JP2019207544A - Travel control device, travel control method, and travel control program - Google Patents

Abstract

To provide a travel control device of a vehicle capable of considering the driving experience level according to a vehicle type of each occupant.SOLUTION: A driving preference database 11 stores a driving experience level and driving preference data according to an occupant and a vehicle type. An occupant recognition part 12 recognizes each occupant on a vehicle. A driving preference data extraction part 13 extracts from the driving preference database 11 the driving experience level and the driving preference data corresponding to the vehicle type on which each occupant recognized by the occupant recognition part 12 boards. A driving preference data synthesizing part 14 synthesizes the driving preference data of each occupant extracted by the driving preference data extraction part 13 by weighting each occupant's driving experience level. A travel command value calculation part 15 uses the driving preference data synthesized by the drive preference data synthesizing part 14 to calculate a travel command value for controlling the travel of the vehicle.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a vehicle travel control device, a travel control method, and a travel control program.

  The driving support system according to Patent Document 1 acquires information representing passenger driving preferences from a database that holds passenger information representing the attributes of the passengers of the vehicle, and provides driving assistance based on the acquired information. Had gone.

Japanese Patent Laying-Open No. 2015-121842

  In the driving support system according to Patent Literature 1, examples of passenger information include name, age, sex, driving history, and license information. This driving history is the driving years of a vehicle owned by the passenger or a vehicle that the passenger is driving on a daily basis. The license information indicating the license type such as the blue license and the gold license is also based on the driving experience of the vehicle owned by the passenger or the vehicle that the passenger is driving on a daily basis. For example, if a passenger who is driving a truck on board a sedan this time, the above-mentioned driving support system will be able to use this driver even if the passenger's truck driving years are long or the license type is a gold license. Passenger information about passengers should not be greatly reflected in driving support for sedans. This is because the behavior of the vehicle is greatly different between the truck and the sedan. However, Patent Document 1 does not consider the type of vehicle. Further, in Patent Document 1, when there are a plurality of passengers, a driving support method that increases the satisfaction level of the entire passengers is not considered.

  The present invention has been made to solve the above-described problems, and an object thereof is to consider the driving experience level corresponding to the vehicle type of each occupant.

  The travel control device according to the present invention includes a driving preference database that stores driving experience and driving preference data according to the occupant and the vehicle type, and an occupant recognition unit that recognizes each occupant on the vehicle. A driving preference data extracting unit for extracting driving experience data and driving preference data corresponding to the type of vehicle on which each occupant is recognized by the occupant recognition unit, and a driving preference data extracting unit; Driving preference data combining unit weighting each passenger's driving preference data extracted in accordance with the driving experience degree of each passenger and combining the driving preference data synthesized by the driving preference data combining unit And a travel command value calculation unit that calculates a travel command value for controlling.

  According to the present invention, the driving preference data corresponding to the vehicle type of each occupant is synthesized by weighting with the driving experience level corresponding to the vehicle type of each occupant. Control can be realized. In addition, when combining the driving preference data of each occupant, weighting considering the driving experience according to each occupant's vehicle type is performed, so the contribution of driving reliability data with low reliability is reduced, and the reliability The contribution of highly-priced driving preference data can be increased.

It is a block diagram which shows the structural example of the traveling control apparatus which concerns on Embodiment 1. FIG. FIG. 2A shows an example of the weight of each occupant used by the driving preference data synthesizer of the first embodiment, FIG. 2A shows the weight before normalization, and FIG. 2B shows the weight after normalization. FIG. 3A shows another example of the weight of each occupant used by the driving preference data synthesizer of Embodiment 1, FIG. 3A shows the weight before normalization, FIG. 3B shows the weight after normalization, and FIG. 3C shows the weight after correction and normalization. It is a weight. 4 is a flowchart illustrating an operation example of the travel control apparatus according to the first embodiment. It is a block diagram which shows the structural example of the traveling control apparatus which concerns on Embodiment 2. FIG. 6 is a flowchart illustrating an operation example of the travel control apparatus according to the second embodiment. FIG. 10 is a block diagram illustrating a configuration example of a travel control device according to a third embodiment. 10 is a flowchart illustrating an operation example of the travel control apparatus according to the third embodiment. It is a figure which shows the hardware structural example of the traveling control apparatus which concerns on each embodiment.

Embodiment 1 FIG.
FIG. 1 is a block diagram illustrating a configuration example of a travel control apparatus 10 according to the first embodiment. A travel control device 10 shown in FIG. 1 is mounted on a vehicle, and includes a driving preference database 11, an occupant recognition unit 12, a driving preference data extraction unit 13, a driving preference data synthesis unit 14, and a driving command value calculation unit. 15. In addition, an in-vehicle camera 1, an input / output device 2, and a vehicle periphery measuring device 3 are connected to the traveling control device 10.

  The in-vehicle camera 1 images each occupant on the vehicle and outputs the captured image to the occupant recognition unit 12.

  The input / output device 2 includes a display device or a voice output device that presents information to each occupant on board the vehicle, and an input device that receives operation input or the like by each occupant. The input / output device 2 according to the first embodiment outputs data such as a vehicle destination, estimated arrival time, and travel route (hereinafter referred to as destination data) to the travel command value calculation unit 15. Further, the input / output device 2 according to the first embodiment outputs information on the driving mode selected and operated by one of the passengers on the vehicle to the driving preference data extraction unit 13.

  The vehicle periphery measuring device 3 measures data related to the vehicle and its surroundings (hereinafter referred to as the vehicle and surrounding data) such as the speed, acceleration, distance between vehicles, and traffic conditions of the vehicle (hereinafter referred to as vehicle and surrounding data), and commands the vehicle and surrounding data to travel. Output to the value calculation unit 15.

