Vehicle control apparatus and vehicle control method

Abstract

This application relates to vehicle control equipment and vehicle control methods. The vehicle control equipment includes: an environment detection unit that detects environmental information about the interior and exterior of a target vehicle; an emotion estimation unit that estimates the driver's emotion based on the environmental information; a driving error behavior prediction unit that predicts the driver's driving error behavior when the estimated emotion is a strong emotion; a driving risk prediction unit that predicts driving risk based on the predicted driving error behavior; and a risk suppression unit that performs predetermined control according to the predicted driving risk.

Description

Technical Field

[0001] This invention relates to a vehicle control device and a vehicle control method. Background Technology

[0002] An information processing device has been proposed that predicts driving risks based on driver biometrics, vehicle behavior, etc., and generates plans to avoid driving risks (see, for example, International Publication No. 2018 / 190152).

[0003] International Publication No. 2018 / 190152 describes estimating the behavior of users of interest (standing still, walking, running, sleeping) based on their acceleration, angular velocity, and air pressure. International Publication No. 2018 / 190152 also describes estimating the emotional state of users of interest based on their biometric data.

[0004] Meanwhile, the specific methods for processing used to estimate the behavior and emotional intensity of interested users, and for processing plans to avoid driving risks based on the behavior and emotional intensity of interested users, as described in International Publication No. 2018 / 190152, are unclear.

[0005] The goal is to improve the accuracy of predicting driving risks using this type of technology.

[0006] To address the aforementioned problems, the purpose of this application is to effectively prevent unexpected situations during driving by improving the accuracy of predicting driving risks. Summary of the Invention

[0007] According to one aspect of this disclosure, a vehicle control device includes: an environment detection unit that detects environmental information about at least one of the interior and exterior of a target vehicle; an emotion estimation unit that estimates a driver's emotion based on the environmental information; a driving error behavior prediction unit that predicts a driving error behavior of the driver when the estimated emotion is a strong emotion; a driving risk prediction unit that predicts a driving risk based on the predicted driving error behavior; and a risk suppression unit that performs predetermined control based on the predicted driving risk.

[0008] According to another aspect of this disclosure, a computer-executed vehicle control method includes: an environment detection step for detecting environmental information about at least one of the interior and exterior of a target vehicle; an emotion estimation step for estimating a driver's emotion based on the environmental information; a driving error behavior prediction step for predicting a driving error behavior of the driver when the estimated emotion is a strong emotion; a driving risk prediction step for predicting a driving risk based on the predicted driving error behavior; and a risk mitigation step for performing predetermined control according to the predicted driving risk.

[0009] According to the present invention, unexpected situations during driving can be effectively prevented by improving the accuracy of predicting driving risks. Attached Figure Description

[0010] Figure 1 This is a diagram showing an overview of the vehicle control device according to this embodiment; Figure 2 This is a diagram illustrating the relationship between strong emotions and driving risks; Figure 3 This is a diagram showing the vehicle control equipment and its peripheral configuration; Figure 4 This is a flowchart illustrating the operation of the vehicle control equipment; Figure 5 It is a diagram used to describe static traffic environment information; Figure 6 It is a diagram used to describe static traffic environment information; Figure 7 It is a diagram used to describe the biological-emotional table; Figure 8 It is a graph used to describe the prediction of driving error behaviors; and Figure 9 It is a graph used to describe the prediction of driving risks.

[0011] List of reference numerals

[0012] 1. Vehicle control equipment (emotion estimation equipment, driving prediction equipment)

[0013] 2. Vehicles

[0014] 3. First sensor

[0015] 4. Second sensor

[0016] 5. Third sensor

[0017] 6. Fourth sensor

[0018] 7. Vehicle drive unit

[0019] 8. Steering Unit

[0020] 9. Braking unit

[0021] 10. Notification Unit

[0022] 21. Processor

[0023] 22. Environmental Monitoring Unit

[0024] 22a. Driving Information Detection Unit

[0025] 22b. Bioinformatics Detection Unit

[0026] 22c. Traffic Environment Monitoring Unit

[0027] 23. Sentiment Estimation Unit

[0028] 23a. First Emotion Estimation Unit

[0029] 23b. Second Emotion Estimation Unit

[0030] 24. Driving Error Behavior Prediction Unit

[0031] 24a. First Driving Error Behavior Prediction Unit

[0032] 24b. Second Driving Error Behavior Detection Unit (Driving Error Behavior Detection Unit)

[0033] 25. Driving Impact Prediction Unit

[0034] 25a. Driving Risk Prediction Unit

[0035] 26. Risk Suppression Unit

[0036] 26a. Control Content Determination Unit

[0037] 26b. Vehicle control unit

[0038] 26c. Notification control unit

[0039] 31. Memory

[0040] 32. Procedure

[0041] 33. Map data

[0042] 34. Statistical Database

[0043] 35. Location database Detailed Implementation

[0044] 1. Overview of Vehicle Control Equipment 1

[0045] Figure 1 This is a diagram showing an overview of the vehicle control device 1 according to this embodiment.

[0046] like Figure 1 As shown, the vehicle control device 1 is a device having the function of performing the following processing: sensing processing S1, detecting information about the target vehicle ( Figure 3 The environmental information D1 of at least one of the interior and exterior of the vehicle 2) shown; the driver's emotion estimation processing S2 based on the environmental information D1; the driving error behavior prediction processing S3 based on the estimated emotion, etc.; the driving risk prediction processing S4 based on the driving error behavior; and the driving risk suppression processing S5.

[0047] The vehicle control device 1 according to this embodiment is an in-vehicle device installed in the target vehicle, but is not limited to this configuration, and can also be a device that remotely controls the target vehicle by communicating with it. The target vehicle is not particularly limited, but a four-wheeled vehicle is described in this embodiment as an example.

[0048] 1.1 Sensing Processing S1

[0049] The processing of detecting environmental information D1 about the interior and exterior of the target vehicle is applied as sensing processing S1. Environmental information D1 is environmental information that affects the driver's driving, emotions, etc. In this embodiment, the processing of detecting driving information D1a about the driver, biological information D1b about the driver, and traffic environment information D1c that allows identification of the traffic environment composed of the surrounding environment is applied as the processing of environmental information D1 about the interior and exterior of the target vehicle.

[0050] Driving information D1a is information that allows for the identification of driver errors. Specifically, driving information D1a is information related to driving operations such as accelerator operation, braking operation, steering wheel operation, etc., and corresponds to information that directly or indirectly detects each driving operation.

[0051] Examples of information obtained by directly detecting each driving operation include accelerator position, braking input, and steering wheel input. Examples of information obtained by indirectly detecting each driving operation include vehicle dynamics information obtained by detecting movement of the target vehicle in front, behind, to the left, to the right, etc.

[0052] Biometric information D1b refers to biological information that influences a driver's driving errors, emotions, etc. Specifically, biometric information D1b corresponds to information that allows the recognition of a driver's facial expressions, voice, and posture, or to physiological signals that allow the recognition of a driver's pulse, blood pressure, etc.

[0053] Traffic environment information (D1c) allows for the identification of traffic environment information that influences driver errors, emotions, and other factors. Specifically, D1c corresponds to information such as the area where the target vehicle is located, road conditions, how the driver grips the steering wheel, time information indicating the time of day (e.g., day or night), in-vehicle environment information indicating the presence of passengers, and weather information indicating the weather. Road conditions include whether there are road mergings, whether there are construction zones, the condition of road infrastructure, traffic control status, junction relationships, and gradient.

