WO2017056454A1 - Dispositif de réduction de dommages, procédé de réduction de dommages et programme - Google Patents

Dispositif de réduction de dommages, procédé de réduction de dommages et programme Download PDF

Info

Publication number
WO2017056454A1
WO2017056454A1 PCT/JP2016/004283 JP2016004283W WO2017056454A1 WO 2017056454 A1 WO2017056454 A1 WO 2017056454A1 JP 2016004283 W JP2016004283 W JP 2016004283W WO 2017056454 A1 WO2017056454 A1 WO 2017056454A1
Authority
WO
WIPO (PCT)
Prior art keywords
collision
action
mobile device
data
predicted
Prior art date
Application number
PCT/JP2016/004283
Other languages
English (en)
Japanese (ja)
Inventor
小柳津　秀紀
雄飛 近藤
康孝 平澤
卓 青木
健人 赤間
Original Assignee
ソニー株式会社
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 ソニー株式会社 filed Critical ソニー株式会社
Publication of WO2017056454A1 publication Critical patent/WO2017056454A1/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
    • B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
    • B60W30/09—Taking automatic action to avoid collision, e.g. braking and steering
    • 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
    • B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
    • B60W30/095—Predicting travel path or likelihood of collision
    • 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
    • B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
    • B60W2030/082—Vehicle operation after collision

