the driving scene; and interface means for providing information to the driver. The system includes: a driving achievement calculation means that calculates a driving achievement level by comparing the driving operations of the driver and / or the behavior of the vehicle detected by the driving detection means with the target driving operations and / or target vehicle behavior calculated by the driving target calculation means; a driving evaluation means that determines whether the driving achievement level calculated by the driving achievement calculation means meets predetermined evaluation criteria; a driving support control means that controls the interface means to provide the driver with a positive evaluation of the driver's driving after the end of a driving scene if it is determined that the driving achievement level in a driving scene subject to driving support meets the evaluation criteria; a driver information acquisition means capable of detecting the physiological state quantities and / or behavior of the driver; a driver state estimation means that estimates the internal state of the driver based on the physiological state quantities and / or behavior detected by the driver information acquisition means; and an evaluation criterion changing means that changes the evaluation criteria for driving evaluation in the driving evaluation means based on the internal state of the driver estimated by the driver state estimation means.

[0008] According to the above solution, the evaluation criteria for driving evaluation in the driving evaluation means are changed according to the driver's internal state (emotional state) estimated based on the driver's physiological state and / or behavior. Therefore, appropriate driving support (provision of positive evaluation) is performed based on appropriate evaluation criteria that take into account the impact of the driver's emotional state on driving performance. Consequently, evaluations praising the driver's driving content are appropriately provided based on criteria suitable for the individual driver's state, thereby promoting desirable cognitive changes in the driver and achieving effective improvement of the driver's internal model and smooth improvement of the driver's driving skills.

[0009] A preferred embodiment based on the above solution method is as described in claim 2 and subsequent claims of the patent. That is, if the evaluation criterion changing means is estimated to have negative feelings while driving a driving scene that is the subject of driving assistance, it relaxes the evaluation criteria for that driving scene (corresponding to claim 2). In this case, if the driver has negative feelings, the evaluation criteria for driving evaluation are relaxed, so even if the degree of driving achievement in the target driving scene does not reach the evaluation criteria before relaxation, if a certain degree of driving achievement is achieved, the driver is given a positive evaluation. Therefore, even if negative feelings arise because the driver has difficulty with that driving scene, and these negative feelings reduce driving performance, resulting in a slightly lower driving achievement, the driver is informed that they are driving at an acceptable level, so the driver's sense of self-efficacy can be appropriately increased and the occurrence of negative feelings can be suppressed.

[0010] The relaxation of the evaluation criteria is achieved by reducing the evaluation criterion value in the evaluation criteria (corresponding to claim 3). In this case, the relaxation of the evaluation criteria can be accurately achieved by changing the evaluation criterion value, with an appropriate amount of change.

[0011] The relaxation of the evaluation criteria is achieved by dividing the driving scene subject to driver assistance into multiple driving scenes, and by performing a determination by the driving evaluation means and providing a positive evaluation by the interface means for each divided driving scene (corresponding to claim 4). In this case, even if the driving does not meet the evaluation criteria for the entire driving scene before division, a positive evaluation is provided for the driving scenes among the divided driving scenes that meet the evaluation criteria, so that the evaluation criteria can be appropriately relaxed and the driver can accurately recognize in which of the divided driving scenes they were able to drive appropriately.

[0012] The driving assistance control means controls the interface means to provide the driver with suggestions regarding the driver's driving operations and / or the vehicle's behavior in a driving scene before or during the driving scene to which driving assistance is to be provided (corresponding to claim 5). In this case, pre-assistance (e.g., providing suggestions regarding driving operations and / or the vehicle's behavior) is provided at the timing of attention allocation when the emotional control process model is applied to driving a vehicle, and post-assistance (providing positive evaluation) is provided at the timing of cognitive change. As a result, driving assistance is provided that matches the emotional control process model from attention allocation to cognitive change, effectively supporting driving based on the driver's appropriate attention allocation and effectively bringing about desirable cognitive changes in the driver. [Effects of the Invention]

[0013] According to the present invention, in a driver assistance device that provides a positive evaluation to a driver after the completion of a driving scene when the driver's driving performance in that driving scene meets the evaluation criteria, the evaluation criteria are changed according to the driver's internal state (the evaluation criteria are relaxed if the driver has negative emotions), so that appropriate driver assistance (provision of a positive evaluation) that appropriately matches the internal state of each individual driver is provided. This effectively enhances the driver's sense of self-efficacy and effectively brings about desirable cognitive changes in the driver (a change in the perception that one can drive that driving scene appropriately), thereby achieving effective improvement of the driver's internal model and smooth improvement of the driver's driving skills. Furthermore, as the driver's internal model is improved through driver assistance, the driver can accumulate successful experiences through driving and gain a sense of happiness through driving a vehicle. [Brief explanation of the drawing]

[0014] [Figure 1] This figure shows the internal driver model in the present invention. [Figure 2]This is a diagram for explaining the relationship between the present invention and the emotion control process model. [Figure 3] This is a block configuration diagram showing an example of the control system of the driving support device of the present invention. [Figure 4] This is a diagram showing a scene where the host vehicle decelerates in the merging parallel section. [Figure 5] This is a diagram showing a scene where the host vehicle accelerates in the merging parallel section. [Figure 6] This is a diagram showing a scene where the lane change is completed during merging. [Figure 7] This is a flowchart showing a control example of the present invention.

