WO2018138767A1 - Procédé d'apprentissage de caractéristiques de déplacement et dispositif de commande de conduite - Google Patents

Procédé d'apprentissage de caractéristiques de déplacement et dispositif de commande de conduite Download PDF

Info

Publication number
WO2018138767A1
WO2018138767A1 PCT/JP2017/002283 JP2017002283W WO2018138767A1 WO 2018138767 A1 WO2018138767 A1 WO 2018138767A1 JP 2017002283 W JP2017002283 W JP 2017002283W WO 2018138767 A1 WO2018138767 A1 WO 2018138767A1
Authority
WO
WIPO (PCT)
Prior art keywords
vehicle
driving
inter
distance
learning method
Prior art date
Application number
PCT/JP2017/002283
Other languages
English (en)
Japanese (ja)
Inventor
平松　真知子
Original Assignee
日産自動車株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 日産自動車株式会社 filed Critical 日産自動車株式会社
Priority to PCT/JP2017/002283 priority Critical patent/WO2018138767A1/fr
Publication of WO2018138767A1 publication Critical patent/WO2018138767A1/fr

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
    • B60W30/14—Adaptive cruise control
    • B60W30/16—Control of distance between vehicles, e.g. keeping a distance to preceding vehicle
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • B60W40/08—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to drivers or passengers