[0054] 1.2 Sentiment Estimation Processing S2

[0055] The process of estimating emotions, including the driver's strong emotions D2, using biometric information D1b and traffic environment information D1c is applied as emotion estimation process S2. Strong emotions D2 indicate strong emotions that significantly influence a person's thoughts, behaviors, etc., and correspond to, for example... Figure 2 The words shown are “anger,” “extreme joy,” “surprise,” and “excitement.”

[0056] Figure 2 This is a diagram that schematically illustrates the relationship between strong emotions (D2) and driving risks.

[0057] Driving risk indicates the likelihood of situations that should be avoided while driving. Examples of situations to be avoided include malfunctions that should be avoided while driving, and in this embodiment, collision events and near-collision events are each set as situations to be avoided.

[0058] Compared to when drivers are in a normal state, strong emotions (D2) tend to significantly influence driving errors, leading to increased driving risks. Compared to other emotions, certain emotions, particularly "anger" and "extreme joy," may further increase driving risks.

[0059] 1.3 Driving Error Behavior Prediction and Processing S3

[0060] The process of predicting future driving errors by using estimated strong emotions D2, etc., is applied as driving error prediction processing S3. In this embodiment, in addition to the process of predicting future driving errors D3a and D3b based on estimated strong emotions D2, driving error prediction processing S3 also includes the process of detecting real-time or past driving errors D3c included in driving information D1a.

[0061] Note that in driving errors D3a and D3b, driving error D3a indicates a driving error caused by the driver's intentional actions. Driving error D3b indicates a driving error caused by the driver's unintentional actions.

[0062] 1.4 Driving Risk Prediction and Processing S4

[0063] The process of predicting driving risks based on predicted driving errors is applied as driving risk prediction processing S4. In this embodiment, the probability D4 of the occurrence of collision events and near-collision events is predicted as driving risks.

[0064] 1.5 Driving Risk Suppression Measures S5

[0065] The control used to suppress the predicted driving risk (probability of occurrence D4) is applied as driving risk suppression processing S5. In this embodiment, driving risk can be reduced and unforeseen situations during driving can be effectively prevented by performing predetermined control based on the predicted driving risk.

[0066] 2. Peripheral configuration of vehicle control equipment 1

[0067] Reference Figure 3 Describe the peripheral configuration of vehicle control device 1.

[0068] Vehicle 2 includes a first sensor 3, a second sensor 4, a third sensor 5, a fourth sensor 6, a vehicle drive unit 7, a steering unit 8, a braking unit 9, and a notification unit 10, each of which is connected to vehicle control equipment 1.

[0069] The first sensor 3 is one or more sensors that detect information related to driving information D1a, and specifically corresponds to accelerometer sensor, brake sensor, steering wheel sensor, acceleration sensor, gyroscope sensor, etc.

[0070] The second sensor 4 is one or more sensors that detect information related to biometric information D1b, and specifically corresponds to an in-vehicle camera that captures images including the driver's facial expressions and postures, a voice input unit that acquires speech, etc.

[0071] The third sensor 5 is one or more sensors that detect information related to traffic environment information D1c, and specifically corresponds to an external camera that captures images of the surrounding environment including road conditions, a distance sensor that detects the presence of other vehicles in the surrounding environment and the distance to other vehicles, a position detection sensor such as a GPS sensor, a clock circuit, a rain sensor, an onboard camera or weight sensor that detects the presence of passengers, etc.

[0072] The third sensor 5 may include a communication unit capable of wireless communication with the third sensor 5. The communication unit can obtain a portion of the traffic environment information D1c from an information providing server via a communication network such as the Internet. In this case, detailed road conditions, weather information, etc., can be easily obtained.

[0073] The fourth sensor 6 is a driver assistance sensor used to implement advanced driver assistance systems (ADAS). For example, the fourth sensor 6 corresponds to a position detection sensor, such as a radar sensor that detects obstacles and other vehicles, a camera that performs object recognition and lane recognition, a light detection and ranging (LiDAR) sensor, or a GPS sensor, thereby enabling functions such as automatic braking, lane maintenance support, and collision warning.

[0074] The vehicle drive unit 7 includes a drive source for the vehicle 2 and an energy supply source that supplies drive energy to the drive source. The vehicle drive unit 7 drives the drive source according to the driver's driving operation. The drive source includes at least one of, for example, a drive motor and an internal combustion engine.

[0075] The steering unit 8 includes a steering mechanism and a drive source. The steering mechanism includes the steering column and steering box of the vehicle 2, and the drive source is such as an electric motor for steering. The steering unit 8 activates the steering mechanism according to the driver's driving operation.

[0076] The braking unit 9 includes the braking device of the vehicle 2 and a drive source such as an electric motor for braking. The braking unit 9 activates the braking device according to the driver's driving operation.

[0077] The notification unit 10 includes a display and a voice output unit. Under the control of the vehicle control device 1, the display shows various information, and the voice output unit outputs various voice commands within the vehicle. This allows various types of information to be notified to the driver, etc. For example, the display can show a map of the road, a navigation image guiding the driver along a route from the current location to the destination, etc. The voice output unit can output voice commands guiding the driver along the route.

[0078] Any standard configuration capable of notifying the driver can be widely used in the notification unit 10.

[0079] 2.1 Configuration of Vehicle Control Equipment 1

[0080] The vehicle control device 1 includes a processor 21 and a memory 31.

[0081] The processor 21 serves as a computer for controlling the vehicle control device 1 and the vehicle 2. The processor 21 also serves as an environment detection unit 22, an emotion estimation unit 23, a driving error behavior prediction unit 24, a driving impact degree prediction unit 25, and a risk suppression unit 26 by loading and executing the program 32 recorded on the memory 31.

[0082] The environmental detection unit 22 includes: a driving information detection unit 22a, which detects driving information D1a based on information from the first sensor 3; a biological information detection unit 22b, which detects biological information D1b based on information from the second sensor 4; and a traffic environment detection unit 22c, which detects traffic environment information D1c based on information from the third sensor 5. The environmental detection unit 22 is responsible for performing... Figure 1 An example of the processing unit “Sensing Processing S1” in the example.

[0083] The emotion estimation unit 23 includes a first emotion estimation unit 23a that estimates strong emotions based on biological information D1b and a second emotion estimation unit 23b that estimates strong emotions based on traffic environment information D1c.

[0084] Emotion estimation unit 23 is the execution Figure 1 An example of the processing unit "emotion estimation processing S2" in the diagram. In this embodiment, a strong emotion estimated by the first emotion estimation unit 23a is represented by reference numeral D2a, and a strong emotion estimated by the second emotion estimation unit 23b is represented by reference numeral D2b. Strong emotions D2a and D2b are each... Figure 1 The example shown is a strong emotion D2.

[0085] The driving error behavior prediction unit 24 includes a first driving error behavior prediction unit 24a and a second driving error behavior detection unit 24b. The first driving error behavior prediction unit 24a predicts driving error behavior D3 by using a statistical database 34 stored in the memory 31 when a strong emotion D2 has been estimated. The second driving error behavior detection unit 24b detects driving error behavior D3c by using driving information D1a.