Embodiments for Carrying Out the Invention

[0015] Hereinafter, embodiments of the present invention will be described based on the accompanying drawings. FIG. 1 shows a driver internal model 1 that models the driver's brain state during vehicle operation. This driver internal model 1 was constructed by the inventor of the present invention by applying the concept of the free energy principle regarding human brain functions, particularly to the brain processing of the driver during vehicle operation. The present invention constitutes a driving support device based on this driver internal model 1. Therefore, first, the content of the driver internal model 1 will be described in detail.

[0016] The free energy principle is a theory proposed by Friston, which uniformly explains various functions of the human brain as attempting to minimize free energy. According to the free energy principle, the internal model of the human brain is updated (changed) so as to minimize the prediction error.

[0017] As shown in FIG. 1, the driver internal model 1 includes a motion system internal model 1A and a body system internal model 1B. Here, the motion system internal model 1A models the function of making inferences in the driver's brain during driving in the motion system (i.e., predicting the behavior of the vehicle with respect to the driving operation) and controlling body movements (driving operations) based on this inference. On the other hand, the body system internal model 1B models the function of making inferences in the driver's brain during driving in the body system (i.e., predicting the internal state of the viscera, blood vessels, etc. during vehicle driving) and controlling the internal state (the state of the viscera, blood vessels, etc.) based on this inference.

[0018] In the driver's brain during vehicle driving, a motion system processing loop 3A for reducing the motion system prediction error 2A centered on the motion system internal model 1A (in the figure, a processing loop including the motion system internal model 1A → prediction of results 11 → driving operation 12 → vehicle behavior 13 → perceptual awareness 14 → motion system prediction error 2A) and a body system processing loop 3B for reducing the body system prediction error 2B centered on the body system internal model 1B (in the figure, a processing loop including the body system internal model 1B → prediction of internal state 15 → internal state 16 → body system prediction error 2B) are being processed.

[0019] Here, the motion system prediction error 2A is recognized by the motion system internal model 1A as the difference between the predicted vehicle behavior (prediction of results 11) by the driver and the actual vehicle behavior 13 (the realized result). On the other hand, the body system prediction error 2B is recognized by the body system internal model 1B as the difference between the predicted internal state 15 of the driver and the actual internal state 16. In the driver internal model 1, as the motion system prediction error 2A and the body system prediction error 2B are reduced in the motion system processing loop 3A and the body system processing loop 3B, the motion system internal model 1A and the body system internal model 1B are updated.

[0020] To explain in more detail below, while driving a vehicle, information regarding the vehicle's driving conditions (for example, the driving scene in which the vehicle is moving and various associated information) is perceived and recognized by the driver (perception / recognition 10). Based on this perception / recognition 10, the internal motion system model 1A infers the vehicle's behavior that should be performed in that situation (driving scene), and this is the predicted result 11 (predicted vehicle behavior).

[0021] Furthermore, the internal motion system model 1A infers the driving operations 12 necessary to obtain the predicted outcome 11 (predicted vehicle behavior). In other words, the internal motion system model 1A has hypotheses (beliefs) for inferring (predicting) what driving operations should be performed to obtain the predicted vehicle behavior, and based on these hypotheses, the driving operations 12 to be performed are inferred. The driver will then perform the driving operations 12 in accordance with this inference (judgment).

[0022] For example, in a driving scenario involving a lane change, the system predicts the necessary vehicle behavior for an appropriate lane change (e.g., a suitable driving path for the lane change) based on perceptions such as the presence of a vehicle at the destination and the presence of a vehicle behind (prediction of the result 11). It then infers and executes specific driving operations 12 necessary to obtain the predicted result, such as steering to change lanes or pressing the pedal to adjust vehicle speed.

[0023] The driving operations 12 performed in this manner result in the actual vehicle behavior 13, and this actual vehicle behavior 13 (the realized result) is perceived and recognized by the driver (perceptual recognition 14). Specifically, the vehicle behavior 13 is recognized through visual (optical flow) and auditory confirmation of the vehicle behavior, as well as perceptions such as the feeling of acceleration (G value).

[0024] The perceived vehicle behavior 13 is compared with the predicted outcome 11, and the difference between the predicted outcome 11 (predicted vehicle behavior) and the actual vehicle behavior 13 (actual result) is recognized as the motion system prediction error 2A. For example, if the timing of the driving operation 12 (e.g., steering timing) that was judged to be necessary to obtain the predicted outcome 11 (predicted vehicle behavior) is actually too late to obtain the predicted outcome 11, the delay in the vehicle behavior 13 relative to the predicted outcome 11 is recognized as the motion system prediction error 2A.