Definitions

  • This technology relates to an apparatus, a method, and a program for reducing the damage of a collision accident that occurs, for example, when driving a car.
  • the present technology can be applied not only to automobiles but also to various mobile devices such as ships, airplanes, and robots, and also to various technical fields such as simulation devices and game machines.
  • Patent Document 1 discloses a technique for controlling a host vehicle such that when a collision between the host vehicle and a three-dimensional object is inevitable, the host vehicle collides at a place where the passenger's damage is reduced from the boarding state of the host vehicle.
  • Japanese Patent Application Laid-Open No. 2004-260688 discloses a technique in which when a collision is inevitable, the brake is applied without waiting for the driver's judgment, but when the driver is performing an avoidance operation, braking is not performed so as not to disturb the driver's operation. Is disclosed.
  • Japanese Patent Application Laid-Open No. 2004-228688 discloses a technique that prevents sensitive movement by changing the level of the warning sound and reducing the influence of the driver's braking / steering operation according to the state of the driver.
  • Patent Documents 1 to 3 are such that when the driver is performing an avoidance operation, the system does not take a risk avoidance action. Therefore, when the driver performs an avoidance operation, the driver is responsible for the avoidance action. There was a risk that collision damage would increase.
  • an object of the present technology is to provide a damage reduction device, a damage reduction method, and a program that can reliably reduce damage caused by a collision.
  • a damage reduction apparatus inputs input of status data related to a situation of a moving direction of a mobile device and operation data related to a moving operation of an operator who operates the mobile device. And a prediction unit that predicts a collision with the object in the moving direction based on the situation data, and when the collision is predicted, the mobile device avoids the predicted collision based on the situation data.
  • a candidate calculation unit that calculates a plurality of behavior candidates for the selection, and a selection unit that selects one behavior corresponding to the operation data when the collision is predicted from the plurality of behavior candidates.
  • the damage reduction device when a collision is predicted, a plurality of action candidates for avoiding the predicted collision based on the situation data regarding the situation of the moving direction of the mobile device are calculated, Since one action corresponding to the operation data when a collision is predicted is selected from a plurality of action candidates, that is, reflecting the operator's intention, the collision is finally determined by the judgment on the device side. Since actions to avoid are selected, damage caused by collision can be surely reduced.
  • the candidate calculation unit may calculate an evaluation value obtained by quantifying damage predicted for the action for each of the plurality of action candidates. Thereby, the calculated action can be appropriately selected.
  • the selection unit selects one action from the plurality of action candidates based on the evaluation value when the operation data is not input when the collision is predicted. May be.
  • the selection unit when the plurality of action candidates do not have an action according to the operation data when the collision is predicted, based on the evaluation value.
  • One action may be selected from a plurality of action candidates.
  • the candidate calculation unit is a candidate in which the selection unit selects an action candidate whose evaluation value is equal to or greater than a predetermined threshold value from the plurality of calculated action candidates. You may leave as. Thereby, since the avoidance action of the evaluation value below a certain level may cause damage, it is possible to prevent such an action from being selected.
  • the damage reduction device may include an output unit that outputs the action data selected by the selection unit to a control unit that controls a movement operation of the mobile device.
  • the input unit inputs braking, acceleration, and steering data of the mobile device as the operation data
  • the candidate calculation unit includes the object in the movement direction.
  • one or more collisions are predicted, one or more objects to be avoided are listed based on the inputted braking, acceleration, and steering data, and the mobile device and the corresponding object are listed for each listed object.
  • To calculate the distance to the object and the relative approach speed between the mobile device and the object, and to avoid collision with these objects as candidates for the plurality of actions based on the calculated distance and speed A plurality of candidates for braking, acceleration and steering of the mobile device are calculated, and the selection unit is configured to input the input unit through the input unit when the collision is predicted from the calculated braking, acceleration and steering candidates.
  • the mobile body The braking, acceleration, and steering of the mobile device for avoiding a collision with the object according to the braking, acceleration, and steering data input from is selected, and the damage reduction device is selected by the selection unit.
  • an output unit that outputs braking, acceleration, and steering data of the mobile device for avoiding a collision with the object to a control unit that controls braking, acceleration, and steering of the mobile device. May be.
  • the action for avoiding the collision is finally selected according to the judgment of the apparatus while reflecting the will of the operator, so that the damage caused by the collision can be surely reduced.
  • a damage reduction method inputs situation data related to a situation in a movement direction of a mobile device, predicts a collision with an object in the movement direction based on the inputted situation data, and the collision is When predicted, the mobile device calculates a plurality of behavior candidates for avoiding the predicted collision based on the situation data, and operates the mobile device when the collision is predicted.
  • the operation data related to the person's movement operation is input, and one action corresponding to the operation data is selected from the plurality of action candidates.
  • the action for avoiding the collision is finally selected according to the judgment of the apparatus while reflecting the will of the operator, so that the damage caused by the collision can be surely reduced.