  The driving preference database 11 stores a driving experience level and driving preference data according to the occupant and the vehicle type. The driving experience level is, for example, the number of boarding times for each vehicle type for each occupant. The driving preference data is, for example, a real multidimensional vector whose elements are driving pattern items. Below, an example of the driving preference data of a certain occupant for a certain vehicle type is shown. The longitudinal acceleration, the lateral acceleration, the accelerator depression amount, the brake depression amount, and the inter-vehicle distance are driving pattern items.

Driving preference data = (longitudinal acceleration, lateral acceleration, accelerator depression amount, brake depression amount, inter-vehicle distance, ...)

  When there are four driving modes, the driving preference data stored in the driving preference database 11 according to the vehicle type of Mr. Y is the following four. The “slow” and “hurry” driving modes are modes that emphasize speed or time, the “safe” driving mode is a mode that emphasizes safety, and the “active” driving mode gives an aggressive ride feeling. This is an important mode. The value of each element of the multidimensional vector is not an actual value of the longitudinal acceleration, the lateral acceleration, the accelerator depression amount, the brake depression amount, and the inter-vehicle distance, but a value represented by, for example, 10 levels.

Y's "slow" data = (3, 3, 2, 6, 5, ...)
Y's “Hurry” data = (7, 5, 8, 5, 4, ...)
Y's "safety" data = (3, 3, 2, 5, 6, ...)
Y's "active" data = (8, 6, 8, 5, 5, ...)

  In this way, the driving preference database 11 stores driving preference data of four driving modes for one vehicle type for Mr. Y. Mr. N's driving preference data reflects Mr. N's driving preference and is different from Mr. Y's driving preference data. Here, an example in which there are a plurality of operation modes has been described, but one operation mode may be used.

  In addition, when the driving experience level and driving preference data of the occupant are not stored in the driving preference database 11, for example, when the occupant operates the input / output device 2, the driving experience level and the driving preference data are converted into the driving preference database. 11 may be registered.

  The occupant recognition unit 12 recognizes each occupant on the vehicle using the image captured by the in-vehicle camera 1. The occupant recognition unit 12 outputs the recognized information on each occupant to the driving preference data extraction unit 13.

  The driving preference data extraction unit 13 extracts the driving experience level and driving preference data of each occupant recognized by the occupant recognition unit 12 from the driving preference database 11. At this time, the driving preference data extraction unit 13 drives the driving experience and driving preference data of the same vehicle type as the vehicle on which each occupant is boarded, or driving of a vehicle type similar to the vehicle on which each occupant is boarding. The experience level and driving preference data are extracted from the driving preference database 11. In addition, when there are a plurality of driving modes, the driving preference data extraction unit 13 selects driving preference data corresponding to the driving mode selected by operating one of the passengers on the vehicle by operating the input / output device 2. Extracted from the driving preference database 11. The driving preference data extraction unit 13 outputs the driving experience level and driving preference data of each occupant extracted from the driving preference database 11 to the driving preference data synthesis unit 14.

  The driving preference data synthesizing unit 14 synthesizes the driving preference data of each occupant extracted by the driving preference data extracting unit 13 by weighting with the driving experience level of each occupant extracted by the driving preference data extracting unit 13. The driving preference data synthesis unit 14 outputs the synthesized driving preference data to the travel command value calculation unit 15.

For example, when Mr. Y's driving preference data is the vector V _Y and Mr. N's driving preference data is the vector V _N , the combined driving preference data vector V is expressed by Expression (1). ω _Y and ω _N are weights.

V = ω _Y V _Y + ω _N V _N (1)

  The driving preference data synthesis unit 14 adjusts the weight by multiplying the weight by a coefficient. However, the driving preference data synthesis unit 14 normalizes the final weight of each occupant used in the calculation of Expression (1) so that the sum of the weights of each occupant becomes “1”.

  The driving preference data synthesis unit 14 uses the driving experience level of each occupant to adjust the weight of each occupant. As described above, the driving experience level is the number of times a vehicle on which each occupant is boarded or the number of times a vehicle similar to the vehicle on which each occupant is boarded. When there are few passengers who are not familiar with the vehicle, the driving preference data of the passenger is considered to have low reliability. For example, if the passenger sitting in the rear seat is the first person to board the vehicle and thus feels uneasy about driving, the driving preference data composition unit 14 adjusts the weight of the driving preference data of the passenger to be small. By doing so, the contribution to driving control of driving preference data with low reliability is made low. On the other hand, since the driving preference data of the occupant with a large number of boarding is considered to be highly reliable, the driving preference data synthesis unit 14 adjusts the weight of the driving preference data of the occupant to increase the reliability. Increase the contribution to driving control of highly-priced driving preference data.

  If the number of passengers sitting in the driver's seat is extremely high and all passengers except the passenger sitting in the driver's seat are low, if the driver's preference data is too important, the entire passenger It is not always appropriate when considering the degree of satisfaction, and it is better to consider the balance with other passengers. Thus, when the variation of the number of boarding of each occupant is large, the driving preference data synthesis unit 14 corrects the weight of each occupant so that the variation is small, so that the driving preference of the occupant with extremely large or small number of boarding Reduce the contribution of data to travel control.