[0025] In vehicle operation, driving operations 12 are adjusted to reduce the motion system prediction error 2A. Then, the internal motion system model 1A is updated by changing the reasoning content (hypothesis for reasoning) regarding driving operations to reduce the motion system prediction error 2A. This positive update of the internal motion system model 1A (an update that reduces the motion system prediction error 2A) enables the driver to perform appropriate driving operations, and the driver's driving skills improve.

[0026] For example, if previously the steering timing for lane changes was delayed, resulting in an inability to perform smooth lane changes, updating the internal model 1A of the motion system to reduce the motion system prediction error 2A (by adopting a new hypothesis that steering needs to be performed earlier for lane changes) will appropriately advance the steering timing for lane changes, allowing steering to be performed at the appropriate time.

[0027] On the other hand, in the internal system processing loop 3B, the internal system internal model 1B makes predictions about the internal state (state of internal organs, blood vessels, etc.) based on perception and cognition 10 (prediction of internal state 15). In other words, the internal system internal model 1B has hypotheses (beliefs) for inferring (predicting) what the internal state will be like in relation to the state recognized by perception and cognition 10, and the internal state is predicted based on these hypotheses.

[0028] This predicted internal state 15 is compared with the actual internal state 16, and the difference between the predicted internal state 15 and the actual internal state 16 becomes the internal system prediction error 2B. The internal system internal model 1B is updated to reduce the internal system prediction error 2B, and as a result, the internal system (autonomic nervous system, etc.) is adjusted so that the predicted internal state 15 and the actual internal state 16 match.

[0029] In the brain processing described above, the motor system internal model 1A and the internal system internal model 1B interact with each other, and if the motor system internal model 1A changes, this change will affect the internal system internal model. Furthermore, the inference and internal system prediction errors 2B in the internal system internal model 1B are the cause of the driver's emotions 17.

[0030] For example, if the vehicle is in a situation it is not comfortable with (for instance, attempting to merge onto a main road), the internal model 1B predicts that the driving will not go well in that situation, causing an increase in heart rate, and the actual heart rate will also increase. Furthermore, such negative reasoning can also trigger negative emotions such as anxiety (fear) about driving and a lack of enjoyment in driving.

[0031] Conversely, when the internal motor system model 1A is improved (positively updated) and driving skills improve, resulting in smoother driving operations, this improvement in the internal motor system model 1A positively impacts the internal body system model 1B. Specifically, the internal body system model 1B is positively updated to reflect confidence in driving in that driving scenario, no longer predicting an increase in heart rate, and the actual increase in heart rate disappears. Furthermore, positive reasoning regarding that driving scenario (the prediction that the driving operations will be performed well) also triggers positive emotions such as confidence, a sense of security (calmness), and joy in driving (happiness) when faced with that driving scenario.

[0032] Thus, if the internal model 1A of the motor system is updated appropriately, driving skills in that driving scenario will naturally improve, and the driver's emotions 17 will also improve. On the other hand, if the positive updates of the internal model 1A of the motor system are not updated appropriately, driving skills will never improve, and inappropriate driving operations (driving operations with a large motor system prediction error 2A) will be repeated. Furthermore, negative emotions such as anxiety about driving and a lack of enjoyment in driving will persist.

[0033] Furthermore, in relation to this driver internal model 1, the inventors of the present invention have found that "by accumulating small successes, a sense of self-efficacy is nurtured, and ultimately, this sense of self-efficacy becomes generalized, making it easier to feel happy on a regular basis (happiness level improves)." The positive updates of the motor system internal model 1A and the internal system internal model 1B according to the present invention are small successes, and are thought to lead to an improvement in the driver's happiness level.

[0034] Therefore, as the appropriate driving assistance in this invention is repeatedly performed while the vehicle is in motion, positive experiences (feelings of happiness) through the driving assistance are accumulated as small successes, ultimately fostering a sense of self-efficacy (for example, confidence in being able to drive the vehicle), and as this sense of self-efficacy generalizes, it is possible to achieve a state in which the driver's happiness (constant sense of well-being) is enhanced.

[0035] Next, following Figure 2, we will describe another insight used to configure the driver assistance device of the present invention: the emotion control process model. The emotion control process model is a model of cognitive processing proposed by Gross, which views emotion regulation in the following five stages.

[0036] As shown in Figure 2, according to the process model of emotion regulation, human emotion regulation is achieved by sequentially executing the following stages: situation selection → situation modification → attention allocation → cognitive modification → response adjustment. Here, in the situation selection stage, when faced with a situation that is expected to evoke some emotion, emotions are controlled by choosing to either select or avoid that situation. In the situation modification stage, emotions are controlled by changing the situation itself.

[0037] Furthermore, in the attention allocation stage (adjusting the direction of attention), emotions are controlled by changing how attention is directed. In the cognitive modification stage (changing cognitions), emotions are controlled by changing the interpretation of events (cognitive reappraisal). In the response regulation stage (adjusting emotional responses), emotions are controlled by adjusting the expression and behavior of emotional responses.

[0038] It is believed that a similar process of emotional control is also at play when driving a vehicle. For example, in a lane change scenario, first, in the situation selection phase, a choice is made whether to change lanes or not. In the situation modification phase, the driver adjusts the position of their vehicle relative to a vehicle traveling in the lane they intend to change lanes into, thereby modifying the situation to allow for the lane change.