  • a program includes a step of inputting situation data relating to a situation in a movement direction of a mobile device to a computer, and a step of predicting a collision with an object in the movement direction based on the inputted situation data. And when the collision is predicted, the mobile device calculates a plurality of behavior candidates for avoiding the predicted collision based on the situation data, and when the collision is predicted, A step of inputting operation data related to a moving operation of an operator who operates the mobile device and a step of selecting one action corresponding to the operation data from the plurality of action candidates are executed. Thereby, the action for avoiding the collision is finally selected by the determination on the apparatus side while reflecting the will of the operator, so that the damage caused by the collision can be surely reduced.
  • FIG. 1 is a block diagram illustrating a configuration of an automobile including a damage reduction device according to an embodiment of the present technology. It is a figure for demonstrating the outline of the operation
  • crew situation in a vehicle by the damage reduction apparatus shown in FIG. 3 is a flowchart illustrating an operation of a driver intention reflection unit illustrated in FIG. 2. It is a flowchart which shows operation
  • FIG. 1 is an external view illustrating a configuration of an automobile as an example of a mobile device on which a damage reduction device according to an embodiment of the present technology is mounted.
  • FIG. 2 is a block diagram thereof.
  • the automobile 100 includes a distance sensor 101 for the moving direction and a front camera front 102.
  • the distance sensor 101 is installed, for example, in the approximate center of the front part of the automobile 100 and outputs data relating to the distance between the automobile 100 and an object existing in the moving direction to the damage reduction apparatus 1. As will be described later, the output of the distance sensor 101 is referred to when calculating a relative distance, a relative speed, or a relative acceleration with respect to an object (a vehicle, a pedestrian, a structure, etc.) existing in front of the host vehicle.
  • the distance sensor 101 is composed of various sensors using, for example, a millimeter wave radar or an infrared laser.
  • the front camera 102 is installed, for example, in the cabin or roof portion of the automobile 100, and photographs the front view of the automobile 100 at a predetermined frame rate. Image data photographed by the front camera 102 is output to the damage reduction apparatus 1, and as will be described later, the type of an object (a vehicle, a pedestrian, a structure, etc.) existing in front of the own vehicle is determined, and It is referred to when calculating the relative position to the car.
  • the front camera 102 is composed of an image sensor such as a CMOS or a CCD, for example.
  • the in-vehicle camera 103 is installed in the cabin of the automobile 100, and images the interior of the cabin at a predetermined frame rate.
  • the image data photographed by the in-vehicle photographing camera 103 is output to the damage reduction apparatus 1 and is referred to in the determination of the presence / absence of the passenger in the own vehicle and the boarding position thereof, as will be described later.
  • the in-vehicle shooting camera 103 is composed of, for example, an image sensor such as a CMOS or a CCD.
  • the distance sensor 101, the front camera 102, and the in-vehicle camera 103 are configured so that their outputs are supplied to the control unit 107 instead of being supplied to the damage reduction apparatus 1 as shown in FIG. May be.
  • Steering device 104 is typically composed of a power steering device, and transmits the steering wheel operation of driver 110 to the steered wheels.
  • the braking device 105 includes a brake operating device attached to each wheel and a hydraulic circuit that operates them, and transmits a depressing operation force of the brake pedal to the brake operating device via the hydraulic circuit.
  • the braking device 105 typically has an ABS control function for preventing wheel lock (slip) and a traction control function for preventing drive slip of the drive wheels.
  • the vehicle body acceleration device 106 includes a throttle valve, a fuel injection device, and the like, and controls the rotational acceleration of the drive wheels.
  • the control unit 107 controls the steering device 104, the braking device 105, and the vehicle body acceleration device 106. That is, the control unit 107 detects the steering amount and the steering direction based on the output of the steering angle sensor 1041 that detects the steering operation of the driver 110 and controls the steering device 104. Further, the control unit 107 calculates the vehicle body speed of the vehicle based on the output of the wheel speed sensor 1051 installed on all wheels or a part of the wheels, and the braking device so as to prevent the wheels from being locked (slip). 105 is controlled.
  • the brake switch 1052 is for detecting a brake operation (depressing the brake pedal) of the driver 110, and is referred to during ABS control or the like. Furthermore, the control unit 107 controls the vehicle acceleration device 106 based on the output of the accelerator sensor 1061 that detects the accelerator pedal operation amount of the driver 110.
  • the control unit 107 may coordinately control a plurality of these units as well as individually controlling the steering device 104, the braking device 105, and the vehicle body acceleration device 106. As a result, the vehicle 100 can be controlled to a desired posture during turning, braking, acceleration, and the like.
  • control unit 107 is configured to be able to control the steering device 104, the braking device 105, and the vehicle body acceleration device 106 irrespective of the various operations of the driver 110 described above.
  • control unit 107 is configured to be able to control at least one of the above devices based on an output of a damage reduction device 1 described later.
  • the control unit 107 may be a collection of ECUs that individually control the steering device 104, the braking device 105, and the vehicle body acceleration device 106, or may be a single controller that comprehensively controls these devices. May be. Further, the steering device 104, the braking device 105, and the vehicle body acceleration device 106 may individually include the ECU. In this case, the control unit 107 is configured to individually output a control signal to the ECU of each device.
  • the measurement data by the distance sensor 101 and the photographing data by the front camera front 102 are input to the damage reduction apparatus 1 as the situation data regarding the situation in the moving direction of the automobile 100.
  • the damage reduction device 1 reflects the intention of the driver 110 by listing candidates for damage reduction processing in an emergency with a collision, and selecting and executing a process close to the action selected by the driver 110 from the list. On the other hand, appropriate coping actions are taken to reduce collision damage.