  For example, the driving preference data synthesis unit 14 multiplies the weight of the occupant whose number of boarding is 1 to 10 times by the coefficient “0.4”, and the coefficient of “0” to the weight of the occupant whose number of boarding is 11 to 50 times. .8 ”and multiplying the weight of the occupant who has boarded 51 times or more by the coefficient“ 1.2 ”to adjust the weight. The coefficient is multiplied by an initial value (for example, “1.0”) of the weight of each occupant. The driving preference data synthesizing unit 14 adjusts the initial value of the weight of each occupant by multiplying the coefficient and then normalizes the sum of the weights of each occupant to be “1”.

  FIG. 2 shows an example of the weight of each occupant used by the driving preference data synthesis unit 14 of Embodiment 1, FIG. 2A shows the weight before normalization, and FIG. 2B shows the weight after normalization. Since the number of boarding of the driver's seat is 50 or more, the weighting coefficient is “1.2”, and for each other passenger, the number of boarding is 11 to 50, and the weighting coefficient is “0”. .8 ". When the initial value of each occupant's weight is “1.0”, the driving preference data composition unit 14 multiplies the initial value of each occupant's weight by the above coefficient, whereby the weight of each occupant is shown in FIG. 2A. Adjust to the value. The driving preference data synthesizing unit 14 determines that the variation in weight of each occupant after adjustment is less than a predetermined threshold value, and does not correct the weight of each occupant after adjustment. Then, the driving preference data composition unit 14 normalizes the adjusted weights of the occupants so that the sum of the weights of the occupants is “1”, and sets the values as shown in FIG. 2B.

  FIG. 3 shows another example of the weight of each occupant used by the driving preference data synthesis unit 14 of Embodiment 1, FIG. 3A shows the weight before normalization, FIG. 3B shows the weight after normalization, and FIG. 3C shows the correction. And the weight after normalization. As shown in FIG. 3A, the weight of each other passenger is as small as “0.4” with respect to the weight “1.2” of the driver's seat occupant, which is only 1/3 of the weight of the driver's seat. Therefore, the driving preference data synthesis unit 14 determines that the variation in weight of each occupant after adjustment shown in FIG. 3A is equal to or greater than the threshold value, and corrects the weight of each occupant after adjustment.

  Here, an example of a correction method for the weight of each occupant after adjustment will be described. For example, the driving preference data combining unit 14 corrects each weight so that the ratio of the largest weight to the smallest weight among the adjusted weights of each occupant is equal to or less than the ideal value. In the example of FIG. 3A, the weight ratio between the driver's seat occupant and each of the other occupants is 1.2 / 0.4 = 3 times. Correct it to be less than double.

First, the driving preference data synthesizing unit 14 arranges the weights ω _i of the occupants in order of magnitude, and sets ω ₁ in ascending order. That is, minimum weight value = ω ₁ ≦ ω ₂ ≦... ≦ ω _n = maximum weight value (n is the total number of passengers).
Subsequently, the driving preference data synthesizing unit 14 sets α = ω _n / ω ₁ and when α exceeds β, which is “ratio of ideal maximum weight and minimum weight” (β = 2 in the above example). Γ = β / α.
Subsequently, the driving preference data combining unit 14, "omega _{i (before} correction)" performs the correction using equation (2) when a minimum weighted value, "omega _{i (before} correction)" is the weight minimum Otherwise, correction is performed using equation (3).
ω _i (after correction) = ω _i (before correction) (2)
ω _i (after correction) = γ · ω _i (before correction) (3)
Finally, the driving preference data composition unit 14 performs normalization on the weight “ω _i (after correction)” of each occupant to obtain a final weight.

  The weight of each occupant after correction and normalization obtained by the above correction method is as shown in FIG. 3C. In FIG. 3C, the weight of each occupant after normalization is “0.4” for the driver's seat occupant and “0.2” for the other occupants, and the ratio is corrected to be twice or less. . On the other hand, as shown in FIG. 3B, the weight of each occupant after normalization that has not been corrected is “0.5” for the occupant in the driver seat and “0.16” for the other occupants. The ratio is over twice.

  In the above example, the driving preference data synthesis unit 14 increases the weight of the occupant in order to give priority to the driving preference data of the occupant having a high driving experience level. In order to prioritize data, the weight of the passenger may be increased.

  In addition, when the value of the driving experience level of a certain occupant is significantly different from the value of the driving experience level of another occupant, the driving preference data synthesis unit 14 may correct the weight of the occupant to “0”. Thereby, the driving preference data of the occupant is excluded from the synthesis target and does not contribute to the travel control.

  In addition, the driving preference data synthesis unit 14 may correct the weight with reference to the value of the driving preference data. For example, when the value of the driving preference data of a certain occupant is significantly different from the value of the driving preference data of another occupant, the driving preference data synthesis unit 14 corrects the weight of the occupant to “0”. Thereby, the driving preference data of the occupant is excluded from the synthesis target and does not contribute to the travel control.

  The travel command value calculation unit 15 includes the driving preference data synthesized by the driving preference data synthesis unit 14, the destination data output from the input / output device 2, and the vehicle and surrounding data output from the vehicle periphery measurement device 3. Is used to calculate a travel command value for controlling the travel of the vehicle. The travel command value is, for example, speed, acceleration, and brake time. The travel command value calculation unit 15 outputs the calculated travel command value to an on-vehicle ECU (Electronic Control Unit) not shown. A vehicle-mounted ECU (not shown) realizes automatic driving by controlling the traveling of the vehicle according to the traveling command value.

  The travel command value calculation unit 15 can also reversely calculate the driving preference data using the destination data, the vehicle and surrounding data, and the travel command value. That is, the driving preference data and the travel command value can be converted bidirectionally. The usage of the calculated driving preference data will be described in detail in the third embodiment.