[0039] In terms of attention allocation, lane changes are executed while paying attention to the driving conditions of one's own vehicle and adjacent vehicles. For example, the distance between the adjacent vehicle to be entered in the lane change is determined, the driver accelerates or decelerates their own vehicle while driving alongside the adjacent vehicle to adjust their position relative to that distance, and then steers to enter that distance and complete the lane change.

[0040] In cognitive transformation, a self-assessment of driving behavior during lane changes is made, and cognitive reappraisal is performed based on this assessment. For example, if a driver who previously struggled with lane changes successfully performs one this time, a cognitive transformation occurs where they believe that this driving behavior was appropriate and that they can successfully perform lane changes by performing this driving behavior.

[0041] In relation to this emotional control process model, the inventors of the present invention have found from experiments using driving simulators that when a driver successfully drives in a driving scenario in which they feel uncomfortable, providing the driver with positive feedback on their driving (praise for their driving) at the timing of cognitive changes in vehicle operation enhances the driver's self-efficacy and positively updates their internal model. For example, it has been confirmed in tests that evaluate drivers' cognitive abilities (divergent convergence tests) that drivers who receive positive feedback at the appropriate time show an improvement in their cognitive abilities.

[0042] Furthermore, it has been found that if drivers are provided with information about driving in a given situation (for example, coaching on the content and timing of driving operations to be performed in that driving scene) at the right time to allocate their attention while driving a vehicle, they will be able to allocate their attention appropriately in that situation.

[0043] Based on the above considerations, the present invention provides a driver assistance device that provides information to the driver. By providing appropriate driver assistance that takes into account the driver's internal model 1 and the process model of emotion control, the invention aims to improve the driver's driving skills and the driver's emotional state. A specific example of the configuration of the driver assistance device of the present invention will be described in detail below.

[0044] Figure 3 shows a block diagram of an example of the control system in the driver assistance device of the present invention. The driver assistance device is installed in a vehicle such as an automobile or a device that simulates driving a vehicle (driving simulator), such as an advanced driver assistance system (ADAS).

[0045] As shown in the figure, the control system includes a control unit U, an external information acquisition means 21, a driving information acquisition means 22, a vehicle information acquisition means 23, a driver monitoring means 24, a storage means 25, and an HMI (Human-Machine Interface) 26.

[0046] The external information acquisition means 21 is a means for acquiring external information about the vehicle's driving state (external environment of the vehicle, such as roads and other vehicles), and consists of, for example, an external camera that takes pictures of the outside of the vehicle and a sensor that detects the conditions outside the vehicle.

[0047] The driving information acquisition means 22 is a means for detecting the state of driving operations and includes, for example, a steering angle sensor, an accelerator sensor, a brake sensor, etc. The vehicle information acquisition means 23 is a means for acquiring (detecting) vehicle conditions such as vehicle speed and includes, for example, a vehicle speed sensor, an acceleration sensor, etc.

[0048] The driver monitoring means 24 is a means for monitoring the driver of a vehicle and acquiring information about the driver's condition, and is composed of, for example, an in-vehicle camera that photographs the driver.

[0049] The storage means 25 is a means for storing various types of data, and is, for example, an external storage device. The storage means 25 stores various types of information, such as road information, information about standard driver models, and information about individual vehicle models, as well as the vehicle's driving history (history of driving operations and vehicle behavior for each driver).

[0050] The HMI26 is a device that provides visual and auditory information to the vehicle driver in driving assistance, and includes notification means such as a display 26A capable of displaying images and characters (e.g., an in-car monitor or a head-up display (HDU)) and a speaker 26B capable of outputting sound. In this embodiment, the HMI26 corresponds to the "interface means" in the claims.

[0051] The control unit U is a control device composed of, for example, a microcomputer, and includes a driving scene recognition means 31, a driving target calculation means 32, a driving operation detection means 33, a vehicle behavior detection means 34, a driving achievement level calculation means 35, a driving evaluation means 36, a driving support control means 37, a driver state estimation means 38, and an evaluation criterion changing means 39. These means are provided as a control program within the control unit U.

[0052] The driving scene recognition means 31 is a means for recognizing (identifying) the vehicle's driving scene (the situation in which the vehicle is driving) based on external information acquired by the external information acquisition means 21. By identifying the driving scene, the driving scene recognition means 31 can determine whether the driving scene is subject to driver assistance and, if so, what kind of driver assistance should be provided. The driving scenes are grouped according to their content (for example, lane changes, hill starts, etc.), and driving scenes belonging to the same group are treated as the same driving scene.

[0053] The driving target calculation means 32 is a means for calculating the (appropriate) driving operations and vehicle behaviors that should be performed in a specific driving scene recognized by the driving scene recognition means 31, as target driving operations and target vehicle behaviors. The calculation of target driving operations and target vehicle behaviors is performed based on the content of the recognized specific driving scene, by referring to various information (driving road information, information on standard driver models, information on individual vehicle models) stored in the storage means.