[0086] Driving error behavior prediction unit 24 is the execution Figure 1 An example of the processing unit "Driving Error Behavior Prediction Processing S3" in this embodiment. In this embodiment, based on the driving error behavior D3 predicted by the first driving error behavior prediction unit 24a, it is determined whether it includes... Figure 1 The driving errors D3a and D3b are included. However, it can also be determined whether driving errors D3a and D3b are included based on driving error D3c detected by the second driving error detection unit 24b.

[0087] The driving impact prediction unit 25, based on the predicted driving error behaviors D3a and D3b, performs processing to predict the degree of impact on future driving. The driving impact prediction unit 25 includes a driving risk prediction unit 25a that predicts the driving risk (occurrence probability D4) as a degree of impact on future driving. The driving impact prediction unit 25 is a process that executes... Figure 1An example of the processing unit "Driving Risk Prediction Processing S4" in the example.

[0088] The risk mitigation unit 26 includes a control content determination unit 26a that determines the control content of the vehicle 2 based on the prediction results of the driving impact degree prediction unit 25, a vehicle control unit 26b that executes the control corresponding to the determined control content, and a notification control unit 26c that executes the notification corresponding to the determined control content.

[0089] Risk suppression unit 26 is the execution Figure 1 An example of the processing unit "Driving Risk Suppression Processing S5" in the example.

[0090] In addition to program 32, memory 31 also stores map data 33, a statistical database 34, and a location database 35. Map data 33 is map data used for navigation and includes, for example, facility data consisting of nodes and links indicating road shapes, as well as point of interest (POI) data stating the names, locations, etc., of facilities that may be destination candidates. Map data 33 may be information appropriately obtained via a communication network.

[0091] The statistical database 34 includes various types of statistical data used by the emotion estimation unit 23, the driving error behavior prediction unit 24, and the driving impact degree prediction unit 25. Specifically, the statistical database 34 includes an environment-emotion table T1 and a bio-emotion table consisting of an expression-emotion table T2a and a posture-emotion table T2b, driving error behavior statistics T3 indicating driving error behaviors involving human error when experiencing specific emotions, driving risk statistics T4 indicating the probability of occurrence of driving error behaviors that cause collision events and near-collision events, etc., each of which will be described below.

[0092] The location database 35 stores information such as the driving position when the emotion estimation unit 23 estimates a specific strong emotion D2. The driving position is location information detected by a location detection sensor (such as a GPS sensor included in the third sensor 5). The vehicle control device 1 (e.g., the emotion estimation unit 23) records this driving position.

[0093] 3. Operation of vehicle control equipment 1

[0094] According to Figure 4 The flowchart shown describes the operation of vehicle control device 1.

[0095] Vehicle control device 1 repeatedly executes the following commands while vehicle 2 is in motion: Figure 4 The processing shown.

[0096] First, the vehicle control device 1 acquires traffic environment information D1c via the traffic environment detection unit 22c (step S1a), and the second emotion estimation unit 23b estimates the emotion including strong emotion D2b based on the acquired traffic environment information D1c (step S2a).

[0097] In this embodiment, in step S1a, "static traffic environment information" and "dynamic traffic environment information" are obtained as traffic environment information D1c.

[0098] "Static traffic environment information" includes information that allows identification of the area where vehicle 2 is located, road conditions, how the driver is gripping the steering wheel, time information indicating whether it is day or night, in-vehicle environment information indicating the presence of passengers, and weather information indicating the weather. "Dynamic traffic environment information" is described below.

[0099] 3.1 Estimation of strong emotions based on static traffic environment information

[0100] Reference Figure 5 and Figure 6 Describes a method for estimating strong emotions D2b based on "static traffic environment information".

[0101] Figure 5 The diagram shows ratios indicating the degree of intense emotion D2b for each condition included in the "Static Traffic Environment Information". These ratios are examples of indices indicating how much higher the probability of a driver experiencing intense emotion is compared to a baseline (ideal condition). The degree of intense emotion D2b can also be interpreted as an index of the probability of occurrence D4.

[0102] like Figure 6 As shown, the ratio indicating the strong emotion D2b corresponding to the static traffic environment information can be calculated by calculating the combined value of the ratios of multiple conditions included in the static traffic environment information. Figure 6 The 'p' symbol in the attached figure indicates the probability of a strong emotion occurring at the baseline.

[0103] exist Figure 6 In the study, the intensity of strong emotions D2b increased in the order of baseline, traffic environment 1, traffic environment 2, and traffic environment 3, and the probability of drivers experiencing strong emotions increased.

[0104] In this embodiment, the second emotion estimation unit 23b is based on... Figure 5 The ratio of traffic environment and strong emotion D2b shown is associated with each other in the environment-emotion table T1. The ratio of strong emotion D2b for each environment is obtained, and the strong emotion D2b corresponding to the static traffic environment information obtained in step S1a is calculated by using the calculation formula for combining each ratio.

[0105] 3.2 Dynamic Traffic Environment Information

[0106] "Dynamic traffic environment information" is categorized into "overall triggers," which are driving-related triggers, and "accidental triggers," which are driving-independent triggers. "Overall triggers" are information primarily related to actions outside of one's own vehicle. Specifically, this information corresponds to situations such as a vehicle ahead driving slowly, a vehicle behind cutting in, reckless driving, or situations where one must apply the brakes when an oncoming vehicle is about to turn left in front of them.

[0107] Note that, as an estimation method for strong emotion D2b based on information related to the behavior outside of one's own vehicle, the ratio of strong emotion D2b can be calculated using statistical values ​​such as speed, acceleration, relative speed to the vehicle in front, and the average or maximum distance between one's own vehicle and the vehicle.

[0108] "Chance triggers" are information primarily related to behaviors other than those of the driver inside the vehicle. Specifically, this information corresponds to conversations with passengers, telephone use, and in-car music.

[0109] In this embodiment, the second emotion estimation unit 23b multiplies the probability D4 of the occurrence of strong emotion D2b calculated using static traffic environment information by a predetermined coefficient (1 or higher) set based on at least one of "overall triggering" and "accidental triggering," or calculates strong emotion D2b by considering dynamic traffic environment information by summing the ratios of strong emotion D2b calculated using dynamic traffic environment information. This improves the accuracy of strong emotion estimation compared to when dynamic traffic environment information is not considered.

[0110] Note that the coefficient for "overall trigger" is set to be greater than that for "accidental trigger," making "overall trigger" more likely to be identified as a strong emotion.

[0111] However, this disclosure is not limited to estimating strong emotions by using both "static traffic environment information" and "dynamic traffic environment information," but can also be used to estimate strong emotions by using only static traffic environment information.

[0112] 3.3 Notification and Control Based on Strong Emotions

[0113] Return to Figure 4 The vehicle control device 1 determines the degree of strong emotion D2b estimated based on traffic environment information D1c via the second emotion estimation unit 23b (step S3a).

[0114] In this embodiment, it is determined whether the degree of strong emotion D2b is "high". In this case, the degree of strong emotion D2b can be determined in multiple levels (such as high, medium and low), and it can be determined whether the degree of strong emotion D2b corresponds to "high". When the degree of strong emotion D2b is equal to or higher than a first predetermined value, the degree of strong emotion D2b can be determined to be "high". When the degree of strong emotion D2b is "high" (step S3a: yes), the vehicle control device 1 performs traffic environment-related notification processing via the notification control unit 26c (step S4a). The first predetermined value can be set to an appropriate value that can determine the degree of strong emotion D2b as "high", in other words, an appropriate value that can appropriately perform traffic environment-related notification processing.