  The method of calculating the travel command value by the travel command value calculation unit 15 is realized by, for example, a technique described in Japanese Patent Application Laid-Open No. 2007-261486. Japanese Patent Application Laid-Open No. 2007-261486 describes a method for converting a plurality of physical parameters related to vehicle performance and a plurality of psychological parameters indicating occupant sensitivity evaluation values related to the performance. The plurality of physical parameters related to the vehicle performance are equivalent to the travel command value, and the plurality of psychological parameters indicating the passenger's sensitivity evaluation value are equivalent to the driving preference parameter.

Next, the operation of the travel control device 10 will be described.
FIG. 4 is a flowchart showing an operation example of the travel control apparatus 10 according to the first embodiment. The travel control device 10 starts the operation shown in the flowchart of FIG. 4 before the vehicle starts traveling.

  In step ST <b> 11, the occupant recognition unit 12 recognizes each occupant on the vehicle using the captured image of the in-vehicle camera 1. This is pre-processing for extracting driving preference data of a passenger on board from the driving preference database 11.

  In step ST12, the input / output device 2 accepts a driving mode selection operation by one of the passengers in the vehicle. The driving preference data extraction unit 13 receives information on the selected operation mode from the input / output device 2. As described above, the operation mode is a mode such as “rush”, “slow”, “safe”, and “active”. The occupant selects one of these operation modes when boarding. The selected operation mode is the basis of travel control during automatic operation in step ST15 described later.

  In step ST13, the driving preference data extracting unit 13 recognizes each occupant by the occupant recognition unit 12, driving mode information from the input / output device 2, and driving preference data and driving experience level corresponding to the vehicle type. Are extracted from the driving preference database 11. For example, when Mr. Y and Mr. N are on the vehicle and “slow” mode is selected as the operation mode, “slow” data of Mr. Y of the same vehicle type as this vehicle or a similar vehicle type and N The “slow” data of Mr. Y, the driving experience level of Mr. Y and the driving experience level of Mr. N of the same vehicle type as this vehicle or similar vehicle types are extracted.

  In step ST14, the driving preference data synthesis unit 14 synthesizes the driving preference data of each occupant extracted by the driving preference data extraction unit 13 with weighting by the driving experience level of each occupant, and creates one new driving preference data. Generate.

  In step ST <b> 15, the travel command value calculation unit 15 includes driving preference data synthesized by the driving preference data synthesis unit 14, destination data from the input / output device 2, and vehicle and surrounding data from the vehicle periphery measurement device 3. Is used to calculate a travel command value for controlling the travel of the vehicle. A vehicle-mounted ECU (not shown) starts automatic driving of the vehicle according to the travel command value. The travel command value calculation unit 15 repeats the calculation of the travel command value from when the vehicle starts traveling until it arrives at the destination.

  As described above, the travel control device 10 according to the first embodiment includes the driving preference database 11, the occupant recognition unit 12, the driving preference data extraction unit 13, the driving preference data synthesis unit 14, and the travel command value calculation unit. 15. The driving preference database 11 stores a driving experience level and driving preference data according to the occupant and the vehicle type. The occupant recognition unit 12 recognizes each occupant on the vehicle. The driving preference data extraction unit 13 extracts, from the driving preference database 11, the driving experience level and driving preference data corresponding to the type of vehicle on which each occupant recognized by the occupant recognition unit 12. The driving preference data synthesizing unit 14 synthesizes the driving preference data of each occupant extracted by the driving preference data extraction unit 13 by weighting each occupant's driving experience level. The travel command value calculation unit 15 uses the drive preference data synthesized by the drive preference data synthesis unit 14 to calculate a travel command value that controls the travel of the vehicle. Since the traveling control device 10 weights and synthesizes the driving preference data of each occupant and reflects it in the vehicle traveling, it is possible to realize traveling control that increases the satisfaction level of the entire occupant. In addition, when combining the driving preference data of each occupant, the traveling control device 10 generates driving preference data that is weighted and combined in consideration of the driving experience level of the vehicle on which the passenger is riding. The contribution of data can be reduced, the contribution of reliable driving preference data can be increased, and stable travel control can be realized.

  In addition, the driving preference data synthesis unit 14 according to the first embodiment adjusts the weight of driving preference data of the occupant with a large value of driving experience to increase the weight of driving preference data of the occupant with a small value of driving experience. Adjust so that becomes smaller. Then, the driving preference data synthesis unit 14 corrects the weight of each occupant after adjustment according to the variation in the weight of each occupant after adjustment. Thus, the traveling control device 10 is a traveling control in which the value of the driving experience level is not excessively influenced by the driving preference data of an occupant that is extremely different from other occupants, and the occupant's overall satisfaction is high and stable. Can be realized.

Embodiment 2. FIG.
FIG. 5 is a block diagram illustrating a configuration example of the travel control device 10 according to the second embodiment. The travel control device 10 according to the second embodiment has a configuration in which an emotion evaluation unit 16 is added to the travel control device 10 according to the first embodiment shown in FIG. An infrared sensor 4 and a heart rate sensor 5 are connected to the travel control device 10. In FIG. 5, the same or corresponding parts as those in FIG.

  The infrared sensor 4 measures the body temperature of each occupant on the vehicle and outputs body temperature data to the emotion evaluation unit 16. The heart rate sensor 5 measures the heart rate of each occupant on the vehicle and outputs heart rate data to the emotion evaluation unit 16. The in-vehicle camera 1 images each occupant boarding the vehicle and outputs the captured image to the emotion evaluation unit 16 in addition to the occupant recognition unit 12. The infrared sensor 4, the heart rate sensor 5, and the in-vehicle camera 1 correspond to a biometric sensor. The biometric sensor is not limited to the infrared sensor 4, the heart rate sensor 5, and the in-vehicle camera 1.