[0054] The driving operation detection means 33 is a means for detecting the driver's driving operations for each driving scene based on the information (detection signal) acquired by the driving information acquisition means 22. The detected driving operation data for each driving scene is used to calculate the driving achievement level, which will be described later, and is also stored in the storage means 25 as data for each driver.

[0055] The vehicle behavior detection means 34 is a means for detecting (calculating) the behavior of the vehicle in a driving scene based on the information (detection signal) acquired by the vehicle information acquisition means 23. The detected vehicle behavior data for each driving scene is used to calculate the driving achievement level described later, and is also stored in the storage means 25 as data for each driver. In this embodiment, the driving operation detection means 33 and the vehicle behavior detection means 34 correspond to the "driving detection means" in the claims.

[0056] The driving achievement calculation means 35 is a means for calculating the driving achievement level (an index indicating the degree of difference between the driving operation and / or the vehicle behavior) by comparing the driver's driving operation and / or the vehicle behavior detected by the driving operation detection means 33 and / or the vehicle behavior detected by the vehicle behavior detection means 34 with the target driving operation and / or target vehicle behavior calculated by the driving target calculation means 32.

[0057] The driving performance level is calculated based on an integrated value obtained by adding up the differences between the target driving operation and the driver's driving operation in the driving scene being supported (e.g., the absolute value of the difference in driving operation amounts such as steering amount) and / or the difference between the target vehicle behavior and the vehicle behavior (e.g., the absolute value of the difference in behavior amounts such as the driving path) over the entire driving scene, with appropriate weights as needed. The system is set so that a larger integrated value indicates a lower driving performance level, and a smaller integrated value indicates a higher driving performance level.

[0058] The driving evaluation means 36 is a means for evaluating driving in a given driving scene based on the driving achievement level calculated by the driving achievement level calculation means 35. For example, the driving evaluation means 36 compares the driving achievement level with a predetermined evaluation standard value. If the driving achievement level is equal to or greater than the evaluation standard value, it performs a positive evaluation (a positive evaluation indicating that appropriate driving was performed in the driving scene). If the driving achievement level is less than the evaluation standard value, it performs a negative evaluation (a negative evaluation indicating that the driving in the driving scene was insufficient). The driving assistance in this invention is performed based on the driving evaluation performed by the driving evaluation means 36.

[0059] The driver assistance control means 37 determines whether the current driving scene is eligible for driver assistance based on the driving scene recognition means 21, and if the driving scene is eligible for assistance, it controls the HMI 26 to perform driver assistance. Driver assistance via the HMI 26 includes pre-assistance (provision of information in advance) and post-assistance (provision of information in later).

[0060] Pre-driving support is a type of driver assistance that provides coaching and guidance to ensure proper driving in a driving scene before (for example, immediately before) or while driving in a driving scene that is subject to driver assistance. This support is provided as the display of images and text on the HMI26's display 26A, and as sound output such as voice and notification sounds from the speaker 26B.

[0061] Specific examples of pre-support include providing information that gives an overall picture of the driving to be performed in a given driving scene, shortly before entering that scene. Furthermore, while driving through the scene, suggestions regarding individual driving operations (e.g., accelerator operation or steering operation) are provided at the time when those operations should be performed (e.g., immediately before performing those operations).

[0062] On the other hand, post-event support is driving assistance that is performed after the supported driving scene has ended (for example, immediately afterward) if the driving evaluation by the driving evaluation means 36 is positive, and is performed by providing the driver with information that positively evaluates the driving in that driving scene. Specifically, for example, a display praising the driving on the display 26A (for example, a display of "Good Job!") and / or an audio output praising the driving from the speaker 26B (for example, an audio output of "Good Job") are provided.

[0063] In this embodiment, post-operation support is performed only if the operation evaluation by the operation evaluation means 36 is positive, and is not performed if the operation evaluation is negative.

[0064] In this way, when a driver successfully performs a driving scenario, a positive evaluation (praise) is provided at the time the scenario is completed. This allows the driver to gain confidence in their ability to drive well in that scenario, increasing their self-efficacy. Furthermore, a positive self-evaluation is formed, leading to the belief that they can perform appropriate driving in the next scenario by exhibiting similar driving behaviors (positive cognitive transformation occurs). As the driver's internal motor system model 1A is positively rewritten in this way, the interaction between the internal motor system model 1A and the internal body system model 1B also positively rewrites the internal body system model 1B, resulting in a positive improvement in the driver's emotions (and even their sensibilities, including value judgments). As a result, the driver's internal model 1 is positively updated, improving the driver's driving skills and increasing their sense of happiness while driving.

[0065] Figures 4 to 6 show a driving scene of merging (for example, merging onto the main lane of a highway) as a specific application example of the present invention. As shown in the figures, when merging, the vehicle 41 enters the acceleration lane (merging lane) 42 and then changes lanes to the driving lane 43 adjacent to the right side of the acceleration lane 42. There are parallel vehicles 44 to 46 traveling in the driving lane 43, and the vehicle 41 attempts to enter the gap 51 or 52 formed between the parallel vehicles 44 to 46.