[0115] In the notification processing of step S4a, a notification is executed to draw attention to the traffic environment. For example, the driver is notified by the notification unit 10 of content such as "Be aware of heavy traffic and the possibility of vehicles cutting in from the side".

[0116] This process is an example of the process of "executing a predetermined notification without using the prediction results of the driving error behavior prediction unit" according to this disclosure. After the notification process is completed, the process proceeds to step S14a.

[0117] On the other hand, when the degree of strong emotion D2b estimated in step S3a is not "high" (step S3a: no), the vehicle control device 1 obtains biological information D1b via the biological information detection unit 22b (step S5a), and estimates strong emotion D2a based on biological information D1b via the first emotion estimation unit 23a (step S6a).

[0118] In step S6a, the driver's facial expressions included in the biometric information D1b are analyzed to determine whether the expression corresponds to a strong emotion. For this determination, the expression-emotion table T2a in the statistical database 34 stored in memory 31 is referenced. The expression-emotion table T2a includes data in which expressions of strong emotions (e.g., "anger") and their probabilities of occurrence are correlated, and... Figure 7 As shown in the image.

[0119] For example, when “angry face” and “shouting” are detected as expressions, the value obtained by combining each probability is calculated (e.g., 0.9 + 0.4 = 1.3).

[0120] Similarly, the values ​​obtained by combining each probability are calculated using the "Face-Emotion Table T2a," which corresponds to another strong emotion (e.g., "extremely happy" or "surprised"). For example, when the combined value is calculated to be 1.3 for "anger," 0.95 for "extremely happy," and 0.9 for "surprised," "anger" is determined to be dominant, with the highest value. These combined values ​​each correspond to "emotional intensity (emotional level)."

[0121] In step S6a, the driver's posture included in the biometric information D1b is analyzed similarly. When determining whether a posture corresponds to a strong emotion, the "Posture-Emotion Table T2b" is referenced. The "Posture-Emotion Table T2b" includes data in which the probabilities of posture and the occurrence of strong emotions are correlated, and... Figure 7 As shown in the image.

[0122] For example, when a "rude gesture" is detected, from Figure 7 The probability of identifying "anger" in the "Posture-Emotion Table T2b" shown is 0.25.

[0123] The probability value is obtained based on posture and combined with the probability value based on facial expression. For example, when the probability of "anger" is calculated as 1.3 based on facial expression, adding 0.25 based on posture results in a final value of 1.55. Figure 7 As shown. Similarly, by adding 0.2 to 0.95, the final value for "extremely happy" is 1.15, and by adding 0.5 to 0.9, the final value for "surprised" is 1.4. Therefore, the level of "angry" is determined to be the highest.

[0124] Based on the corrected probability values, facial expressions and posture are taken into account to determine the most dominant strong emotion. In this embodiment, accuracy can be improved by applying correction coefficients as needed.

[0125] Strong emotions can also be categorized based on the driver's facial expressions and the intensity of those emotions can be assessed based on the driver's posture. For example, when the probability of "anger" is identified as 1.3 based on facial expression, a correction value (e.g., 0.25) is considered due to posture, and the emotion level is determined to be "moderate." A method could be considered that sets "low" (if the probability of posture is less than 0.1), "moderate" (if the probability is between 0.1 and 0.3), and "high" (if the probability is equal to or greater than 0.3) as the criteria for this determination, but these are freely chosen criteria.

[0126] Return to Figure 4The vehicle control device 1 determines whether the level of the intense emotion D2a estimated in step S6a is "low" (step S7a). When the determination result indicates that the level of the intense emotion D2a is "high" or "medium" (step S7a: No), the control unit 26c is notified to execute notification control for encouraging emotional well-being (step S8a). On the other hand, when the level of the intense emotion D2a is "low" (step S7a: Yes). Figure 4 The process shown in the flowchart has ended.

[0127] Note that step S7a is not limited to determining whether the degree of strong emotion D2a is "low" from multiple levels such as high, medium, and low. It can also determine whether the degree of strong emotion D2b is "low" when the degree of strong emotion D2b is equal to or lower than a second predetermined value, or by considering the ratio of strong emotion D2b to determine whether the degree of strong emotion D2a is "low". For example, the determination in step S7a can be performed based on the result of the degree of strong emotion D2a + β (correction coefficient) × the degree of strong emotion D2b. The second predetermined value can be set as an appropriate value that determines the degree of strong emotion D2b as "low," in other words, an appropriate value for determining whether to perform the notification control in step S8a.

[0128] In step S8a, notification control is performed to reduce the intensity of strong emotions or guide the driver to a normal emotional state. Specifically, notification control corresponds to controls for outputting sounds or images that allow the driver to relax, calm the driver, or distract the driver. These sounds include dialogue, music, etc.

[0129] Subsequently, the vehicle control device 1 acquires biometric information D1b again via the biometric information detection unit 22b (step S9a). Next, the vehicle control device 1 estimates a strong emotion D2a based on the acquired biometric information D1b via the first emotion estimation unit 23a (step S10a). The processing in step S10a is similar to that in step S6a, and at this point, the most prominent strong emotion is identified.

[0130] Next, the vehicle control device 1 determines whether the strong emotion estimated by the first emotion estimation unit 23a is a specific strong emotion with relatively high driving risk (step S11a). Specific strong emotions with high driving risk include "anger," "extreme happiness," etc. Figure 2 As shown.

[0131] When it is determined that the estimated intense emotion is a specific intense emotion that poses a high driving risk (step S11a: Yes), the vehicle control device 1 performs alarm processing via the notification control unit 26c (step S12a). In the alarm processing, processing is performed to reduce the intensity of the intense emotion or guide the driver to a normal emotional state.

[0132] In this embodiment, alarm processing is performed based on the category and intensity of the strong emotion. Specifically, when the specific strong emotion is "anger" and the intensity of the strong emotion is "high," a sound or image is output at a low volume or using language that the driver can easily associate with. When the specific strong emotion is "anger," a clear sound or image is output when the intensity of the strong emotion is "moderate," and a weakened sound or image is output when the intensity of the strong emotion is "low."

[0133] After the alarm processing in step S12a is performed, or when it is determined in step S11a that the estimated strong emotion is not a specific strong emotion (step S11a: no), the vehicle control device 1 acquires driving information D1a and traffic environment information D1c (step S13a), and predicts future driving error behavior via the second driving error behavior detection unit 24b and the first driving error behavior prediction unit 24a (step S14a).

[0134] The detection of driving errors by the second driving error detection unit 24b involves using driving information D1a to identify real-time or past driving error behavior D3c. Since driving information D1a includes information indicating the driver's real-time or past driving conditions, the second driving error detection unit 24b uses the real-time or past driving conditions to detect the driver's real-time or past driving error behavior D3c, and identifies the identified driving error behavior D3c as a future driving error behavior. Specifically, the second driving error detection unit 24b detects driving error behaviors D3a and D3b involving human error during periods of strong emotion (see [link to relevant documentation]). Figure 8 ).