Definitions

  • the present invention relates to a driving characteristic learning method for learning driving data during manual driving by a driver in a vehicle capable of switching between manual driving and automatic driving by a driver, and driving in which the learning result is applied to driving characteristics of automatic driving.
  • the present invention relates to a control device.
  • Patent Document 1 is disclosed as an automatic travel control device that learns the operation method of a driver during manual driving and reflects it in the automatic travel control in order to provide automatic travel control according to the driver's preference. ing.
  • the automatic travel control device disclosed in Patent Document 1 the relationship between the vehicle speed and the inter-vehicle distance is learned in consideration of environmental conditions such as road width, brightness, and weather.
  • the present invention has been proposed in view of the above-described circumstances, and an object thereof is to provide a driving characteristic learning method and a driving control device that can accurately learn the inter-vehicle distance that captures the driver's feeling.
  • the driving characteristic learning method and the driving control device give priority to learning the inter-vehicle distance during the deceleration operation in the driver's manual driving.
  • the inter-vehicle distance that captures the driver's feeling can be learned with high accuracy.
  • FIG. 1 is a block diagram showing a configuration of an operation control system including an operation control apparatus according to an embodiment of the present invention.
  • FIG. 2 is a flowchart showing a processing procedure of a travel characteristic learning process by the operation control apparatus according to the embodiment of the present invention.
  • FIG. 3 is a diagram illustrating an example of data input in the travel characteristic learning process according to the embodiment of the present invention.
  • FIG. 4 is a diagram for explaining the coefficients of the multiple regression analysis executed in the travel characteristic learning process according to the embodiment of the present invention.
  • FIG. 5 is a diagram illustrating an example of data indicating a relationship between the vehicle speed and the inter-vehicle distance during the deceleration operation.
  • FIG. 1 is a block diagram showing a configuration of an operation control system including an operation control apparatus according to an embodiment of the present invention.
  • FIG. 2 is a flowchart showing a processing procedure of a travel characteristic learning process by the operation control apparatus according to the embodiment of the present invention.
  • FIG. 3 is a diagram
  • FIG. 6 is a diagram illustrating an example of data indicating the relationship between the vehicle speed and the inter-vehicle distance not only during the deceleration operation but in all cases.
  • FIG. 7 is a diagram for explaining a method for determining the degree of effort by the travel characteristic learning process according to the embodiment of the present invention.
  • FIG. 8 is a diagram for explaining a method for determining the degree of carefulness by the running characteristic learning process according to the embodiment of the present invention.
  • FIG. 9 is a flowchart showing a processing procedure of automatic driving control processing by the driving control device according to the embodiment of the present invention.
  • FIG. 1 is a block diagram illustrating a configuration of an operation control system including an operation control device according to the present embodiment.
  • the driving control system 100 includes a driving control device 1, a driving state detection unit 3, a driving environment detection unit 5, a driving changeover switch 7, and a control state presenting unit 9. It has. Furthermore, the operation control system 100 is connected to an actuator 11 mounted on the vehicle.
  • the driving control device 1 learns driving data during manual driving by the driver in a vehicle that can be switched between manual driving and automatic driving by the driver, and executes processing for applying the learning result to the driving characteristics of the automatic driving. Controller.
  • the driving control device 1 executes driving characteristic learning processing for learning the inter-vehicle distance from the preceding vehicle of the vehicle using the driving data during the deceleration operation in the driver's manual driving with priority.
  • travel characteristic learning process travel data during deceleration operation is selected from travel data during manual operation, and learning is performed using the selected travel data during deceleration operation. That is, learning is performed using only the traveling data during the deceleration operation.
  • the learning is performed in consideration of the traveling data of the distance between the stopped vehicles and the environment information of the environment in which the vehicle is traveling.
  • the driving control device 1 includes a learning data storage unit 21, a travel characteristic learning unit 23, and an automatic driving control execution unit 25.
  • this embodiment demonstrates the case where the driving control apparatus 1 is mounted in a vehicle, you may install a communication apparatus in a vehicle and install the driving control apparatus 1 in an external server.
  • deceleration operation from when the accelerator pedal is turned off until it stops, from when the brake pedal is turned on until it stops, from when the acceleration becomes negative, until it stops, etc. The start time is not questioned.
  • the driving control device 1 When the driving control device 1 is mounted on a vehicle, the driving characteristics of the driver who owns or uses the vehicle can be learned. Further, traveling data for a predetermined period (for example, the latest one month) can be stored and reflected in the automatic driving of the vehicle owned or used by the driver. On the other hand, when it is installed on an external server, it is possible to learn using long-term driving data of the driver himself, so that a more stable learning result can be calculated. In addition, when learning is not completed, the driving data of other drivers can be utilized to reflect the average driving characteristics of the driver in the area in automatic driving.
  • the traveling state detection unit 3 indicates the traveling state of the vehicle such as the vehicle speed, the steering angle, the acceleration, the inter-vehicle distance with the preceding vehicle, the relative speed with the preceding vehicle, the current position, the direction indicator display state, the wiper operating state, and the like. Detect driving data. For example, an in-vehicle network such as CAN (Controller Area Network), a navigation device, a laser radar, a camera, and the like. In particular, the traveling state detection unit 3 detects the operation amount of the brake pedal and the accelerator pedal of the vehicle and the deceleration of the vehicle as data for determining whether or not the driver is decelerating.