  The emotion evaluation unit 16 recognizes the facial expression of each occupant on the vehicle using the image captured by the in-vehicle camera 1. The emotion evaluation unit 16 uses each facial expression recognized from the captured image of the in-vehicle camera 1, body temperature data measured by the infrared sensor 4, heart rate data measured by the heart rate sensor 5, and the like. Evaluate feelings. Specifically, the emotion evaluation unit 16 calculates an emotion evaluation value obtained by quantifying the positive emotion and the negative emotion of each occupant during vehicle travel. Positive emotions are emotions such as comfort or satisfaction. Negative emotions are emotions such as discomfort, dissatisfaction, or anxiety. The emotion evaluation unit 16 outputs the calculated emotion evaluation value to the driving preference data synthesis unit 14. Here, it is assumed that the occupant feels uncomfortable with the current driving as the emotion evaluation value of the occupant is low, and the occupant feels comfortable with the current driving as the emotion evaluation value of the occupant is high.

  The driving preference data synthesizing unit 14 periodically re-synthesizes the driving preference data while the vehicle is traveling, and outputs it to the traveling command value calculation unit 15. At that time, the driving preference data synthesis unit 14 re-adjusts the weight of each occupant adjusted according to the degree of driving experience of each occupant according to the emotion evaluation value of each occupant calculated by the emotion evaluation unit 16, Normalization is performed so that the sum of the weights of each occupant after readjustment becomes “1”.

  For example, when the occupant feels uncomfortable while driving the vehicle, the validity of the driving preference data synthesized by the driving preference data synthesis unit 14 is suspected. That is, it is assumed that the current driving does not match the driving preference of the occupant. Therefore, the driving preference data composition unit 14 re-synthesizes the driving preference data of each occupant after readjusting so that the weight of the driving preference data of the occupant having a low emotion evaluation value is increased. As a result, the driving preference data of the occupant with discomfort is greatly reflected in the travel command value, and the contribution of the driving preference data of the occupant to the travel control is increased, and as a result, the satisfaction level of the occupant is increased.

  However, the emotion evaluation value itself contains a lot of measurement noise, and also reflects inevitable emotions such as disturbance of emotions of each occupant of the own vehicle due to sudden braking of the preceding vehicle during driving There is a possibility that. Therefore, it is necessary to pay attention to resynthesis of driving preference data using emotion evaluation values.

  For example, for the measurement noise included in the emotion evaluation value, the emotion evaluation unit 16 calculates a moving average of a predetermined time length using time series data of the emotion evaluation value for the purpose of removing spike noise. You may output to the driving preference data synthetic | combination part 14. FIG. In this case, the driving preference data synthesizing unit 14 re-synthesizes the driving preference data by re-adjusting the weight only when the moving average of the emotion evaluation value of a certain occupant maintains a low value for a predetermined time.

  In addition, the emotion evaluation unit 16 determines that the distance between the preceding vehicle and the preceding vehicle is narrowed when the speed of the own vehicle does not change greatly in relation to disturbance of the passenger's emotion according to the behavior of the preceding vehicle. You may judge using the vehicle of the vehicle periphery measurement apparatus 3, and surrounding data. When the emotion evaluation unit 16 determines that the distance between the preceding vehicle and the preceding vehicle is narrowed using the vehicle and the surrounding data, even if the emotion evaluation value varies, the variation is not caused by the driving preference data. The output of the emotion evaluation value to the driving preference data synthesizer 14 may be temporarily suspended as a result of other responsibility. The driving preference data synthesizing unit 14 stops the resynthesis of the driving preference data while the output of the emotion evaluation value from the emotion evaluation unit 16 is stopped.

  In addition, when the emotion evaluation value output from the emotion evaluation unit 16 changes, the driving preference data synthesis unit 14 does not change the weight at once using the changed emotion evaluation value, but with the passage of time. You may make it change gradually.

  FIG. 6 is a flowchart showing an operation example of the travel control apparatus 10 according to the second embodiment. The operations in steps ST11 to ST13 shown in FIG. 6 are the same as those in steps ST11 to ST13 shown in FIG.

  In step ST21, the emotion evaluation unit 16 calculates the emotion evaluation value of each occupant using the measurement data of each occupant measured by a biometric sensor such as the in-vehicle camera 1, the infrared sensor 4, or the heart rate sensor 5. To do.

  In step ST <b> 22, the driving preference data synthesizing unit 14 uses the driving preference data of each occupant extracted by the driving preference data extracting unit 13 to evaluate each occupant's emotion calculated by the driving experience level and the emotion evaluation unit 16. A new driving preference data is generated by weighting the values and combining them.

  In step ST <b> 23, the travel command value calculation unit 15 includes driving preference data synthesized by the driving preference data synthesis unit 14, destination data from the input / output device 2, and vehicle and surrounding data from the vehicle periphery measurement device 3. Is used to calculate a travel command value for controlling the travel of the vehicle. A vehicle-mounted ECU (not shown) starts automatic driving of the vehicle according to the travel command value. The emotion evaluation unit 16, the driving preference data synthesis unit 14, and the travel command value calculation unit 15 calculate the emotion evaluation value and re-synthesize the driving preference data from when the vehicle starts to travel until it arrives at the destination. And the calculation of the travel command value is repeated.