[0135] Reference Figure 8 The method for predicting driving error behavior D3 by the first driving error behavior prediction unit 24a is described.

[0136] Figure 8 This shows the probability of driving errors involving human error occurring when experiencing strong emotions such as "anger". Driving errors involving human error can be classified into driving errors D3a caused by the driver's intentional actions and driving errors D3b caused by the driver's unintentional actions.

[0137] Driving errors D3a caused by the driver's intentional actions specifically correspond to aggressive driving, driving in narrow spaces between vehicles, etc. Driving errors D3b caused by the driver's unintentional actions specifically correspond to sudden braking, ignoring traffic lights, etc.

[0138] In this embodiment, the statistical database 34 includes data, referred to as "driving error behavior statistics T3," which allows for probabilistic identification of driving error behaviors D3a and D3b involving human error for each type of strong emotion. "Driving error behavior statistics T3" is based on... Figure 8 The statistics shown are created from data such as driving error statistics. For example, driving error behavior statistics T3 is data in which driving error behaviors D3a and D3b involving human error, and the probability of occurrence of driving error behaviors D3a and D3b, are correlated with each other for each strong emotion.

[0139] By using "driving error behavior statistics T3", driving error behaviors D3a and D3b, which have a relatively high probability of occurrence, can be predicted as future driving error behaviors based on strong emotions.

[0140] Furthermore, by making the "driving error statistics T3" into data that allows identification of whether future driving errors are caused by the driver's intentional actions (D3a) or by the driver's unintentional actions (D3b), it is easy to identify whether future driving errors are caused by the driver's intentional actions (D3a) or by the driver's unintentional actions (D3b).

[0141] return Figure 4 The vehicle control device 1 predicts driving risk (probability of occurrence D4) based on the predicted driving error behavior via the driving risk prediction unit 25a (step S15a).

[0142] Reference Figure 9 Describe an example of a method for predicting driving risks.

[0143] Figure 9 The driving risk (probability D4) is shown for driving errors that lead to avoidable situations (collision events and near-collision events) when experiencing strong emotions such as "anger". Figure 9 The probability of occurrence, D4, is shown as a ratio.

[0144] In this embodiment, the statistical database 34 includes data on driving error behaviors D3a and D3b involving human error and driving risk (probability of occurrence D4) that are correlated with each strong emotion as "driving risk statistics T4".

[0145] "Driving Risk Statistics T4" is based on Figure 9 The statistics shown are created from data such as driving risk statistics. For example, driving risk statistics T4 are data in which driving errors involving human error D3a and D3b and driving risk (probability of occurrence D4) are correlated with each strong emotion.

[0146] By using “Driving Risk Statistics T4”, driving risks (probability of occurrence D4) associated with predicted driving errors D3a and D3b can be easily identified based on strong emotions.

[0147] Return to Figure 4 The vehicle control device 1 determines whether the predicted driving risk is higher than a predetermined threshold via the driving risk prediction unit 25a (step S16a). When the driving risk is high (step S16a: yes), the risk suppression unit 26 performs control based on the driving risk (step S17a).

[0148] Specifically, in step S17a, the control content determination unit 26a identifies whether the high-risk driving error is caused by the driver's intentional behavior (D3a) or by the driver's unintentional behavior (D3b), and determines the control content based on the identified driving error. In the case of driving error D3a, the notification control unit 26c performs a warning process to suppress the judgment error by informing the driver of the predicted future fault and the resulting damage. For example, in the case of aggressive driving, a notification such as "Aggressive driving may violate criminal law. Please avoid aggressive driving" can be provided.

[0149] In the event of driving error D3b, ADAS control to avoid predicted future malfunctions is executed via vehicle control unit 26b. Specifically, vehicle 2 is controlled to adjust the vehicle speed to an appropriate range, maintain distance between vehicles through braking control, and prevent lane departure through lane assist support, etc.

[0150] Note that the data structures of the Environment-Emotion Table T1, Driving Risk Statistics Table T4, Facial Expression-Emotion Table T2a, Posture-Emotion Table T2b, Driving Error Behavior Statistics Table T3, and Driving Risk Statistics Table T4 can be appropriately modified.

[0151] The above is based on Figure 3 The flowchart shown illustrates the operation of vehicle control device 1.

[0152] Note that the steps S4a, S8a, S12a, and S17a, which are processes performed by the risk suppression unit 26, are examples of the "predetermined control" according to this disclosure. The content of this process may be modified as appropriate.

[0153] For example, the risk suppression unit 26 can perform route guidance processing via the vehicle control unit 26b to guide the driver along an appropriate route using navigation technology, or state recovery processing to restore the driver's state. When performing route guidance processing in step S4a, it is possible to consider performing route guidance processing in step S4a to select a traffic environment suitable for reducing the intensity of strong emotions D2b, performing route guidance processing suitable for restoring the emotional state in step S8a, and performing route guidance processing suitable for suppressing driving errors with high driving risks in step S17a.

[0154] State recovery processing involves using verbal communication, smell, music, etc., to maintain or improve the driver's attention.

[0155] In the event of a specific strong emotion with relatively high driving risk (e.g., a specific strong emotion predicting that driving error behavior has a relatively high driving risk), the location of vehicle 2 is stored in the location database 35, and when the driving risk prediction unit 25a predicts that the driver will pass through the location stored in the location database 35 based on information about the driving route already set by the driver, the risk suppression unit 26 performs route guidance processing to guide the driver along a route that avoids the stored location.

[0156] In this embodiment, the processing of the environmental detection unit 22 is an example of the environmental detection steps of this disclosure. The processing of the emotion estimation unit 23 is an example of the emotion estimation steps of this disclosure. The processing of the driving error behavior prediction unit 24 is an example of the driving error behavior prediction steps of this disclosure. The processing of the driving risk prediction unit 25a is an example of the driving risk prediction steps of this disclosure. The processing of the risk suppression unit 26 is an example of the risk suppression steps of this disclosure. The processing of the control content determination unit 26a is an example of the control content determination steps of this disclosure. The processing of the vehicle control unit 26b is an example of the vehicle control steps of this disclosure. The processing of the notification control unit 26c is an example of the notification control steps of this disclosure.

[0157] 4. Operation and Effects

[0158] 4.1 Overall Algorithm

[0159] As described above, the vehicle control device 1 according to this embodiment includes: an environment detection unit 22 that detects environmental information D1 about the interior and exterior of the target vehicle; an emotion estimation unit 23 that estimates the driver's emotion D2 based on the environmental information D1; a driving error behavior prediction unit 24 that predicts the driver's driving error behaviors D3a and D3b when the estimated emotion D2 is a strong emotion; a driving risk prediction unit 25a that predicts driving risk based on the predicted driving error behaviors D3a and D3b; and a risk suppression unit 26 that performs predetermined control according to the predicted driving risk.

[0160] With this configuration, driving risks can be predicted with high accuracy by anticipating driving errors when the driver experiences strong emotions. This high accuracy allows for appropriate control based on driving risks. Therefore, improving the accuracy of driving risk prediction can effectively prevent unforeseen situations during driving.

[0161] Note that the case of detecting environmental information D1 about the interior and exterior of the target vehicle and detecting the driver's emotion D2 based on the detected environmental information D1 has been described. However, it is possible to detect environmental information D1 about at least one of the interior and exterior of the target vehicle and to estimate the driver's emotion D2 based on the detected environmental information D1.