  • CAN Controller Area Network
  • the traveling environment detection unit 5 includes the number of lanes on the road on which the vehicle is traveling, the speed limit, the road gradient, the display state of the traffic signal in front of the vehicle, the distance to the intersection in front of the vehicle, the number of vehicles traveling in front of the vehicle, and the intersection in front of the vehicle.
  • Environment information representing an environment in which the vehicle is traveling, such as a scheduled route, is detected.
  • the display state of the traffic light in front of the vehicle may be detected using road-to-vehicle communication, and the number of vehicles traveling in front of the vehicle may be detected using vehicle-to-vehicle communication or a cloud service linked to a smartphone. .
  • the planned course at the intersection in front of the vehicle is obtained from the display state of the navigation device or the direction indicator. Furthermore, the illuminance, temperature, and weather conditions around the vehicle are acquired from the illuminance sensor, the outside temperature sensor, and the wiper switch, respectively. However, the illuminance may be obtained from a headlight switch.
  • the operation changeover switch 7 is a switch that is mounted on the vehicle and is switched between automatic operation and manual operation when operated by a vehicle occupant.
  • a switch installed on the steering of the vehicle.
  • the control state presentation unit 9 displays whether the current control state is manual operation or automatic operation on a meter display unit, a display screen of a navigation device, a head-up display, or the like. In addition, a notification sound that informs the start and end of automatic driving is also output to indicate whether or not learning of driving characteristics has ended.
  • Actuator 11 receives an execution command from operation control device 1 and drives each part such as an accelerator, a brake, and a steering of the vehicle.
  • the learning data storage unit 21 acquires travel data related to the travel state of the vehicle and environmental information related to the travel environment around the vehicle from the travel state detection unit 3 and the travel environment detection unit 5, and stores data necessary for the travel characteristic learning process. To do.
  • the learning data storage unit 21 stores travel data during a deceleration operation that is used for learning the inter-vehicle distance during manual driving. At this time, the learning data storage unit 21 stores the traveling data during the deceleration operation in association with the traveling state and traveling environment of the vehicle.
  • the inter-vehicle distance when stopped is followed.
  • Store data such as duration.
  • environmental information is stored.
  • the environmental information includes the number of lanes on the road on which the vehicle is traveling, the speed limit, the road gradient or traffic light display status, the distance from the vehicle to the intersection, the number of vehicles ahead of the vehicle, the display status of the direction indicator, and the weather around the vehicle. Temperature or illuminance.
  • the driving characteristic learning unit 23 reads the driving data stored in the learning data storage unit 21 and learns the driving characteristic of the vehicle in consideration of the influence state from the driving state and the driving environment.
  • the driving data during the deceleration operation in the driver's manual driving is preferentially used to learn the inter-vehicle distance from the preceding vehicle among the driving characteristics of the vehicle.
  • the travel characteristic learning unit 23 selects travel data during the deceleration operation from travel data during the manual operation, and learns using the selected travel data during the deceleration operation. That is, the inter-vehicle distance from the preceding vehicle is learned using only the traveling data during the deceleration operation.
  • learning is performed in consideration of travel data on the distance between the vehicles being stopped and environmental information of the environment in which the vehicle is traveling. Further, the travel characteristic learning unit 23 learns for each trip of the vehicle. Further, the driving style of the driver may be determined based on the learning result of the inter-vehicle distance from the preceding vehicle. The learning result calculated in this way is stored in the running characteristic learning unit 23 as needed.
  • the automatic operation control execution unit 25 executes automatic operation control when an automatic operation section is entered or when the driver selects automatic operation using the operation changeover switch 7. At this time, the automatic driving control execution unit 25 applies the learning result learned by the driving characteristic learning unit 23 to the driving characteristic of automatic driving. In particular, the learning result of the inter-vehicle distance from the preceding vehicle is applied to the inter-vehicle distance during automatic driving.
  • the operation control device 1 includes a general-purpose electronic circuit including a microcomputer, a microprocessor, and a CPU, and peripheral devices such as a memory. And by operating a specific program, it operates as the above-described learning data storage unit 21, travel characteristic learning unit 23, and automatic driving control execution unit 25.
  • Each function of the operation control apparatus 1 can be implemented by one or a plurality of processing circuits.
  • the processing circuit includes a programmed processing device such as, for example, a processing device including an electrical circuit, and an application specific integrated circuit (ASIC) or conventional circuit arranged to perform the functions described in the embodiments. It also includes devices such as parts.
  • step S ⁇ b> 101 the learning data storage unit 21 determines whether or not the vehicle is in manual operation according to the state of the operation changeover switch 7. If the vehicle is in manual driving, the process proceeds to step S103. If the vehicle is in automatic driving, the driving characteristic learning process is terminated and automatic driving control is executed.
  • the learning data storage unit 21 detects travel data related to the travel state of the vehicle and environmental information related to the travel environment around the vehicle from the travel state detection unit 3 and the travel environment detection unit 5.