[0066] To explain in more detail, when vehicle 41 enters the acceleration lane 42, it begins to travel alongside vehicles 44-46 traveling in the driving lane 43. In this embodiment, the timing of the start of this parallel travel is the start of the merging driving scene.

[0067] Next, the system determines the gap between vehicles to enter when changing lanes (the space formed by the vehicles driving in front and behind), and then accelerates or decelerates its own vehicle 41 so that it is positioned in the optimal location for entering this gap (for example, to the side of the position to be entered).

[0068] Figure 4 shows a case where a vehicle changes lanes into a gap 51 formed between parallel vehicles 45 and 46 behind its own vehicle 41. In this case, the vehicle 41 is positioned relative to the gap 51 by decelerating. On the other hand, Figure 5 shows a case where a vehicle changes lanes into a gap 52 formed between parallel vehicles 44 and 45 in front of its own vehicle 41. In this case, the vehicle 41 is positioned relative to the gap 52 by accelerating its own vehicle 41.

[0069] Once vehicle 41 is in an appropriate position relative to the gap 51 or 52, the steering wheel is turned toward the driving lane 43, moving vehicle 41 toward the gap 51 or 52. Figure 6 shows the state after the lane change to the gap 52 between parallel vehicles 44 and 45 has been completed. With the completion of this lane change, the merging driving scene ends (that is, the timing of the completion of the lane change marks the end of the merging driving scene).

[0070] In such merging driving scenarios, the driver assistance system provides pre-merging support at the start of merging (when parallel driving begins) and / or during merging (while performing driving operations such as accelerating, decelerating, or changing lanes), and post-merging support after the merging is completed (after the lane change is completed).

[0071] Specifically, as pre-merging support, suggestions regarding the content of driving operations and various information required throughout the entire merging process are provided at the start of or immediately before the start of parallel merging. Furthermore, during merging, suggestions regarding the content and timing of driving operations to be performed at each stage of merging (such as accelerator operation for accelerating or decelerating the vehicle 41, and steering operation to maintain a safe distance between vehicles 51 or 52) are provided via the HMI 26 at the optimal timing for each driving operation (for example, immediately before each operation). This allows the driver to appropriately allocate their attention to driving during merging.

[0072] On the other hand, post-merging support is provided by offering a positive evaluation (a compliment on driving) via the HMI 26 after the merging is complete, provided that the driving performance during the merging meets the evaluation criteria (is above the evaluation criteria value). For example, at the moment when the lane change of the vehicle 41 to the space between vehicles 51 or 52 is completed (immediately after completion), the display 26A shows "Good Job!" and the speaker 26B outputs the voice message "Good Job".

[0073] Thus, when merging is performed appropriately, the driver is given a positive driving evaluation at the appropriate time. As a result, even drivers who previously had difficulty merging gain confidence that they can merge appropriately next time (positive cognitive change occurs), and the driver's internal model 1 is updated favorably (self-efficacy increases).

[0074] Returning to Figure 3, the driver state estimation means 38 is a means for estimating the driver's internal state (emotions and brain state) based on information acquired by the driver monitoring means 24. More specifically, the driver state estimation means 38 first identifies who the driver is by analyzing information acquired by the driver monitoring means 24 (for example, images of the driver taken by the in-car camera), and also detects (estimates) the driver's physiological state (for example, measured values ​​such as heart rate, respiration, and sweating, and characteristic quantities obtained from various measured values ​​(for example, irregularities in heart rhythm)), and also detects the driver's behavior (for example, behaviors indicating psychological distress or anxiety, irregular behavior, etc.).

[0075] Next, the driver state estimation means 38 estimates the driver's internal state (emotions and psychological state) based on the detected physiological state quantities and behaviors of the driver. Specifically, if the driver state estimation means 37 detects changes in physiological state quantities or characteristic behaviors that suggest a particular emotion of the driver, it estimates that the driver is experiencing that particular emotion. For example, if a driver is experiencing negative emotions (e.g., anxiety or stress), it is thought that their heart rate will increase and their sweating will increase. Therefore, if the driver's heart rate rises above a predetermined threshold or their sweating increases above a predetermined threshold, it is estimated that the driver is experiencing negative emotions (e.g., anxiety). The physiological state quantities and behaviors used for estimation can be combined as appropriate.

[0076] In relation to the claims, the driver monitoring means 24 and the driver state estimation means 38's function for detecting the physiological state and behavior of the driver in this embodiment correspond to the "driver information acquisition means" in the claims.

[0077] Furthermore, while the above example shows a method of estimating the driver's physiological state by analyzing images of the driver, the driver's physiological state can also be directly detected by means of detecting the physiological state that are directly attached to the driver (for example, a heart rate monitor to measure heart rate or a sweat sensor to detect the amount of sweat).

[0078] The evaluation criteria changing means 39 is a means for changing the evaluation criteria for driving evaluation in the driving evaluation means 36 based on the driver's internal state estimated by the driver state estimation means 38. Specifically, if it is estimated that the driver is experiencing negative emotions (e.g., anxiety or stress) while driving in a driving scene to be supported, the evaluation criteria for that driving scene are relaxed, and the evaluation criterion value compared to the driving achievement level is changed to a lower value.