[0162] The environmental detection unit 22 includes: a bio-information detection unit 22b, which detects the driver's bio-information D1b as environmental information D1; and a traffic environment detection unit 22c, which detects the target vehicle's traffic environment information D1c as environmental information D1. The emotion estimation unit 23 estimates a strong emotion based on each of the driver's bio-information D1b and the traffic environment information D1c.

[0163] This configuration can improve the accuracy of estimating strong emotions, which in turn improves the accuracy of predicting driving risks and helps to provide more appropriate control and suppress the comfort of over-control.

[0164] Environmental information D1 includes driving information D1a of the driver in the target vehicle. The driving error behavior prediction unit 24 predicts the driver's driving error behavior based on real-time driving conditions, including driving error behavior statistics T3 when experiencing strong emotions and driving error behavior D3c detected using driving information D1a.

[0165] This configuration can improve the accuracy of predicting driving errors, which in turn improves the accuracy of predicting driving risks and helps to provide more appropriate control and suppress the comfort of over-control.

[0166] The driving risk prediction unit 25a predicts driving risk based on driving risk statistics T4, which allows for the identification of the relationship between driving errors and situations to be avoided (collision events / near-collision events). With this configuration, the accuracy of predicting driving risks leading to situations to be avoided can be improved. Note that driving risk statistics T4 is an example of statistics that allows for the identification of the relationship between driving errors and situations to be avoided. The situations to be avoided can be appropriately modified.

[0167] The risk mitigation unit 26 executes at least one of the following as predetermined controls based on driving risk: vehicle speed control, braking control, route guidance processing, state recovery processing, and warning processing. Driving risk can be directly reduced by executing vehicle speed control and braking control, and indirectly reduced by route guidance processing, state recovery processing, and warning processing. This allows for appropriate control of the target vehicle based on predicted driving risk, thereby reducing driving risk.

[0168] In this embodiment, driving risk can be effectively reduced by implementing a predetermined control that suppresses driving errors with high driving risk among those predicted by the driving error prediction unit 24.

[0169] The vehicle control device 1 includes a memory 31 that stores the location of the target vehicle when the estimated emotion is a specific strong emotion. When the driving risk prediction unit 25a predicts that the target vehicle will pass through the location stored in the memory 31, the risk suppression unit 26 performs route guidance processing to guide the driver along a route used to avoid the stored location. With this configuration, the driver can be guided along a route to avoid locations where the driver has previously experienced strong emotions, thereby preventing situations where driving risks increase. Note that the memory 31 corresponds to the storage unit storing the location of the target vehicle, and any recording medium can be used.

[0170] In the case of driving error behaviors D3a and D3b predicted by the driving error behavior prediction unit 24, the risk suppression unit 26 performs a notification process to notify the driver as a predetermined control in the case of driving error behavior D3a caused by the driver's intentional behavior, and performs driving control of the target vehicle as a predetermined control in the case of driving error behavior D3b caused by the driver's unintentional behavior.

[0171] With this configuration, control can be modified based on whether the predicted driving error is an intentional driving error by the driver, thus enabling appropriate control based on the driving error.

[0172] 4.2 Sentiment Estimation

[0173] In this embodiment, the vehicle control device 1 includes: a biometric detection unit 22b that detects the biometric information D1b of the driver of the target vehicle; an environmental detection unit 22 that detects environmental information D1 in the area surrounding the target vehicle; and an emotion estimation unit 23 that estimates the driver's strong emotions based on the biometric information D1b and the environmental information D1. According to this configuration, an emotion estimation device can be provided that helps to quickly and appropriately estimate the emotions affecting the driver's erroneous driving behavior. The emotion estimation device may only have the emotion estimation-related functions of the vehicle control device 1, or it may combine these functions with functions not related to emotion estimation.

[0174] The emotion estimation unit 23 estimates strong emotions using environmental information D1 based on the environment-emotion table T1, which indicates the relationship between the environment around the vehicle and strong emotions. According to this configuration, strong emotions can be effectively identified from the environment around the vehicle.

[0175] Environmental information D1 includes at least one of road conditions, traffic density, time of day, region, and the presence or absence of passengers. Based on this configuration, strong emotions affected by at least one of road conditions, traffic density, time of day, region, and the presence or absence of passengers can be estimated.

[0176] The emotion estimation unit 23 estimates strong emotions using the driver's biological information D1b, based on bio-emotion tables (facial expression-emotion table T2a and posture-emotion table T2b) that indicate the relationship between the driver's biological information and strong emotions. According to this configuration, strong emotions can be effectively identified using the biological information D1b.

[0177] Biometric D1b includes information that allows for the identification of each of the driver's facial expressions and postures. Based on this configuration, strong emotions can be effectively identified from mood and posture.

[0178] The emotion estimation unit 23 estimates strong emotions by using driving risk statistics T4, which is a statistical data set that allows identification of the relationship between the surrounding environment of the target vehicle and strong emotions, and driving error behaviors detected from the driver's driving information D1a in the target vehicle. This configuration improves the accuracy of strong emotion estimation.

[0179] 4.3 Driving Prediction

[0180] In this embodiment, the vehicle control device 1 includes: a biometric detection unit 22b that detects the biometric information D1b of the driver of the target vehicle; a traffic environment detection unit 22c that detects traffic environment information D1c, which serves as environmental information about the surrounding area of ​​the target vehicle; an emotion estimation unit 23 that estimates the driver's strong emotions based on the driver's biometric information D1b and the traffic environment information D1c; a second driving error behavior detection unit 24b that detects the driver's driving error behavior D3c; and a driving impact degree prediction unit 25 that predicts the degree of impact on future driving error behaviors based on the estimation result of the driver's strong emotions and the detection result of the driving error behavior D3c.

[0181] According to this configuration, rapid and appropriate estimation of strong emotions influencing driving errors is facilitated by using biometric information D1b and traffic environment information D1c. By considering the relationship between the estimation results of strong emotions and the driver's driving errors, a driving prediction device can be provided that can predict the degree of influence on future driving errors with high accuracy. Note that the driving prediction device can also be referred to as a driving influence degree prediction device.

[0182] Note that in this embodiment, the driving impact prediction unit 25 has been shown to predict the driving risk as the degree of impact on future driving errors, but the present invention is not limited to driving risk.

[0183] In this embodiment, the vehicle control device 1 includes a notification control unit 26c that executes predetermined notifications based on the prediction results of the driving influence degree prediction unit 25. With this configuration, appropriate notifications can be executed according to the degree of influence on future driving errors, effectively preventing unpredictable situations during driving.

[0184] The second driving error detection unit 24b uses the driver's driving information D1a to detect driving error behavior D3c. Based on this structure, the driver's driving error behavior D3c can be detected with high accuracy, and therefore, the degree of influence on future driving error behaviors can be predicted with high accuracy.

[0185] 4.4 Emotion-based vehicle control

[0186] In this embodiment, the vehicle control device 1 includes: an environment detection unit 22 that detects environmental information D1 about at least one of the interior and exterior of the target vehicle; an emotion estimation unit 23 that estimates the emotion D2 of the driver of the target vehicle based on the environmental information D1; a control content determination unit 26a that determines the control content of the target vehicle when the estimated emotion D2 is a strong emotion; and a vehicle control unit 26b that performs control of the target vehicle based on the determined control content.