  • the detected travel data includes vehicle speed, steering angle, acceleration, deceleration, inter-vehicle distance from the preceding vehicle, relative speed with the preceding vehicle, current position, planned route at the front intersection, brake pedal and accelerator pedal operation amount, The duration of following the preceding vehicle, the operating state of the wiper, etc. are detected.
  • the environmental information includes the number of lanes of the road on which the vehicle is traveling, the speed limit, the road gradient or the display state of the traffic light, the distance from the vehicle to the intersection, the number of vehicles ahead of the vehicle, the display state of the vehicle direction indicator, the vehicle Detect ambient weather, temperature, illuminance, etc.
  • step S105 the learning data storage unit 21 determines whether or not the current vehicle is being decelerated or stopped.
  • a method of determining whether or not the vehicle is decelerating it is determined that the vehicle is decelerating when the deceleration operation is performed, for example, when the brake pedal operation is ON or when the accelerator pedal operation is OFF. . Alternatively, it may be determined that the vehicle is decelerating when a deceleration greater than a predetermined value is generated in the vehicle.
  • the method for determining whether or not the vehicle is stopped determines that the vehicle is stopped when the vehicle speed is zero. If it is determined that the vehicle is decelerating or stopped, the process proceeds to step S107. If it is determined that the vehicle is not decelerating or stopped, the process returns to step S103.
  • the learning data storage unit 21 determines whether or not the current running state of the vehicle matches the learning condition.
  • the learning condition is a condition for determining whether or not the current driving state is appropriate for acquiring data used for learning of driving characteristics.
  • the learning conditions (A) and (B) it is possible to exclude data in an excessive state immediately after the interruption of the preceding vehicle or immediately after leaving, and to apply the learning condition (C).
  • the learning condition (D) it is possible to exclude data targeted for vehicles other than the preceding vehicle existing in the intersection or ahead of the intersection while the vehicle is stopped. Therefore, by setting these learning conditions (A) to (D), the driving characteristics can be learned using the driving data when the vehicle is in a stable condition.
  • the learning condition (E) the driving characteristics with higher accuracy can be learned by setting the learning condition (E) at a place where the driver is likely to adjust the inter-vehicle distance sensitively. Therefore, the learning condition (E) may not always be applied, and may be applied only when it is desired to improve learning accuracy. Further, these learning conditions are not necessarily applied, and may not be applied depending on the situation.
  • step S109 the learning data storage unit 21 stores the travel data and environment information detected in step S103 and selected in the processes in steps S105 and 107 as learning data.
  • the data is stored after being selected in advance has been described. However, after all the data during manual operation is stored once, the above-described steps S105 and 107 may be performed for selection. Good.
  • stores one data about one stop. This is to prevent the same data from being stored repeatedly.
  • FIG. 3 an example of the learning data stored in the learning data storage unit 21 is shown in FIG.
  • data of the inter-vehicle distance D during the deceleration operation, the vehicle speed V during the deceleration operation, x1 to x7, and y1 are recorded.
  • x1 to x7 and y1 are data set based on the environment information, and a value of 0 or 1 is set according to the setting method shown in FIG.
  • x1 is set to 1 when the vehicle is traveling on a road with two or more lanes on one side when data on the inter-vehicle distance D and speed V shown in FIG. 0 is set when driving.
  • the speed limit may be used instead of the number of lanes.
  • 1 is set when the speed limit of the road on which the vehicle is traveling is lower than a predetermined value (40 or 50 km / h), and 0 is set when the speed limit is equal to or higher than the predetermined value.
  • x2 is set to 1 when the vehicle is traveling on an uphill, 0 is set otherwise (flat road and downhill), and x3 is set when the traffic light ahead of the vehicle is a red signal. Is set to 1; otherwise, 0 is set (blue light or no traffic light). However, a yellow signal may be included in the red signal.
  • x4 is set to 1 when the distance from the vehicle to the intersection is less than a predetermined value J [m]
  • 0 is set when the distance is not less than the predetermined value J [m]
  • x5 is set to L in front of the vehicle. 1 is set when there are N or more vehicles within the predetermined value [m], and 0 is set when there are N-1 or less vehicles.
  • the degree of congestion may be determined using VICS (registered trademark) information.
  • x6 is set to 1 when the turn indicator for turning right or left of the vehicle is ON, and is set to 0 when it is OFF.
  • y1 is set to 1 when the distance to the stop line is equal to or greater than a predetermined value K [m] while the vehicle is stopped, and is set to 0 when the distance is less than the predetermined value K [m].
  • 1 may be set when the weather around the vehicle is bad, and 0 may be set when the weather is not bad.
  • a method for determining whether or not the weather is bad when the wiper of the vehicle is set to OFF or intermittent, it is determined that the weather is not bad, and when it is ON, it is determined that the weather is bad.
  • conditions such as temperature and illuminance may be added.
  • the temperature is set to 1 when the outside air temperature sensor is negative, and is set to 0 when it is positive. Thereby, it can respond to the difference in the characteristic by road surface freezing.
  • the illuminance may be set to 1 when the illuminance sensor is bright and 0 when it is dark. Further, it may be set depending on whether or not the headlight is turned on.
  • the case of classifying into two levels of 0 or 1 is described, but it may be classified into three or more levels.
  • the environmental information of x1 to x6 and y1 is associated with the travel data of the inter-vehicle distance D during the deceleration operation and the vehicle speed V during the deceleration operation. Therefore, in the present embodiment, learning is performed using travel data of the inter-vehicle distance D during the deceleration operation and the vehicle speed V during the deceleration operation, and further, the travel characteristics are learned by associating the environment in which the vehicle is traveling with the inter-vehicle distance. can do.