  As described above, the travel control apparatus 10 according to the second embodiment includes the emotion evaluation unit 16 that calculates the emotion evaluation value of each occupant using the measurement result of each occupant measured by the biometric sensor. The driving preference data synthesis unit 14 adjusts the weight of each passenger's driving preference data according to the emotion evaluation value of each passenger calculated by the emotion evaluation unit 16. Thereby, the traveling control device 10 can realize more detailed traveling control according to the degree of feeling of each occupant with respect to the current driving, and can improve the satisfaction of the entire occupant.

  In the second embodiment, the emotion evaluation unit 16 calculates the emotion evaluation value of each occupant, but may detect a medical condition such as fever in addition to the emotion evaluation value. For example, the emotion evaluation unit 16 detects a sick person based on the body temperature data of the infrared sensor 4 and outputs information about the sick person to the driving preference data synthesis unit 14. The driving preference data synthesis unit 14 adjusts the weight of the driving preference data of the sick person detected by the emotion evaluation unit 16 so as to increase. Alternatively, when the sick person is detected by the emotion evaluation unit 16, the driving preference data synthesis unit 14 temporarily stops the synthesis of the driving preference data and outputs the driving preference data of the sick person to the driving command value calculation unit 15 as it is. Also good.

Embodiment 3 FIG.
FIG. 7 is a block diagram illustrating a configuration example of the travel control device 10 according to the third embodiment. The travel control device 10 according to the third embodiment has a configuration in which a scoring unit 17 and a learning unit 18 are added to the travel control device 10 of the second embodiment shown in FIG. In FIG. 7, parts that are the same as or correspond to those in FIGS. 1 and 5 are given the same reference numerals, and descriptions thereof are omitted.

  The scoring unit 17 stores the destination data output from the input / output device 2 while the vehicle is traveling, the vehicle and surrounding data output from the vehicle periphery measuring device 3, and the emotion evaluation value output from the emotion evaluation unit 16. To do. Then, the scoring unit 17 uses the stored data to score the travel content after the vehicle travels. Alternatively, the scoring unit 17 may score the traveling content up to that point during traveling of the vehicle. The scoring unit 17 may generate a scoring result for each occupant on the vehicle, or may generate one scoring result for all occupants on the vehicle.

  The scoring unit 17 scores items such as the difference between the arrival time and the scheduled arrival time at the destination, the number of sudden braking and sudden start during traveling, or the magnitude of the emotion evaluation value of each occupant. The scoring unit 17 increases the score as the difference between the actual arrival time at the destination and the estimated arrival time is smaller. In addition, the scoring unit 17 increases the points as the number of sudden braking during traveling or the number of sudden starts decreases. In addition, the scoring unit 17 adds points if the emotion evaluation value during the running continuously maintains a high value (that is, a positive emotion), and conversely maintains a low value (that is, a negative emotion). If you do, you will lose points.

  Further, the scoring unit 17 may weight or add scoring suitable for each occupant, reflecting the contents of the driving preference data of each occupant. For example, when it can be read from the driving preference data that Mr. Y prefers speed, the scoring unit 17 increases points for scoring items related to speed.

  As described in the second embodiment, the emotion evaluation value includes a lot of measurement noise. Therefore, the emotion evaluation unit 16 may calculate a moving average of a predetermined time length using the time series data of emotion evaluation values and output it to the scoring unit 17. In this case, the scoring unit 17 adds or subtracts points when a moving average of a certain passenger's emotion evaluation value maintains a high value or a low value for a predetermined time. In addition, the emotion evaluation unit 16 suspends the output of the emotion evaluation value to the scoring unit 17 when the passenger's occupant's emotions are disturbed according to the behavior of the preceding vehicle. May be excluded from the scoring target.

  The scoring unit 17 outputs the scoring results of each occupant to the input / output device 2 and inquires of each occupant whether or not to update the driving preference database 11. The scoring unit 17 receives from the input / output device 2 an answer indicating whether or not the driving preference database 11 can be updated. The scoring unit 17 instructs the travel command value calculation unit 15 to reversely calculate the driving preference data of the occupant who has replied to update the driving preference database 11 from the travel command value and the like, and the driving preference data that is the reverse calculation result is used as the travel command. The value is received from the value calculation unit 15 and output to the learning unit 18.

  The input / output device 2 presents the results of scoring of each occupant by the scoring unit 17 to each occupant. In addition, the input / output device 2 accepts an update operation on whether or not the driving preference database 11 can be updated by each occupant presenting the scoring result, and outputs information on whether or not the updating is possible to the scoring unit 17.

  The travel command value calculation unit 15 follows the instruction from the scoring unit 17 and stores the destination data, vehicle and surrounding data obtained during the travel of the vehicle, and the travel command value calculated by itself, stored in the scoring unit 17. Is used to calculate the driving preference data of the occupant who has replied that the driving preference database 11 is updated. The reversely calculated driving preference data is teacher data to be learned by the learning unit 18. This teacher data is output to the learning unit 18 via the scoring unit 17.

  The learning unit 18 learns the driving preference data stored in the driving preference database 11 of the occupant who replied to update the driving preference database 11 to be close to the occupant's teacher data calculated backward by the driving command value calculation unit 15. Let Thereby, the driving preference data stored in the driving preference database 11 of the occupant who replied that the driving preference database 11 is updated is updated.

  As a learning method by the learning unit 18, for example, there is a method using Kohonen's LVQ2 net, which is one of neural networks. In the learning method applying the LVQ2 net, the driving preference data stored in the driving preference database 11 is changed so as to approach the teacher data. Note that the learning unit 18 may change a coefficient that indicates the degree to which the driving preference data approaches the teacher data in accordance with the difference value between the learned driving preference data and the teacher data. As a result, even when learning is performed using the same teacher data, a learning result corresponding to the driving preference data of each occupant is obtained.