[0187] Based on this configuration, a vehicle control device 1 can be provided to control the vehicle according to the strong emotions influencing future driving errors, by determining the control content of the target vehicle when the driver experiences strong emotions. This allows appropriate measures to be taken based on the driver's future driving errors.

[0188] In this embodiment, an example has been given of predicting driving error behavior when the estimated emotion D2 is a strong emotion, and performing control over the target vehicle based on the predicted driving error behavior and driving risk. However, the prediction of driving error behavior and driving risk can be omitted. In other words, when the estimated emotion D2 is a strong emotion, determining the control content of the target vehicle can be appropriately and broadly applied to the degree to which the vehicle can be controlled based on the strong emotion that influences future driving error behavior.

[0189] The control content determination unit 26a determines the control content based on at least one of the degree of intense emotion and the category of intense emotion. According to this configuration, the control content executed when experiencing intense emotions can be appropriately modified.

[0190] Environmental information D1 includes traffic environment information D1c for the target vehicle and biometric information D1b for the driver. Emotion estimation unit 23 estimates emotion D2 based on each of the traffic environment information D1c and biometric information D1b. Control content determination unit 26a performs processing to determine control content based on the degree of strong emotion estimated based on traffic environment information D1c and processing to determine control content based on the category of strong emotion estimated based on biometric information D1b. According to this configuration, the control content executed when experiencing strong emotions, the control priority, etc., can be changed based on the traffic environment, biometric information D1b, etc., so that appropriate measures can be taken.

[0191] The vehicle control device 1 includes a driving error behavior prediction unit 24. When the estimated emotion D2 is a strong emotion, the driving error behavior prediction unit 24 predicts driving error behaviors D3a and D3b of the driver. Among the driving error behaviors D3a and D3b predicted by the driving error behavior prediction unit 24, the control content determination unit 26a sets the control content of the target vehicle to notification processing for the driver in the case of driving error behavior D3a caused by the driver's intentional behavior, and sets the control content of the target vehicle to driving control in the case of driving error behavior D3b caused by the driver's unintentional behavior. According to this configuration, the control content can be appropriately changed based on whether the predicted driving error behavior is an intentional driving error behavior performed by the driver.

[0192] 4.5 Vehicle control when the environment intensifies emotional stress

[0193] In this embodiment, the vehicle control device 1 includes: an environment detection unit 22 that detects environmental information D1 regarding at least one of the interior and exterior of the target vehicle; an emotion estimation unit 23 that estimates the driver's strong emotions based on the environmental information D1; a driving error behavior prediction unit 24 that predicts the driver's driving error behavior; and a risk suppression unit 26 that executes predetermined controls to avoid situations based on the prediction results of the driving error behavior prediction unit 24. When the level of strong emotion estimated by the emotion estimation unit 23 is high, the risk suppression unit 26 executes the predetermined controls without using the prediction results of the driving error behavior prediction unit 24 (steps S1a to S4a).

[0194] With this configuration, appropriate controls can be implemented using predictions of the driver's erroneous driving behaviors. When the environment intensifies the driver's emotional state, pre-defined controls can be executed quickly without relying on predictions of erroneous driving behaviors. This allows for rapid action based on the situation.

[0195] The emotion estimation unit 23 includes a first emotion estimation unit 23a that estimates a strong emotion D2a based on the driver's biometric information D1b, which constitutes part of the environmental information D1, and a second emotion estimation unit 23b that estimates a strong emotion D2b based on traffic environment information D1c, which constitutes part of the environmental information D1. When the level of the strong emotion D2b estimated by the second emotion estimation unit 23b is high, the risk suppression unit 26 performs predetermined control if the first emotion estimation unit 23a does not estimate the strong emotion D2a.

[0196] According to this configuration, since each of the strong emotion D2a based on biometric information D1b and the strong emotion D2b based on traffic environment information D1c can be estimated, predetermined control can be executed based on each estimation result. When the level of strong emotion estimated based on traffic environment information D1c is high, predetermined control can be executed quickly because it is performed without estimating the strong emotion D2a based on the driver's biometric information D1b. Therefore, when the traffic environment increases the level of strong emotion, measures can be taken quickly.

[0197] When the intensity of the strong emotion D2b estimated by the second emotion estimation unit 23b is high, the risk suppression unit 26 performs a predetermined control that induces attention to the traffic environment. Based on this configuration, appropriate notification can be executed quickly.

[0198] When the intensity of the strong emotion D2b estimated by the second emotion estimation unit 23b is not higher than the first predetermined value, the first emotion estimation unit 23a estimates the driver's strong emotion D2a based on the biometric information D1b. When the estimated intensity of the strong emotion D2a is not lower than the second predetermined value, the risk suppression unit 26 executes a notification control to reduce the intensity of the strong emotion as a predetermined control.

[0199] According to this configuration, when the level of intense emotion D2a estimated based on bioinformation D1b is not lower than a second predetermined value, emotional health can be encouraged by executing notification control to reduce the level of intense emotion.

[0200] After executing notification control to reduce the intensity of strong emotions, the first emotion estimation unit 23a estimates the driver's strong emotion D2a based on biometric information D1b, and when the estimated strong emotion D2a is a specific strong emotion, the risk suppression unit 26 performs alarm processing according to the category and intensity of the strong emotion (step S12a).

[0201] According to this configuration, when the strong emotion estimated based on bioinformation D1b is a specific strong emotion following encouraged emotional health, alarm processing can be simplified or eliminated and excessive alarms can be suppressed compared to performing alarm processing before encouraging emotional health, since alarm processing is performed according to the category and degree of this strong emotion.

[0202] 5. Other implementation methods

[0203] The above-described embodiments are merely one embodiment of the present invention, and can be freely modified and applied without departing from the spirit of the present invention.

[0204] For example, vehicle control device 1 is illustrated in Figure 3 The present invention is not limited to the following steps: step S13a involves acquiring driving information D1a and traffic environment information D1c, and step S14a involves predicting future driving errors via the second driving error detection unit 24b and the first driving error prediction unit 24a. For example, the vehicle control device 1 can acquire driving information D1a in step S13a and detect (predict) future driving errors via the second driving error detection unit 24b in step S14a.

[0205] Configurations implemented through software, hardware, or a combination of both can be widely applied. Figure 3The configuration of each unit is shown. A portion of the configuration of the vehicle control device 1, the emotion estimation device, and the driving prediction device described above can be located in a system external to the target vehicle, such as a server on a communication network. In this case, information acquired in the target vehicle is transmitted to the external system, which performs at least a portion of the processing performed by the vehicle control device 1, the emotion estimation device, and the driving prediction device. Output data based on the processing results is transmitted to the target vehicle, and various controls within the target vehicle are implemented.

[0206] The way processing units are divided in the flowchart and their processing order are not limited to the example shown, and can be changed appropriately.

[0207] The case where the program 32 for implementing the vehicle control method, emotion estimation method, and driving prediction method according to the present invention is recorded on the vehicle control device 1 has been described; however, the program 32 can be obtained from an external device via communication. The program 32 can be recorded on a suitable recording medium so that it can be read by a computer. A magneto-optical recording medium or a semiconductor memory device can be used as the recording medium.