  • the damage reduction device 1 calculates braking, acceleration, and steering processes for avoiding the possibility that the automobile 100 collides with the object 1000, and avoids the avoidance route 1001 and List 1002.
  • the selection of the driver 110 is assigned to the steering angle by the steering wheel operation, and the right turn, straight advance, and left turn can be selected.
  • the driver 110 steers the steering wheel turned to the right by the determination, the driver 110 can detect the intention by the rudder angle sensor 1041 and can transmit the intention to the damage reduction apparatus 1 via the control unit 107.
  • the damage reduction apparatus 1 selects and executes the avoidance route 1002 that is avoidance in the right direction among the avoidance routes 1001 and 1002.
  • the execution of the avoidance route 1002 in this case is not performed by the operation of the driver 110, but the control unit 107 controls the steering by the steering device 104 so as to pass through the avoidance route 1002 regardless of the steering operation by the driver 110. .
  • the driver 110 can select the driver 110 by detecting, for example, depression of the brake pedal or depression of the accelerator pedal via the brake switch 1052 or the accelerator sensor 1061. It is also possible to detect the driver's intention to select by analyzing the operation of the driver 110 based on the shooting data of the in-vehicle shooting camera 103 for shooting the passenger situation in the vehicle.
  • the damage reduction apparatus 1 includes an input unit 10, a prediction unit 20, a candidate calculation unit 30, a driver intention reflection unit 40, a selection unit 50, and an in-vehicle occupant.
  • a situation grasping unit 60 and an output unit 70 are provided.
  • the input unit 10 inputs status data related to the status of the moving direction of the automobile 100, operation data related to the drive operation of the driver 110, and occupant status data in the automobile 100.
  • the situation data is, for example, the measurement data of the distance sensor 101 and the shooting data of the front camera 102
  • the operation data is, for example, the steering operation of the driver 110 detected by the steering angle sensor 1041, the brake switch 1052, and the accelerator sensor 1061, and the brake operation. It is data regarding the depression and the depression of the accelerator.
  • the in-vehicle situation data is, for example, shooting data from the in-vehicle shooting camera 103.
  • the input unit 10 includes a distance sensor data input unit 11 that inputs data from the distance sensor 101, a front camera data input unit 12 that inputs data from the front camera 102, and in-vehicle shooting that inputs data from the in-vehicle shooting camera 103.
  • a camera data input unit 13 and an operation data input unit 14 for inputting operation data from the control unit 107 are provided.
  • the prediction unit 20 predicts a collision with the object 1000 in the movement direction of the automobile 100 based on the situation data regarding the situation in the movement direction of the automobile 100.
  • the object 1000 refers to an object that can exist on the road, such as a person, a trash can, a car, a traffic light, a guardrail, and the like.
  • the prediction unit 20 inputs situation data regarding the situation of the moving direction of the automobile 100 from the distance sensor 101 and the front camera 102 via the input unit 10, and calculates whether the object 1000 exists in the moving direction, If it exists, the possibility of a collision between the automobile 100 and the object 1000 is calculated from the distance to the object 1000, the relative speed, the relative acceleration, the size of the object 1000, and the like.
  • the prediction unit 20 estimates that there is a collision with the object 1000 and determines that an emergency avoidance action is necessary, the prediction unit 20 instructs the candidate calculation unit 30 to create an avoidance action classification table.
  • the candidate calculation unit 30 calculates a plurality of action candidates for avoiding a collision predicted based on the situation data regarding the situation of the moving direction of the automobile 100 and the situation data in the car. In addition, an evaluation value obtained by quantifying damage predicted for the action is calculated for each of the plurality of action candidates. For example, the candidate calculation unit 30 receives an instruction from the prediction unit 20, and receives status data from the distance sensor 101 and the front camera 102 via the input unit 10 regarding the situation of the moving direction of the automobile 100 and the in-vehicle occupant situation grasping unit 60.
  • the following is a list of actions for braking, acceleration, and steering that can integrate situation data in the car and avoid collisions or reduce damage, and table the avoidance actions that consist of these processes and their respective evaluation values.
  • the candidate calculation unit 30 creates an avoidance action classification table (to be described later) based on the generated avoidance action table and sends the avoidance action classification table to the selection unit 50.
  • the driver intention reflecting unit 40 is a control unit for operating data regarding the steering wheel operation of the driver 110, the brake depression amount, and the accelerator depression amount detected by the steering angle sensor 1041, the brake switch 1052, and the accelerator sensor 1061 via the input unit 10.
  • the intention of the driver 110 is detected from the operation data, and the intention is reflected in the selection operation by the damage reduction apparatus 1.
  • the driver intention reflecting unit 40 is “right” when the steering operation by the driver 110 is cut to the right, “left” when the steering operation is turned to the left, and “center” when neither is turned.
  • the brake pedal is depressed by the driver 110
  • “plus” is output
  • “minus” is output when the accelerator pedal is depressed
  • “0” is output when neither is depressed. It should be noted that when such an output is made from the control unit 107 side of the automobile 100, the damage reduction device 1 does not need to have the driver intention reflection unit 40.
  • the selection unit 50 selects one action corresponding to the operation data when a collision is predicted, from a plurality of action candidates calculated by the candidate calculation unit 30. For example, when the prediction unit 20 predicts a collision with the object 1000, the selection unit 50 receives an output from the driver intention reflection unit 40, and performs a corresponding avoidance action from an avoidance action classification table described later based on this output. select. The selection unit 50 sends the selected avoidance behavior data to the control unit 107 of the automobile 100 via the output unit 70. The control unit 107 executes an avoidance action sequence based on this data, and performs steering, braking, and acceleration of the automobile 100 under control without intervention of the driver.