  • FIG. 5 shows an example of data indicating the relationship between the vehicle speed and the inter-vehicle distance during the deceleration operation
  • FIG. 6 shows not only the case where the process of step S105 is not performed, that is, not only during the deceleration operation.
  • An example of data showing the relationship between all vehicle speeds and inter-vehicle distances is shown.
  • FIG. 6 when the speed is not limited during the deceleration operation, the data varies widely. Therefore, even if the relationship between the vehicle speed and the inter-vehicle distance is learned, the learning accuracy cannot be improved.
  • the driver positively adjusts the inter-vehicle distance, so that data variation is suppressed as shown in FIG.
  • the inter-vehicle distance that matches the driver's feeling can be learned with high accuracy, and the learning accuracy can be improved.
  • step S111 the learning data storage unit 21 determines whether or not a predetermined amount of learning data has been stored. If the predetermined amount is not reached, the process returns to step S103. Proceed to step S113.
  • step S113 the travel characteristic learning unit 23 learns the travel characteristics of the vehicle.
  • the inter-vehicle distance from the preceding vehicle is learned among the travel characteristics using travel data during the deceleration operation in the manual operation of the driver.
  • a multiple regression model represented by the following equation (1) is created and learned using a data set as shown in FIG.
  • Vf is the current vehicle speed
  • Df is the inter-vehicle distance from the preceding vehicle calculated from the model.
  • x1 to x7 and y1 are environmental factors
  • a0 to a6, b0 and b1 are coefficients obtained by learning.
  • the term (a0 to a6 ⁇ 6) in the equation (1) is the time to the preceding vehicle while traveling (the vehicle head time, but the time to the position obtained by subtracting the stop inter-vehicle distance).
  • (b0 + b1y1) is the distance between the stopped vehicles, and is the distance between the vehicle and the preceding vehicle when the vehicle speed becomes zero.
  • the multiple regression model represented by the equation (1) indicates that the inter-vehicle distance from the preceding vehicle and the inter-vehicle distance at the time of stop vary depending on environmental factors.
  • a0 is a reference value set for each trip, and the average value of the vehicle head time within the trip when the values of x1 to x6 are 0 It is.
  • b0 is a reference value set for each driver, and is the inter-vehicle distance when the vehicle stops when the value of y1 is zero. For example, an average value of the inter-vehicle distance at the time of stopping may be used.
  • the running characteristic learning unit 23 performs a multiple regression analysis using learning data as shown in FIG. 3, and calculates coefficients of a0 to a6, b0, and b1 in Expression (1). Since the learning data used here is only the driving data during the deceleration operation by the driver as shown in FIG. 5, the variation is suppressed, and as a result, the distance between the preceding vehicle calculated from the equation (1) The distance Df is a straight line F in FIG. As described above, in the present embodiment, since the inter-vehicle distance from the preceding vehicle is learned using only the traveling data during the deceleration operation, the inter-vehicle distance that matches the driver's feeling can be learned with high accuracy.
  • learning can be performed in consideration of environmental information of the environment in which the vehicle is traveling according to the terms a1x1 to a6x6. That is, the correction can be made based on the environmental information. Furthermore, in the present embodiment, learning can be performed in consideration of traveling data of the distance between the vehicles being stopped by b0 + b1y1. That is, it can correct
  • the learning data may use a plurality of trip data, or may use only one trip data. If environmental factor data is not available with only one trip, environmental factor coefficients are calculated using multiple trip learning data, and reference a0 and b0 coefficients are the learning data in the trip. You may calculate using. In this case, it is possible to provide a learning result with no sense of incongruity even when the trip of the day tends to be slow or rushed compared to other trips.
  • the distance between vehicles while traveling and the distance between vehicles when stopped may have different characteristics for each trip. For example, when you are in a hurry to the destination, want to drive slowly, or when a passenger is present, the mood and conditions when driving may differ. Therefore, by performing multiple regression analysis for each trip, the characteristics of the inter-vehicle distance for each trip can be obtained. Furthermore, by performing inter-vehicle distance control during automatic driving with the characteristics of inter-vehicle distance learned for each trip, it is possible to provide automatic driving control that matches the driver's mood and conditions during the trip.
  • the inter-vehicle distance Df with the preceding vehicle in equation (1) is This value is larger than when driving on a road with two or more lanes. Therefore, when the vehicle is traveling on a road with one lane or less on one side, the inter-vehicle distance Df with the preceding vehicle is corrected to be longer than when the vehicle is traveling on a road with two or more lanes.
  • the speed limit may be used instead of the number of lanes, when the speed limit of the road on which the vehicle is traveling is equal to or higher than a predetermined value, the speed limit is lower than that of the preceding vehicle. The inter-vehicle distance Df is corrected to be longer.
  • the inter-vehicle distance Df with the preceding vehicle in Expression (1) is other than the red signal. A larger value than in the case of. Therefore, when the traffic light ahead of the vehicle is a red signal, the inter-vehicle distance Df from the preceding vehicle is corrected to be longer than when the signal is not a red signal.
  • x4 in equation (1) is 1 and a4 is a positive value, so the inter-vehicle distance Df with the preceding vehicle in equation (1) is The distance to the intersection is larger than when the distance is equal to or greater than a predetermined value. Therefore, when the distance to the intersection ahead of the vehicle is less than the predetermined value, the inter-vehicle distance Df with the preceding vehicle is corrected to be longer than when the distance is greater than or equal to the predetermined value.