[0208] 6. Configurations supported by the above implementation methods

[0209] The above implementation method is a specific example of the following configuration.

[0210] (Configuration 1) The vehicle control device includes: an environment detection unit that detects environmental information about at least one of the interior and exterior of the target vehicle; an emotion estimation unit that estimates the driver's emotion based on the environmental information; a driving error behavior prediction unit that predicts the driver's driving error behavior when the estimated emotion is a strong emotion; a driving risk prediction unit that predicts driving risk based on the predicted driving error behavior; and a risk suppression unit that performs predetermined control according to the predicted driving risk.

[0211] This configuration allows for highly accurate prediction of driving risks by anticipating erroneous driving behaviors when the driver is experiencing strong emotions. This high accuracy enables appropriate control based on the driving risks. Therefore, it allows for effective prevention of unforeseen situations during driving by improving the accuracy of driving risk prediction.

[0212] (Configuration 2) According to the vehicle control device of Configuration 1, the environment detection unit includes: a bio-information detection unit, which detects the driver's bio-information as the environment information; and a traffic environment detection unit, which detects the traffic environment information of the target vehicle as the environment information, and the emotion estimation unit estimates whether the driver's emotion is the strong emotion based on the driver's bio-information, and estimates whether the driver's emotion is the strong emotion based on the traffic environment information.

[0213] This configuration can improve the accuracy of estimating strong emotions, which in turn improves the accuracy of predicting driving risks and helps to control and suppress over-control comfort more appropriately.

[0214] (Configuration 3) The vehicle control device according to Configuration 1 or 2, wherein the environmental information includes driving information about the driver in the target vehicle, and the driving error behavior prediction unit predicts the driver's driving error behavior based on real-time driving conditions, the real-time driving conditions including driving error behavior statistics when experiencing the strong emotion and the driving error behavior detected using the driving information.

[0215] This configuration can improve the accuracy of predicting driving errors, which in turn improves the accuracy of predicting driving risks and helps to control and suppress over-control comfort more appropriately.

[0216] (Configuration 4) The vehicle control device according to any one of Configurations 1 to 3, wherein the driving risk prediction unit predicts the driving risk based on statistical data that allows identification of the relationship between the driving error and the situation to be avoided.

[0217] This configuration can improve the accuracy of predicting driving risks that could lead to situations that need to be avoided.

[0218] (Configuration 5) The vehicle control device according to any one of Configurations 1 to 4, wherein the risk suppression unit performs at least one of vehicle speed control, braking control, route guidance processing, state recovery processing and warning processing as the predetermined control according to the driving risk.

[0219] This configuration allows for appropriate vehicle control based on predicted driving risks, thereby reducing driving risks.

[0220] (Configuration 6) According to any one of Configurations 1 to 4, the vehicle control device further includes a storage unit that stores the location of the target vehicle when the estimated emotion is a specific strong emotion, and when the driving risk prediction unit predicts that the target vehicle will pass through the stored location, the risk suppression unit performs route guidance processing to guide the driver along a route to avoid the stored location.

[0221] This configuration allows the driver to be guided along a route that avoids locations where the driver has previously experienced strong emotions, thereby preventing situations that could increase driving risks.

[0222] (Configuration 7) The vehicle control device according to any one of configurations 1 to 6, wherein when the driving error behavior predicted by the driving error behavior prediction unit is a driving error behavior caused by the driver's intentional behavior, the risk suppression unit performs notification processing to notify the driver as the predetermined control, and when the driving error behavior predicted by the driving error behavior prediction unit is a driving error behavior caused by the driver's unintentional behavior, the risk suppression unit performs driving control of the target vehicle as the predetermined control.

[0223] According to this configuration, the control can be modified based on whether the predicted driving error behavior is an intentional act by the driver, so that appropriate control can be performed according to the driving error behavior.

[0224] (Configuration 8) A vehicle control method executed by a computer, the vehicle control method comprising: an environment detection step for detecting environmental information about at least one of the interior and exterior of a target vehicle; an emotion estimation step for estimating a driver's emotion based on the environmental information; a driving error behavior prediction step for predicting a driving error behavior of the driver when the estimated emotion is a strong emotion; a driving risk prediction step for predicting a driving risk based on the predicted driving error behavior; and a risk mitigation step for performing predetermined control according to the predicted driving risk.

[0225] According to this method, when a driver is experiencing strong emotions, highly accurate predictions of driving risks can be achieved by forecasting erroneous driving behaviors. This high accuracy enables appropriate control based on driving risks. Therefore, improving the accuracy of driving risk prediction can effectively prevent unforeseen situations during driving.

Claims

1. A vehicle control device, the vehicle control device comprising: An environmental detection unit that detects environmental information about at least one of the interior and exterior of the target vehicle; An emotion estimation unit estimates the driver's emotion based on the environmental information; The driving error behavior prediction unit predicts the driver's driving error behavior when the estimated emotion is a strong emotion. A driving risk prediction unit that predicts driving risk based on predicted driving errors; as well as A risk suppression unit that performs predetermined controls based on predicted driving risks.

2. The vehicle control device according to claim 1, wherein, The environmental detection unit includes: a bio-information detection unit that detects the driver's bio-information as the environmental information; and a traffic environment detection unit that detects the target vehicle's traffic environment information as the environmental information. The emotion estimation unit estimates whether the driver's emotion is the strong emotion based on the driver's biometric information, and also estimates whether the driver's emotion is the strong emotion based on the traffic environment information.

3. The vehicle control device according to claim 1, wherein, The environmental information includes driving information about the driver in the target vehicle, and The driving error behavior prediction unit predicts the driver's driving error behavior based on real-time driving conditions, which include driving error behavior statistics when experiencing the strong emotion and the driving error behavior detected using the driving information.

4. The vehicle control device according to claim 1, wherein, The driving risk prediction unit predicts the driving risk based on statistical data that allows identification of the relationship between the driving error and the situation to be avoided.

5. The vehicle control device according to claim 1, wherein, The risk mitigation unit performs at least one of the following as the predetermined control based on the driving risk: vehicle speed control, braking control, route guidance processing, state recovery processing, and warning processing.

6. The vehicle control device according to claim 1, further comprising a storage unit, wherein when the estimated emotion is a specific strong emotion, the storage unit stores the location of the target vehicle, wherein, When the driving risk prediction unit predicts that the target vehicle will pass through the stored location, the risk suppression unit performs route guidance processing to guide the driver along a route used to avoid the stored location.

7. The vehicle control device according to claim 1, wherein, When the driving error behavior predicted by the driving error behavior prediction unit is caused by the driver's intentional behavior, the risk suppression unit performs notification processing to notify the driver as the predetermined control. When the driving error behavior predicted by the driving error behavior prediction unit is caused by the driver's unintentional behavior, the risk suppression unit performs driving control of the target vehicle as the predetermined control.

8. A vehicle control method executed by a computer, the vehicle control method comprising: The environmental testing procedure involves detecting environmental information about at least one of the interior and exterior of the target vehicle. The emotion estimation step estimates the driver's emotion based on the environmental information; The driving error behavior prediction step predicts the driver's driving error behavior when the estimated emotion is a strong emotion. The driving risk prediction process involves predicting driving risks based on the predicted driving errors. as well as Risk mitigation steps involve implementing predetermined controls based on the predicted driving risks.

Images