  • the in-vehicle occupant status grasping unit 60 determines which seat in the vehicle is occupant based on the imaging data from the in-vehicle imaging camera 103 which is a camera for imaging the in-vehicle occupant status, and sends it to the candidate calculation unit 30.
  • the candidate calculation unit 30 is a route in which there is a possibility that the automobile 100 may collide with the target object 1000, and a route that makes the steering direction by the steering device 104 a right direction is a left route.
  • the evaluation value of the action candidate that becomes a route that collides with the side on which the passenger of the automobile 100 is not on board is increased based on the passenger situation in the vehicle.
  • the intention of the driver 110 is assigned to the steering operation of the driver 110, the brake depression amount, and the accelerator depression amount detected by the steering angle sensor 1041, the brake switch 1052, and the accelerator sensor 1061.
  • the driver intention reflection unit 40 initializes a steering value and a braking (acceleration) value (S501).
  • the initial value of the steering value is the center, and the initial value of the braking (acceleration) value is zero.
  • the driver intention reflection unit 40 determines whether or not the steering operation by the driver 110 is turned to the right (S502), and if it is turned to the right, sets the left and right values to “right” (S503). ).
  • the driver intention reflection unit 40 judges whether the steering operation by the driver 110 is turned to the left ( If it is cut to the left, the left and right values are set to “left” (S505). If the steering wheel operation by the driver 110 is not turned to either the left or right in step 502, the left and right values remain at the “center” of the initial value.
  • the driver intention reflection unit 40 determines whether or not the brake pedal is depressed by the driver 110 (S506), and if it is depressed, sets the value of braking (acceleration) to “plus” (S507).
  • the driver intention reflecting unit 40 determines whether the accelerator pedal is depressed by the driver 110 (S508). If so, braking (acceleration) is set to a "minus" value (S509). If neither the brake pedal nor the accelerator pedal is depressed by the driver 110, the braking (acceleration) value remains the initial value “0”. Finally, the driver intention reflection unit 40 outputs the braking (acceleration) value and the left and right values to the selection unit 50 (S510).
  • the candidate calculation unit 30 lists objects 1000 that are likely to collide from information of the distance sensor 101 and the front camera 102 (S601). For example, as illustrated in FIG. 7, when the candidate calculation unit 30 detects from the situation data from the distance sensor 101 and the front camera 102 that a pedestrian 1100 and a trash can 1200 are present as the object 1000 in the moving direction of the automobile 100. The pedestrian 1100 and the trash can 1200 are listed as the target object 1000 having the possibility of collision.
  • the candidate calculation unit 30 obtains the distance between the automobile 100 and the object 1000 and the relative speed between the automobile 100 and the object 1000 for each object 1000 listed (S602).
  • Such speed can be obtained, for example, by differentiating data from the distance sensor 101.
  • speed data from the wheel speed sensor 1051 of the automobile 100 may be used, and either value is corrected. It may be used as data.
  • the candidate calculation unit 30 obtains a plurality of candidates for avoidance actions such as braking (acceleration) and steering angle of the vehicle 100 to avoid the listed object 1000 based on the obtained distance and speed. Then, it writes together with each evaluation value in the avoidance action table described later (S603).
  • the evaluation value is a value obtained by quantifying damage predicted for each of the avoidance action candidates.
  • the candidate calculation unit 30 sorts the avoidance action table in descending order of evaluation value (S604), and leaves only the avoidance action having an evaluation value equal to or greater than a certain threshold value in the avoidance action table (S605). ). This is because an avoidance action with an evaluation value below a certain level may cause damage, so that it cannot be selected later.
  • the candidate calculation unit 30 searches for the avoidance action candidate corresponding to the same type of operation among the action table items among the items of the avoidance action classification table divided for each of the braking (acceleration) and steering operations.
  • the avoidance action candidate having the highest evaluation value is written in the avoidance action classification table described later (S606).
  • FIG. 8 is an example of an avoidance action table corresponding to the example of the emergency avoidance situation shown in FIG.
  • the avoidance action route a, b, and c are obtained as shown in FIG.
  • the avoidance action route a is an avoidance action when steering leftward by 50 ° with strong braking and an evaluation value of 0.7.
  • the avoidance action route b is an action to be avoided when the braking is weakened and the vehicle is steered in the right direction by 30 °, and the evaluation value is 0.8.
  • the avoidance action route c is an action when the vehicle is steered straight with the braking being maximized, and the evaluation value is 0.5.
  • FIG. 9 is a diagram illustrating an example of the avoidance action classification table.
  • the avoidance action classification table associates the braking (acceleration) operation and steering operation of the driver 110 with avoidance action candidates, and the action taken by the driver 110 indicates which process in the process list obtained by the candidate calculation unit 30. That is, it is a table showing which avoidance action is close to. However, if there is no corresponding process, the field is blank (process action ID field).
  • FIG. 10 is a flowchart showing the operation of the selection unit 50.
  • the selection unit 50 compares the avoidance action correspondence table with the intention of the driver 110 in S1006 and determines that there is a corresponding avoidance action candidate, the selection unit 50 selects the avoidance action (S1007) and selects the avoidance Action braking and steering operation data is sent to the control unit 107 (S1008).
  • the driver 110 unconsciously or consciously depresses the brake pedal to the left. If the steering wheel operation is performed, the data output from the driver intention reflecting unit 40 indicates that the braking (acceleration) type is “plus” and the steering type is “left”. Select the avoidance action classification ID “5” in the avoidance action classification table shown in FIG. 9, select the avoidance action route a sequence (braking force 70, left 50 °) in the avoidance action table, and output the sequence The data is sent to the control unit 107 via the unit 70.