  Further, the learning unit 18 stores not only the LVQ2 net but also a neural network, and stores a plurality of teacher data for one operation mode, and the plurality of teacher data stored in the past at the time of learning. The driving preference data in the driving preference database 11 may be learned using the teacher data and the current teacher data. Learning accuracy is improved by using a plurality of teacher data.

  FIG. 8 is a flowchart illustrating an operation example of the travel control apparatus 10 according to the third embodiment. The travel control device 10 starts the operation shown in the flowchart of FIG. 8 after the vehicle arrives at the destination.

  In step ST31, the scoring unit 17 reads out the stored destination data, vehicle and surrounding data, and emotion evaluation values of each occupant.

  In step ST32, the scoring unit 17 scores the traveling content for each occupant using the data read in step ST31.

  In step ST <b> 33, the scoring unit 17 uses the input / output device 2 to present each occupant's scoring result to each occupant and inquires whether or not to update the driving preference database 11 based on the current traveling content. By not permitting the occupant who is not satisfied with the scoring result by the scoring unit 17 to update, it is possible to prevent the travel content unfavorable for the occupant from being reflected in the driving preference data of the occupant and to reduce the reliability of the driving preference data. Is prevented.

  In step ST <b> 34, the scoring unit 17 receives, from the input / output device 2, an answer indicating whether or not each driver can update the driving preference database 11. When the scoring unit 17 replies that at least one of the passengers on the vehicle updates the driving preference database 11 ("YES" in step ST34), the scoring unit 17 proceeds to step ST35. On the other hand, the scoring unit 17 ends the operation shown in the flowchart of FIG. 8 when all the occupants in the vehicle reply that the driving preference database 11 is not updated (step ST34 “NO”).

  In step ST35, the scoring unit 17 instructs the travel command value calculation unit 15 to generate teacher data for each occupant who has replied that the driving preference database 11 is updated. Upon receiving this instruction, the travel command value calculation unit 15 uses the destination data, the vehicle and surrounding data obtained during the travel of the vehicle, and the travel command value calculated by itself, stored in the scoring unit 17. The driver's driving preference data (that is, teacher data) that has replied that the driving preference database 11 is updated is reversely calculated and output to the scoring unit 17.

  In step ST <b> 36, the scoring unit 17 outputs the teacher data of each occupant generated by the travel command value calculation unit 15 to the learning unit 18. The learning unit 18 learns the driving preference data stored in the driving preference database 11 of the occupant who replied to update the driving preference database 11 to be close to the occupant's teacher data calculated backward by the driving command value calculation unit 15. Let

  The driving preference data updated by the learning unit 18 is only the driving preference data in the driving mode corresponding to the driving mode selected while the vehicle is traveling. For example, when the driving mode selected while the vehicle is running is “slow” and when Mr. Y replies to update the driving preference database 11, the learning unit 18 is stored in the driving preference database 11. Only “slow” data of the vehicle type of Mr. Y is updated, and other driving preference data are not updated.

  In addition, the traveling control apparatus 10 may start the operation | movement shown by the flowchart of FIG. 8 during driving | running | working of a vehicle. In that case, the scoring unit 17 performs scoring on the traveling content up to the present time. In addition, the scoring unit 17 repeatedly performs scoring while the vehicle is traveling. When the scoring result is extremely low, the scoring unit 17 indicates that the difference between the currently selected operation mode and the actual traveling content is large. It may be used to warn each occupant and confirm whether or not to change the operation mode for each occupant. When the traveling control device 10 receives an instruction to change the operation mode from the occupant during traveling of the vehicle, steps ST12 to ST15 shown in the flowchart of FIG. 4 or steps ST12, ST13, shown in the flowchart of FIG. The operations of ST21 to ST23 are performed.

  As described above, the travel control device 10 according to the third embodiment updates the driving preference database 11 based on the data related to the vehicle, the data related to the periphery of the vehicle, and the travel command value obtained during the travel of the vehicle. A learning unit 18 is provided. Thereby, the traveling control apparatus 10 can improve the precision of the driving preference data stored in the driving preference database 11, and as a result, the driving control with higher satisfaction can be realized.

  Moreover, the traveling control apparatus 10 according to the third embodiment includes a scoring unit 17 for scoring the traveling content of the vehicle. The learning unit 18 updates the driving preference database 11 when an operation for permitting the updating of the driving preference database 11 is performed by each occupant presented with the scoring result of the scoring unit 17. Thereby, the traveling control device 10 can prevent erroneous learning based on the traveling content that does not match the driving preference of the occupant. Therefore, the traveling control device 10 can improve the accuracy of the driving preference data stored in the driving preference database 11, and can realize the driving control with higher satisfaction.

  The travel control device 10 according to the third embodiment has a configuration in which a scoring unit 17 and a learning unit 18 are added to the travel control device 10 according to the second embodiment, but the travel control device according to the first embodiment. 10 may be configured such that a scoring unit 17 and a learning unit 18 are added.

Finally, the hardware configuration of the travel control apparatus 10 according to each embodiment will be described.
FIG. 9 is a diagram illustrating a hardware configuration example of the travel control device 10 according to each embodiment. The driving preference database 11 in the travel control device 10 is a memory 102. The scoring unit 17 stores destination data and the like in the memory 102. The functions of the occupant recognition unit 12, driving preference data extraction unit 13, driving preference data synthesis unit 14, driving command value calculation unit 15, emotion evaluation unit 16, scoring unit 17, and learning unit 18 in the travel control device 10 are processing circuits. It is realized by. That is, the traveling control device 10 includes a processing circuit for realizing the above functions. The processing circuit is a processor 101 that executes a program stored in the memory 102.