  • x5 in equation (1) is 1 and a5 is a positive value, so the inter-vehicle distance Df with the preceding vehicle in equation (1) is the vehicle
  • the number is larger than when the number is less than a predetermined value. Therefore, when the number of vehicles ahead of the vehicle is greater than or equal to a predetermined value, the inter-vehicle distance from the preceding vehicle is corrected to be longer than when the number of vehicles is less than the predetermined value.
  • x6 in equation (1) is 1 and a6 is a positive value, so the inter-vehicle distance Df from the preceding vehicle in equation (1) is determined by the direction indicator. It becomes a larger value than when it is OFF. Therefore, when the vehicle direction indicator is ON, the inter-vehicle distance Df from the preceding vehicle is corrected to be longer than when it is OFF. Similarly, when the weather around the vehicle is bad, the inter-vehicle distance Df with the preceding vehicle is corrected to be longer than when the weather is not bad.
  • the inter-vehicle distance Df with the preceding vehicle is corrected to be longer.
  • the inter-vehicle distance from the preceding vehicle in equation (1) Df is a larger value than when the distance to the stop line is less than a predetermined value. Therefore, when the distance to the stop line ahead of the vehicle is greater than or equal to a predetermined value, the inter-vehicle distance Df with the preceding vehicle is corrected to be longer than when the distance is less than the predetermined value.
  • the driving characteristic learning unit 23 may determine the driving style of the driver based on the learning result of the inter-vehicle distance from the preceding vehicle.
  • the characteristics of the inter-vehicle distance may show a tendency to match the individual driving style of the driver.
  • the value of a0 + a1 in equation (1) indicates the characteristics of the inter-vehicle distance of roads with two or more lanes on one side, and reflects the driver's impatient tendency as shown in FIG. Since the value of a1 is negative, the degree of effort increases as the value of a0 + a1 decreases. That is, since there is a tendency to travel favorably on roads with two or more lanes rather than roads with one lane, it can be determined to be impatient.
  • the average value and standard deviation of the sum (b0 + b1) of the coefficients of the inter-vehicle distance at the time of the stop reflect the driver's merit, and the average + standard deviation ( ⁇ ) value or standard
  • the personal driving style determined in this way may be provided to the driver himself, or by using an external server to compare with other drivers, how much of the overall tendency
  • the driver or the manager may be provided with information by determining whether or not the screen has a tendency.
  • step S115 the driving characteristic learning unit 23 stores the calculated coefficients a0 to a6, b0, and b1 of the equation (1) as calculation results, and ends the driving characteristic learning process according to the present embodiment.
  • step S ⁇ b> 201 the automatic driving control execution unit 25 determines whether or not learning of the inter-vehicle distance from the preceding vehicle has been completed by the travel characteristic learning process shown in FIG. 2. If learning has been completed, the process proceeds to step S203, and if learning has not been completed, the process proceeds to step S211.
  • step S ⁇ b> 203 the automatic driving control execution unit 25 detects travel data regarding the travel state of the vehicle and environment information regarding the travel environment around the vehicle from the travel state detection unit 3 and the travel environment detection unit 5.
  • step S205 the automatic driving control execution unit 25 sets the inter-vehicle distance from the preceding vehicle based on the learning result. Specifically, the coefficients of learning results a0 to a6, b0, b1 are set in equation (1), and the detected vehicle speed of the current vehicle is input into equation (1). The distance Df is calculated. Then, the automatic driving control execution unit 25 sets the calculated inter-vehicle distance Df as the inter-vehicle distance applied to the automatic driving. That is, the learning result of the inter-vehicle distance with the preceding vehicle is applied to the inter-vehicle distance during automatic driving.
  • step S207 the automatic driving control execution unit 25 executes the automatic driving control using the set inter-vehicle distance. Specifically, the automatic driving control execution unit 25 transmits a control execution command to the actuator 11 and executes operations such as an accelerator, a brake, and a steering necessary for automatic driving.
  • step S209 the automatic operation control execution unit 25 determines whether or not the automatic operation has ended. If not, the automatic operation control execution unit 25 returns to step S203 and continues the automatic operation. On the other hand, when the automatic operation is switched to the manual operation and the automatic operation is finished, the automatic operation control process according to the present embodiment is finished.
  • step S ⁇ b> 211 the automatic driving control execution unit 25 detects traveling data related to the traveling state of the vehicle and environmental information related to the traveling environment around the vehicle from the traveling state detection unit 3 and the traveling environment detection unit 5.
  • step S213 the automatic driving control execution unit 25 sets a predetermined value set in advance as the inter-vehicle distance from the preceding vehicle.
  • a predetermined value a general inter-vehicle distance value or average value may be used.
  • step S215 the automatic driving control execution unit 25 executes the automatic driving control using the set inter-vehicle distance. Specifically, the automatic driving control execution unit 25 transmits a control execution command to the actuator 11 and executes operations such as an accelerator, a brake, and a steering necessary for automatic driving.
  • step S217 the automatic operation control execution unit 25 determines whether or not the automatic operation has ended. If not, the automatic operation control execution unit 25 returns to step S211 and continues the automatic operation. On the other hand, when the automatic operation is switched to the manual operation and the automatic operation is finished, the automatic operation control process according to the present embodiment is finished.
  • the driving characteristic learning method it is detected whether or not the driver is decelerating from at least one of brake pedal operation, accelerator pedal operation, and vehicle deceleration.
  • the inter-vehicle distance when the driver is decelerating can be acquired with certainty.