  • the selection unit 50 returns to S1001 again to check whether or not the collision with the object 1000 is predicted by the prediction unit 20, and when the collision is predicted, the operation after S1002 is performed again.
  • the selection part 50 complete finishes a process, when a collision is not estimated.
  • the automobile 110 is equipped with the damage reduction device 1 so that the driver 110 is optimally selected from the several options of automatic braking and automatic collision avoidance obtained by the damage reduction device 1 in a situation where collision avoidance is difficult. Therefore, it is possible to reliably perform an operation to reduce the damage caused by the collision, which reflects high-order judgment criteria that are difficult to judge mechanically.
  • collision accident avoidance technology and damage mitigation technology using automatic emergency brakes and collision avoidance systems are disclosed, but they perform processing that seems to be optimal based on predetermined evaluation values.
  • a high-order human judgment is required, for example, when it is necessary to avoid a person over a thing or a judgment including the action of the person, there is a possibility that an optimal action cannot be taken.
  • this damage reduction apparatus 1 when a collision is predicted, the intention of the driver 110 who is a person is reflected from a plurality of action candidates, and finally the judgment by the damage reduction apparatus 1 side. Since the optimum action for avoiding the collision is selected, damage caused by the collision can be surely reduced. (Other)
  • This technology does not rely on the disclosed avoidance behavior algorithm, but recognizes objects that may collide and calculates avoidance behavior that does not collide with them or reduces damage. Anything is acceptable.
  • the present technology can be applied not only to automobiles but also to various mobile devices such as ships, airplanes, and robots, and also to various technical fields such as simulation devices and game machines.
  • the present technology can be configured as follows. (1) An input unit for inputting status data relating to a situation of a moving direction of the mobile device and operation data relating to a moving operation of an operator who operates the mobile device; A prediction unit that predicts a collision with the object in the moving direction based on the situation data; When the collision is predicted, a candidate calculation unit that calculates a plurality of behavior candidates for the mobile device to avoid the predicted collision based on the situation data; A damage reduction apparatus comprising: a selection unit that selects one action according to the operation data when the collision is predicted from the plurality of action candidates.
  • the said candidate calculation part calculates the evaluation value which quantified the damage estimated with respect to the said action to each of the said some action candidate The damage reduction apparatus.
  • the damage reducing device according to (2) above The damage reducing device selects one action from the plurality of action candidates based on the evaluation value when the operation data is not input when the collision is predicted.
  • the damage reducing device according to (2) above
  • the selection unit selects one action from the plurality of action candidates based on the evaluation value when the plurality of action candidates do not have an action according to the operation data when the collision is predicted.
  • Damage reduction device (5) The damage reducing device according to (2) above, The candidate calculation unit leaves a candidate for an action whose evaluation value is equal to or greater than a predetermined threshold from the plurality of calculated action candidates as a candidate to be selected by the selection unit.
  • the damage reducing device according to any one of (1) to (6), The input unit inputs braking, acceleration and steering data of the mobile device as the operation data,
  • the candidate calculation unit lists one or more objects to be avoided based on the input braking, acceleration, and steering data when a collision with the object in the moving direction is predicted, and lists For each target object, a distance between the mobile device and the target object and a relative approach speed between the mobile device and the target object are calculated, and the plurality of behaviors are calculated based on the calculated distance and speed. Calculate multiple candidates for braking, acceleration and steering of the mobile device to avoid collision with these objects as candidates,
  • the selection unit responds to braking, acceleration and steering data input from the mobile device via the input unit when the collision is predicted from the calculated braking, acceleration and steering candidates. Select braking, acceleration and steering of the mobile device to avoid collision with the object, Damage reduction equipment
  • Data on braking, acceleration and steering of the mobile device for avoiding a collision with the object selected by the selection unit is output to a control unit which controls braking, acceleration and steering of the mobile device.
  • Damage reduction device with output section (8) Enter status data regarding the status of the mobile device in the direction of movement, Predicting a collision with the object in the moving direction based on the input situation data; When the collision is predicted, the mobile device calculates a plurality of action candidates for avoiding the predicted collision based on the situation data; Input operation data relating to a moving operation of an operator who operates the mobile device when the collision is predicted, A damage reduction method for selecting one action according to the operation data from the plurality of action candidates.
  • (9) On the computer, Inputting status data relating to the status of the mobile device in the direction of movement; Predicting a collision with an object in the moving direction based on the input situation data; When the collision is predicted, the mobile device calculates a plurality of action candidates for avoiding the predicted collision based on the situation data; Inputting operation data relating to a moving operation of an operator who operates the mobile device when the collision is predicted; A step of selecting one action corresponding to the operation data from the plurality of action candidates.
  • (10) A mobile device equipped with the device according to the above (1) to (7).
  • (11) A mobile device using the method according to (8) above.
  • the mobile device according to (10) or (11) is any one of an automobile, a ship, an airplane, and a robot.