[0079] This means that if a driver feels uncomfortable driving in a particular driving scenario and experiences negative emotions, they will receive a positive evaluation (a compliment) even if their driving performance is slightly lower than usual. The relaxed evaluation criteria are set so that the evaluation criteria are met only if the driving performance reaches a certain level of passability, even if it falls short of the original criteria. This is to ensure that positive evaluations are not arbitrarily given when driving performance is too low.

[0080] To explain in more detail, drivers who have a fear of certain driving scenarios may actually possess sufficient driving ability to navigate those scenarios, but due to the influence of negative emotions (internal system processing loop 3B in the driver's internal model 1), they may not be able to fully utilize their driving ability in those scenarios (resulting in an increased motion system prediction error 2A). On the other hand, for drivers whose driving performance is reduced due to such fear of certain driving scenarios, it is thought that their driving performance can be significantly improved by enhancing their self-efficacy.

[0081] Therefore, when a driver experiences negative emotions while driving in a driving scenario that is subject to driver assistance, the evaluation criteria for driving are relaxed. If the driver meets these relaxed criteria, they receive a positive evaluation (indicating that they are driving at an acceptable level). This effectively enhances the driver's self-efficacy and leads to a cognitive change, making them believe they can drive that scenario well, even if their driving performance was previously reduced due to a lack of confidence or a fear of the driving scenario. As a result, the driver's internal model 1 is effectively improved, and the driver's driving skills improve smoothly.

[0082] For example, in the merging driving scenarios shown in Figures 4 to 6, the goal is to make full use of the acceleration lane 42 that extends to the end point 42A when changing lanes. The degree of driving achievement is calculated from the difference between this target driving route and the actual driving route, and this degree of driving achievement is compared with the evaluation standard value. Therefore, if the driver fails to make full use of the acceleration lane 42 when merging, a positive evaluation will not be provided.

[0083] Conversely, if it is estimated that the driver is experiencing negative emotions while driving, the evaluation threshold value is changed to a lower value. This means that, for example, even if the timing of a lane change is slightly earlier or later than the target, resulting in the lane change point being too far or too close to the end point 42, and thus widening the difference between the target route and the actual route, and thus slightly lowering the driving performance, if a certain level of driving performance is achieved (for example, if the lane change is made before the end point 42), the evaluation will be above the relaxed threshold value, and a relaxed positive evaluation (an evaluation that acknowledges and praises that the driving is at a sufficiently acceptable level) will be provided. Therefore, even drivers who feel insecure about merging will be actively provided with positive post-evaluation support, and the self-efficacy of drivers who lacked confidence will be increased.

[0084] The extent to which the evaluation criteria are changed according to the driver's condition will be set appropriately depending on the content of the driving scene and the driver's condition. For example, the range of change in the evaluation criteria may be adjusted according to the magnitude of the driver's negative emotions (magnitude of fluctuations in physiological state), so that the evaluation criteria are set appropriately according to the magnitude of the impact of negative emotions on driving performance.

[0085] The method for changing the evaluation criteria for driving performance in the evaluation criteria changing means 39 is not limited to changing the evaluation criterion value for driving performance; any method can be adopted as long as it changes the likelihood of receiving a positive evaluation.

[0086] For example, by dividing a single driving scene into multiple driving scenes, and providing a driving evaluation (positive evaluation) at the end of each divided driving scene, it is possible to relax the evaluation criteria for driving evaluation, even if a positive evaluation is not obtained for the entire driving scene, as a positive evaluation may be obtained for one of the divided driving scenes.

[0087] In other words, by dividing the driving scene into smaller segments, the number of tasks (driving operations) that the driver must perform in each segment decreases. Therefore, the likelihood of the driver exceeding the evaluation standard in any one of the driving scenes increases. Consequently, dividing the driving scene into smaller segments is equivalent to loosening the evaluation standards for driving performance.

[0088] For example, in the merging driving scene shown in Figures 4 to 6, the initial driving scene (driving scene from the start of parallel driving to the end of lane change) is divided into the parallel driving section driving scene (driving scene from the start of parallel driving, accelerating and decelerating (Figures 4 and 5), until reaching the lateral position between the vehicles to be entered) and the lane change driving scene (driving scene from the lateral position between the vehicles to turning the steering wheel and completing entry into the gap (Figure 6)).

[0089] As a result, for example, even if the driver's performance in a lane-changing driving scene does not meet the evaluation criteria, if the driver's performance in the parallel driving scene preceding the lane-changing scene met the evaluation criteria, a positive evaluation will be provided after the parallel driving scene ends. This allows the driver to recognize that they drove well in the parallel driving scene, and as a result, their sense of self-efficacy will increase.

[0090] Next, an example of the control procedure for the driver assistance control of the present invention will be explained according to the flowchart in Figure 7. In the driver assistance control, in step S1, the driving scene is recognized. In the following step S2, it is determined whether or not the recognized driving scene is subject to driver assistance, and if it is not subject to driver assistance, the process ends there.