  • the brake pedal when the brake pedal is operated, it is a clear deceleration operation, and therefore the inter-vehicle distance with the least variation can be acquired.
  • the inter-vehicle distance when the accelerator pedal is not operated it is possible to acquire the data including the deceleration preparation action data.
  • it is determined that the vehicle is decelerating when the deceleration is equal to or greater than a predetermined value it is possible to detect a deceleration operation in any scene.
  • traveling characteristic learning method learning is performed using traveling data of the vehicle speed during the deceleration operation and the inter-vehicle distance during the deceleration operation. As a result, it is possible to accurately learn the inter-vehicle distance that captures the driver's feeling at each vehicle speed, and to give the driver a sense of security.
  • the traveling characteristic learning method the distance between vehicles when the vehicle is stopped is learned. As a result, it is possible to learn including the inter-vehicle distance during stoppage, so that it is possible to perform learning that captures the driver's senses even during stoppage.
  • the driving characteristic learning method learning is performed in correspondence with the environment in which the vehicle is traveling and the inter-vehicle distance.
  • the inter-vehicle distance during traveling and the inter-vehicle distance during stoppage have different characteristics depending on environmental conditions. Therefore, by learning the environment in which the vehicle is traveling in correspondence with the inter-vehicle distance, it is possible to learn the inter-vehicle distance that captures the driver's feeling in each environment. By applying the learning result to automatic driving, the environmental conditions can be accurately reflected in the inter-vehicle distance during automatic driving.
  • the number of lanes of the road on which the vehicle is traveling, the speed limit, the road gradient, or the display state of the traffic light is used as the environment in which the vehicle is traveling. Further, the distance from the vehicle to the intersection, the number of vehicles in front of the vehicle, the display state of the direction indicator of the vehicle, the weather around the vehicle, the temperature or the illuminance are used. Accordingly, the inter-vehicle distance can be corrected by individually reflecting different environmental conditions.
  • learning is performed for each trip of the vehicle. Since the distance between vehicles while traveling and the distance between vehicles that are stopped may have different characteristics for each trip, learning by each trip provides a distance that reflects the driver's mood and conditions during that trip. be able to.
  • the inter-vehicle distance is learned in which the continuation time during which the vehicle follows the preceding vehicle is a predetermined time or more.
  • learning can be performed by excluding the transitional driving state immediately after interruption of the preceding vehicle or immediately after leaving, so that it is possible to learn accurately using the inter-vehicle distance under stable conditions.
  • the inter-vehicle distance when the absolute value of the relative speed with respect to the preceding vehicle is equal to or less than a predetermined value is learned.
  • the inter-vehicle distance is learned when the absolute value of the vehicle steering angle is equal to or less than a predetermined value.
  • the driving characteristic learning method learning is performed when the inter-vehicle distance while the vehicle is stopped is equal to or less than a predetermined value.
  • the driving characteristic learning method when the learning result is applied to the driving characteristic of automatic driving, when the vehicle is traveling on a road with one lane or less on one side, a road with two or more lanes is selected.
  • the inter-vehicle distance from the preceding vehicle is made longer than when traveling.
  • the speed limit of the road on which the vehicle is traveling is greater than or equal to a predetermined value
  • the speed limit is The inter-vehicle distance with the preceding vehicle is made longer than when it is lower than the predetermined value.
  • the driving characteristic learning method when the learning result is applied to the driving characteristic of the automatic driving, when the vehicle is traveling on the downhill, than when the vehicle is traveling on the uphill, Increase the inter-vehicle distance from the preceding vehicle. As a result, safety can be improved and automatic driving can be executed on a downhill where braking is difficult, so that a sense of security can be given to the driver.
  • the driving characteristic learning method when the learning result is applied to the driving characteristic of automatic driving, when the traffic light ahead of the vehicle is a red signal, the preceding vehicle is more than the case other than the red signal. Increase the inter-vehicle distance. As a result, the safety can be improved and the automatic operation can be executed in the case of a red light that needs to be stopped, so that the driver can feel safe.
  • the driving characteristic learning method when the learning result is applied to the driving characteristic of automatic driving, when the number of vehicles ahead of the vehicle is equal to or greater than a predetermined value, the number of vehicles is less than the predetermined value.
  • the inter-vehicle distance from the preceding vehicle is made longer than in some cases.
  • the driving characteristic learning method when the learning result is applied to the driving characteristic of the automatic driving, when the weather around the vehicle is bad weather, the distance between the preceding vehicle and the preceding vehicle is lower than when the bad weather is not. Increase the distance. Thereby, when the surroundings of the vehicle are in bad weather, the safety can be improved and the automatic driving can be executed, so that it is possible to give the driver a sense of security.
  • the driving control device is installed in an external server to learn the driving characteristic of the vehicle. Thereby, the processing load in the vehicle can be reduced.
  • the driving style of the driver is determined based on the learning result of the inter-vehicle distance from the preceding vehicle. Therefore, since a qualitative tendency of the driver can be known, safety can be improved by referring to the manual driving. Further, when there are a plurality of control modes such as a sports mode, an eco mode, and an elderly person mode, an appropriate control mode may be selected with reference to this driving style.
  • the learning result of the inter-vehicle distance from the preceding vehicle is applied to the inter-vehicle distance during automatic driving of the vehicle.