Landscapes

  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Traffic Control Systems (AREA)

Abstract

L'invention concerne un dispositif de réduction de dommages (1) qui comporte : une unité d'entrée (10) qui entre des données d'état associées à un état dans une direction de déplacement d'un dispositif d'entité mobile (100) et des données d'actionnement associées à une opération de déplacement d'un opérateur (110) actionnant le dispositif d'entité mobile (100) ; une unité de prédiction (20) qui prévoit une collision avec un objet (1000) dans la direction de déplacement sur la base des données d'état ; une unité de calcul de candidats (30) qui calcule, si une collision est prévue, sur la base des données d'état, une pluralité de candidats d'action pour le dispositif d'entité mobile (100) afin d'éviter la collision prévue ; une unité de sélection (50) qui sélectionne, parmi la pluralité de candidats d'action, une action correspondant aux données d'actionnement au moment où la collision a été prévue.
PCT/JP2016/004283 2015-09-28 2016-09-20 Dispositif de réduction de dommages, procédé de réduction de dommages et programme WO2017056454A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2015189508 2015-09-28
JP2015-189508 2015-09-28

Publications (1)

Publication Number Publication Date
WO2017056454A1 true WO2017056454A1 (fr) 2017-04-06

Family

ID=58422999

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2016/004283 WO2017056454A1 (fr) 2015-09-28 2016-09-20 Dispositif de réduction de dommages, procédé de réduction de dommages et programme

Country Status (1)

Country Link
WO (1) WO2017056454A1 (fr)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109345020A (zh) * 2018-10-02 2019-02-15 北京航空航天大学 一种完全信息下的无信号交叉口车辆驾驶行为预测模型
WO2020188717A1 (fr) * 2019-03-18 2020-09-24 三菱電機株式会社 Système de génération d'itinéraire, procédé de génération d'itinéraire et programme de génération d'itinéraire
WO2023166725A1 (fr) * 2022-03-04 2023-09-07 日本電気株式会社 Dispositif de commande de surveillance, procédé et support non transitoire lisible par ordinateur

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050090955A1 (en) * 2003-10-27 2005-04-28 Engelman Gerald H. Threat level identification and quantifying system
JP2011186878A (ja) * 2010-03-10 2011-09-22 Nissan Motor Co Ltd 移動体走行経路生成装置
JP2013218678A (ja) * 2012-03-23 2013-10-24 Institut Francais Des Sciences Et Technologies Des Transports De L'Amenagement Et Des Reseaux 路上車両の経路についての情報の決定方法
JP2015058890A (ja) * 2013-09-20 2015-03-30 株式会社デンソー 走行軌道生成装置、および走行軌道生成プログラム
JP2015072611A (ja) * 2013-10-03 2015-04-16 トヨタ自動車株式会社 運転支援装置