  The functions of the occupant recognition unit 12, the driving preference data extraction unit 13, the driving preference data synthesis unit 14, the travel command value calculation unit 15, the emotion evaluation unit 16, the scoring unit 17, and the learning unit 18 are software, firmware, or software. Realized by combination with firmware. Software or firmware is described as a program and stored in the memory 102. The processor 101 reads out and executes a program stored in the memory 102, thereby realizing the function of each unit. That is, the traveling control device 10 includes a memory 102 for storing a program that, when executed by the processor 101, results in the steps shown in the flowchart of FIG. This program also includes procedures or methods for the occupant recognition unit 12, the driving preference data extraction unit 13, the driving preference data synthesis unit 14, the travel command value calculation unit 15, the emotion evaluation unit 16, the scoring unit 17, and the learning unit 18. It can be said that it is what is executed by a computer.

Here, the processor 101 is a CPU (Central Processing Unit), a processing device, an arithmetic device, a microprocessor, or the like.
The memory 102 may be a RAM (Random Access Memory), a ROM (Read Only Memory), an EPROM (Erasable Programmable ROM), or a nonvolatile or volatile semiconductor memory such as a flash memory, a hard disk, a flexible disk, or the like. The magnetic disk may be a CD (Compact Disc) or a DVD (Digital Versatile Disc).

  In the present invention, within the scope of the invention, any combination of each embodiment, any component of each embodiment can be modified, or any component of each embodiment can be omitted.

  In the above description, the driving preference database 11 is mounted on the vehicle, but the driving preference database 11 may be configured on a network. In this case, the traveling control device 10 includes a wireless communication device that performs wireless communication with the driving preference database 11 built on the network. Or the traveling control apparatus 10 may perform radio | wireless communication between the driving preference databases 11 constructed | assembled on the network using the smart phone etc. which were brought into the vehicle.

  1 in-vehicle camera, 2 input / output device, 3 vehicle periphery measuring device, 4 infrared sensor, 5 heart rate sensor, 10 travel control device, 11 driving preference database, 12 occupant recognition unit, 13 driving preference data extraction unit, 14 driving preference data Synthesis unit, 15 travel command value calculation unit, 16 emotion evaluation unit, 17 scoring unit, 18 learning unit, 101 processor, 102 memory.

Claims (7)

A driving preference database storing driving experience and driving preference data according to the occupant and vehicle type;
An occupant recognition unit for recognizing each occupant on the vehicle;
A driving preference data extraction unit that extracts driving experience and driving preference data corresponding to the type of the vehicle on which each of the passengers recognized by the occupant recognition unit from the driving preference database;
A driving preference data synthesizing unit that weights and combines the driving preference data of each occupant extracted by the driving preference data extraction unit with the driving experience level of each occupant;
A travel control device comprising: a travel command value calculation unit that calculates a travel command value for controlling the travel of the vehicle using the drive preference data synthesized by the driving preference data synthesis unit.   The driving preference data synthesizer adjusts the weight of driving preference data of the occupant with a large value of driving experience level, and adjusts the weight of driving preference data of the occupant with a small value of driving experience level, The travel control device according to claim 1, wherein the weight of each occupant after adjustment is corrected according to variation in the weight of each occupant after adjustment. Using the measurement result of each occupant measured by the biometric sensor, the apparatus includes an emotion evaluation unit that calculates the emotion evaluation value of each occupant,
The driving according to claim 1, wherein the driving preference data synthesis unit adjusts the weight of the driving preference data of each occupant according to the emotion evaluation value of each occupant calculated by the emotion evaluation unit. Control device.   2. The learning unit according to claim 1, further comprising a learning unit configured to update the driving preference database based on data relating to the vehicle obtained during the traveling of the vehicle, data relating to the periphery of the vehicle, and the traveling command value. Travel control device. A scoring unit for scoring the traveling content of the vehicle;
The said learning part updates the said driving preference database, when operation which permits the update of the said driving preference database is performed by each said passenger who was shown the scoring result of the said scoring part. The travel control device described. The occupant recognition unit recognizes each occupant on the vehicle,
The driving preference data extracting unit refers to the driving preference database storing the driving experience level and driving preference data according to the occupant and the vehicle type, and the boarding of each occupant recognized by the occupant recognition unit Driving experience level and driving preference data corresponding to the type of vehicle being extracted from the driving preference database;
The driving preference data synthesis unit weights and synthesizes the driving preference data of each occupant extracted by the driving preference data extraction unit with the degree of driving experience of each occupant,
A travel control method in which a travel command value calculation unit calculates a travel command value for controlling the travel of the vehicle using the driving preference data synthesized by the driving preference data synthesis unit. An occupant recognition procedure for recognizing each occupant in the vehicle;
Refer to the driving preference database storing the driving experience level and driving preference data according to the occupant and the vehicle type, and the type of the vehicle on which each occupant is recognized by the occupant recognition procedure. Driving preference data extraction procedure for extracting corresponding driving experience and driving preference data from the driving preference database;
Driving preference data combining procedure for weighting and combining the driving preference data of each occupant extracted by the driving preference data extraction procedure with the driving experience level of each occupant;
A travel control program for causing a computer to execute a travel command value calculation procedure for calculating a travel command value for controlling the travel of the vehicle, using the drive preference data synthesized by the driving preference data synthesis procedure.

Images