[0091] On the other hand, if it is determined in step S2 that the recognized driving scene is subject to driver assistance, the system proceeds to step S3, where pre-assistance is performed before or during the driving scene.

[0092] In step S4, driving operations and vehicle behavior during the driving scene are detected, and in step S5, physiological state data of the driver during the driving scene is acquired. In the following step S8, it is determined whether the driving scene has ended or not, and if the driving scene has not ended, the process returns to step S4, and the detection of driving operations and vehicle behavior and the acquisition of physiological state data of the driver are continued.

[0093] If it is determined in step S6 that the driving scene has ended, the process proceeds to step S7, where the driver's state during the driving scene is estimated based on the physiological state data acquired during the driving scene. In the following step S8, the evaluation criteria values ​​for the driving evaluation are revised based on the estimated driver state.

[0094] In step S9, the driving performance is calculated based on the driving operations and vehicle behavior detected while driving the driving scene. In the following step S10, the calculated driving performance is compared with an evaluation standard value corrected based on the driver's state, and if the driving performance is not equal to or greater than the evaluation standard value, the process proceeds to step 12.

[0095] On the other hand, if it is determined in step S10 that the driving performance is equal to or above the evaluation standard value, the process proceeds to step S11, where a positive evaluation (praise) is given to the driver, and then to step S12. In step S12, data is updated (e.g., the updated evaluation standard value is stored), and the process cycle is completed.

[0096] While embodiments of the present invention have been described above, the present invention is not limited to the above embodiments, and appropriate modifications can be made within the scope of the claims. For example, in the above embodiments, the evaluation criteria were lowered when the driver experienced negative emotions while driving in a driving scene, but the present invention is not limited to this form. For example, the evaluation criteria could be initially set low, and then raised if the driver's emotional state during driving in a driving scene is not negative. [Industrial applicability]

[0097] This invention can be used for driver assistance in vehicles such as automobiles. [Explanation of Symbols]

[0098] 1. Driver internal model 1A Internal model of the motor system 1B Internal Model of the Body System 2A Motor system prediction error 2B Internal System Prediction Error 3A Motion System Processing Loop 3B Internal System Processing Loop U Controller 21 External information acquisition means 22 Means for acquiring driving information 23. Means for acquiring vehicle information 24 Driver monitoring means 25 Memory means 26 HMI 26A Display 26B Speaker 31. Means for recognizing driving scenes 32 Driving target calculation means 33. Driving operation detection means 34. Vehicle behavior detection means 35. Means for calculating the degree of driving achievement 36. Means for evaluating operation 37 Driving support control means 38 Driver state estimation means 39. Means for changing evaluation criteria 41. My vehicle 42 Acceleration lane 42A End of acceleration lane 43 Lane 44-46 Vehicles running side-by-side 51, 52 car distance

Claims

1. In a driver assistance system that assists the driver of a vehicle, A driving scene recognition means for recognizing the driving scene of the aforementioned vehicle, Driving detection means for detecting the driving operations of the driver and / or the behavior of the vehicle, A driving target calculation means for calculating target driving operations and / or target vehicle behavior that are suitable for the aforementioned driving scene, An interface means for providing information to the driver, A driving achievement calculation means calculates the degree of driving achievement by comparing the driving operations of the driver and / or the behavior of the vehicle detected by the driving detection means with the target driving operations and / or target vehicle behavior calculated by the driving target calculation means. An operation evaluation means for determining whether the operation achievement level calculated by the operation achievement level calculation means meets predetermined evaluation criteria, A driver assistance control means controls the interface means to provide the driver with a positive evaluation of the driver's driving after the end of a driving scene in which the driver assistance is provided, when it is determined that the level of driving performance in a driving scene in which the driver assistance is provided meets the evaluation criteria. Driver information acquisition means capable of detecting the physiological state quantities and / or behavior of the driver, A driver state estimation means that estimates the internal state of the driver based on the physiological state quantities and / or behavior detected by the driver information acquisition means, Based on the internal state of the driver estimated by the driver state estimation means, an evaluation criterion changing means changes the evaluation criteria for the driving evaluation in the driving evaluation means. A driver assistance device equipped with this device.

2. In the driving support device according to claim 1, The aforementioned evaluation criteria changing means is a driver assistance device that relaxes the evaluation criteria for a driving scene if it is estimated that the driver is experiencing negative emotions while driving that driving scene.

3. In the driving support device according to claim 2, A driving assistance device in which the relaxation of the evaluation criteria is achieved by reducing the evaluation criteria value in the evaluation criteria.

4. In the driving support device according to claim 2, The relaxation of the evaluation criteria is achieved by dividing the driving scene subject to driving assistance into multiple driving scenes, and by performing a determination by the driving evaluation means and providing a positive evaluation by the interface means for each divided driving scene.

5. In the driving support device according to claim 1, The aforementioned driving assistance control means controls the interface means to provide the driver with suggestions regarding the driver's driving operations and / or the vehicle's behavior in a driving scene, either before or during the driving scene to which driving assistance is to be provided.

Images