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050090955A1 (en) * 2003-10-27 2005-04-28 Engelman Gerald H. Threat level identification and quantifying system
JP2011186878A (ja) * 2010-03-10 2011-09-22 Nissan Motor Co Ltd 移動体走行経路生成装置
JP2013218678A (ja) * 2012-03-23 2013-10-24 Institut Francais Des Sciences Et Technologies Des Transports De L'Amenagement Et Des Reseaux 路上車両の経路についての情報の決定方法
JP2015058890A (ja) * 2013-09-20 2015-03-30 株式会社デンソー 走行軌道生成装置、および走行軌道生成プログラム
JP2015072611A (ja) * 2013-10-03 2015-04-16 トヨタ自動車株式会社 運転支援装置

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109345020A (zh) * 2018-10-02 2019-02-15 北京航空航天大学 一种完全信息下的无信号交叉口车辆驾驶行为预测模型
CN109345020B (zh) * 2018-10-02 2022-04-01 北京航空航天大学 一种完全信息下的无信号交叉口车辆驾驶行为预测方法
WO2020188717A1 (fr) * 2019-03-18 2020-09-24 三菱電機株式会社 Système de génération d'itinéraire, procédé de génération d'itinéraire et programme de génération d'itinéraire
JPWO2020188717A1 (ja) * 2019-03-18 2021-09-27 三菱電機株式会社 ルート生成システム、ルート生成方法、およびルート生成プログラム
CN113544030A (zh) * 2019-03-18 2021-10-22 三菱电机株式会社 路线生成系统、路线生成方法和路线生成程序
JP6991390B2 (ja) 2019-03-18 2022-01-12 三菱電機株式会社 ルート生成システム、ルート生成方法、およびルート生成プログラム
CN113544030B (zh) * 2019-03-18 2023-11-17 三菱电机株式会社 路线生成系统、路线生成方法和计算机能读取的存储介质
WO2023166725A1 (fr) * 2022-03-04 2023-09-07 日本電気株式会社 Dispositif de commande de surveillance, procédé et support non transitoire lisible par ordinateur

Similar Documents

Publication Publication Date Title
US11772644B2 (en) Damage reduction device, damage reduction method, and program
JP6123873B2 (ja) 車両用先進運転者支援システム及びその制御方法
US10308249B2 (en) Adaptive cruise control system and vehicle comprising an adaptive cruise control system
US8457359B2 (en) Method and assistance system for detecting objects in the surrounding area of a vehicle
EP3305620B1 (fr) Dispositif et procédé de commande de véhicule
CN109572690B (zh) 车辆控制装置
CN106184202A (zh) 一种用于车辆的自动紧急转向系统及其控制方法
JP2016051465A (ja) 衝突後操縦計画を行う方法及びシステム、並びに当該システムを備える車両
JP2012111263A (ja) 自動制御装置
US11993276B2 (en) Integrated control apparatus and method for normal and emergency control of vehicle
US11292463B2 (en) Determination of a control signal for an in-part-autonomous vehicle
EP3133454A1 (fr) Procédé et appareil pour commander un véhicule équipé avec des fonctions de commande d'entraînement automatisée
WO2017056454A1 (fr) Dispositif de réduction de dommages, procédé de réduction de dommages et programme
JP2022103827A (ja) 車両制御装置、車両制御方法、およびプログラム
CN114523928A (zh) 控制安全装置的方法和设备、车辆的安全系统和存储介质
US20200298885A1 (en) Vehicle control apparatus, vehicle control method, vehicle, and storage medium
CN114523929A (zh) 用于控制车辆安全装置的方法和设备及用于车辆的安全系统
JP2012045984A (ja) 衝突軽減装置
US20200033863A1 (en) Adaptation of an evaluable scanning area of sensors and adapted evaluation of sensor data
CN114523927A (zh) 控制安全装置的方法和设备、车辆的安全系统和存储介质
JP2016002898A (ja) 車両用衝突制御装置
KR20170068062A (ko) 차량의 후방추돌 회피장치 및 방법
CN114454839A (zh) 控制车辆安全装置的方法和设备、车辆安全系统和存储介质
WO2017056375A1 (fr) Dispositif de réduction de dommages, procédé de réduction de dommages, et programme
US20210300439A1 (en) Vehicle control device

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: 16850634

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

NENP Non-entry into the national phase

Ref country code: JP

122 Ep: pct application non-entry in european phase

Ref document number: 16850634

Country of ref document: EP

Kind code of ref document: A1