Landscapes

  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Control Of Driving Devices And Active Controlling Of Vehicle (AREA)

Abstract

L'invention concerne un procédé d'apprentissage de caractéristiques de déplacement qui apprend de préférence la distance entre véhicules pendant une opération de décélération pendant une conduite manuelle par un conducteur dans un véhicule qui peut être commuté par le conducteur entre une conduite manuelle et une conduite automatique.
PCT/JP2017/002283 2017-01-24 2017-01-24 Procédé d'apprentissage de caractéristiques de déplacement et dispositif de commande de conduite WO2018138767A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/JP2017/002283 WO2018138767A1 (fr) 2017-01-24 2017-01-24 Procédé d'apprentissage de caractéristiques de déplacement et dispositif de commande de conduite

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2017/002283 WO2018138767A1 (fr) 2017-01-24 2017-01-24 Procédé d'apprentissage de caractéristiques de déplacement et dispositif de commande de conduite

Publications (1)

Publication Number Publication Date
WO2018138767A1 true WO2018138767A1 (fr) 2018-08-02

Family

ID=62978122

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2017/002283 WO2018138767A1 (fr) 2017-01-24 2017-01-24 Procédé d'apprentissage de caractéristiques de déplacement et dispositif de commande de conduite

Country Status (1)

Country Link
WO (1) WO2018138767A1 (fr)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2021037794A (ja) * 2019-08-30 2021-03-11 株式会社デンソー 車両制御装置
JP2021084502A (ja) * 2019-11-27 2021-06-03 株式会社Ｓｕｂａｒｕ 制御装置
US11650067B2 (en) 2019-07-08 2023-05-16 Toyota Motor North America, Inc. System and method for reducing route time using big data

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0689400A (ja) * 1992-09-08 1994-03-29 Fujitsu Ten Ltd 警報車間距離制御装置
JPH07108849A (ja) * 1993-10-13 1995-04-25 Hitachi Ltd 車の自動走行制御装置
JP2013068115A (ja) * 2011-09-21 2013-04-18 Toyota Motor Corp 車両用情報処理装置及び車両用情報処理方法
WO2015166721A1 (fr) * 2014-05-02 2015-11-05 エイディシーテクノロジー株式会社 Dispositif de commande de véhicule

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0689400A (ja) * 1992-09-08 1994-03-29 Fujitsu Ten Ltd 警報車間距離制御装置
JPH07108849A (ja) * 1993-10-13 1995-04-25 Hitachi Ltd 車の自動走行制御装置
JP2013068115A (ja) * 2011-09-21 2013-04-18 Toyota Motor Corp 車両用情報処理装置及び車両用情報処理方法
WO2015166721A1 (fr) * 2014-05-02 2015-11-05 エイディシーテクノロジー株式会社 Dispositif de commande de véhicule

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11650067B2 (en) 2019-07-08 2023-05-16 Toyota Motor North America, Inc. System and method for reducing route time using big data
JP2021037794A (ja) * 2019-08-30 2021-03-11 株式会社デンソー 車両制御装置
JP2021084502A (ja) * 2019-11-27 2021-06-03 株式会社Ｓｕｂａｒｕ 制御装置
JP7414490B2 (ja) 2019-11-27 2024-01-16 株式会社Subaru 制御装置

Similar Documents

Publication Publication Date Title
WO2018163288A1 (fr) Procédé d'aide à la conduite et dispositif de commande de conduite
CA3025643C (fr) Procede de commande de conduite et dispositif de commande de conduite
JP6645587B2 (ja) 運転支援方法及び運転支援装置
US11097730B2 (en) Implicit activation and control of driver assistance systems
US10850740B2 (en) Driving assistance method and driving assistance device
WO2018193765A1 (fr) Dispositif de commande de présentation, dispositif de commande de conduite automatisée, procédé de commande de présentation et procédé de commande de conduite automatisée
US20210163015A1 (en) Method for Learning Travel Characteristics and Travel Assistance Device
JPWO2019003280A1 (ja) 車両の走行支援方法及び走行支援装置
WO2018138767A1 (fr) Procédé d'apprentissage de caractéristiques de déplacement et dispositif de commande de conduite
JP2018092525A (ja) 運転支援装置及び運転支援方法
JP6756402B2 (ja) 走行制御方法及び運転制御装置
US20240062590A1 (en) Method for executing a test drive with at least one test vehicle
US20210122396A1 (en) Vehicle control device
WO2022091308A1 (fr) Procédé et dispositif de commande de conduite
WO2019021430A1 (fr) Procédé de commande de déplacement et dispositif de commande de déplacement
JP2019172167A (ja) 自動運転システムおよび自動運転プログラム
WO2024105863A1 (fr) Procédé et dispositif de commande de conduite

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17894242

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17894242

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: JP