WO2021217752A1 - 车辆-行人碰撞风险域的计算方法及安全评价系统 - Google Patents

车辆-行人碰撞风险域的计算方法及安全评价系统 Download PDF

Info

Publication number
WO2021217752A1
WO2021217752A1 PCT/CN2020/091061 CN2020091061W WO2021217752A1 WO 2021217752 A1 WO2021217752 A1 WO 2021217752A1 CN 2020091061 W CN2020091061 W CN 2020091061W WO 2021217752 A1 WO2021217752 A1 WO 2021217752A1
Authority
WO
WIPO (PCT)
Prior art keywords
vehicle
pedestrian
pedestrians
collision
ttc
Prior art date
Application number
PCT/CN2020/091061
Other languages
English (en)
French (fr)
Inventor
聂冰冰
李泉
甘顺
李升波
Original Assignee
清华大学
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 清华大学 filed Critical 清华大学
Priority to US17/920,805 priority Critical patent/US20230192078A1/en
Publication of WO2021217752A1 publication Critical patent/WO2021217752A1/zh

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
    • B60W60/00—Drive control systems specially adapted for autonomous road vehicles
    • B60W60/001—Planning or execution of driving tasks
    • B60W60/0027—Planning or execution of driving tasks using trajectory prediction for other traffic participants
    • B60W60/00274—Planning or execution of driving tasks using trajectory prediction for other traffic participants considering possible movement changes
    • 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
    • B60W30/0956—Predicting travel path or likelihood of collision the prediction being responsive to traffic or environmental parameters
    • 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
    • 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
    • B60W30/0953—Predicting travel path or likelihood of collision the prediction being responsive to vehicle dynamic parameters
    • 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
    • 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
    • B60W40/02—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 ambient conditions
    • 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
    • B60W2554/00—Input parameters relating to objects
    • B60W2554/40—Dynamic objects, e.g. animals, windblown objects
    • B60W2554/402—Type
    • B60W2554/4029—Pedestrians
    • 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
    • B60W2554/00—Input parameters relating to objects
    • B60W2554/40—Dynamic objects, e.g. animals, windblown objects
    • B60W2554/404—Characteristics
    • B60W2554/4042—Longitudinal speed
    • 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
    • B60W2554/00—Input parameters relating to objects
    • B60W2554/40—Dynamic objects, e.g. animals, windblown objects
    • B60W2554/404—Characteristics
    • B60W2554/4045—Intention, e.g. lane change or imminent movement
    • 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
    • B60W2554/00—Input parameters relating to objects
    • B60W2554/80—Spatial relation or speed relative to objects
    • 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
    • B60W2554/00—Input parameters relating to objects
    • B60W2554/80—Spatial relation or speed relative to objects
    • B60W2554/802—Longitudinal distance
    • 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
    • B60W2554/00—Input parameters relating to objects
    • B60W2554/80—Spatial relation or speed relative to objects
    • B60W2554/806—Relative heading

Definitions

  • the present disclosure relates to the field of automobile safety technology, and in particular to a calculation method and a safety evaluation system for a vehicle-pedestrian collision risk domain.
  • the AEB system mainly includes three modules: control module (ECU), ranging module and braking module.
  • the core of the ranging module includes microwave radar, face recognition technology and video system, which can provide safe, accurate, real-time images and road condition information on the road ahead.
  • the AEB system can detect dangerous working conditions and automatically activate the emergency braking function.
  • the vehicle In the pedestrian detection module of the AEB system, the vehicle first recognizes the pedestrian, and then assumes that the pedestrian is stationary or moving at a constant speed to predict the pedestrian's future motion behavior. But the actual working condition is that when pedestrians find that the vehicle is in danger, the pedestrian will take the initiative to retreat to give way or accelerate forward to avoid the vehicle. Since the current AEB system does not fully consider the pedestrian's active avoidance behavior, the identification of the collision risk between the vehicle and the pedestrian is insufficient.
  • the present disclosure provides a method for determining a collision risk domain between a vehicle and a pedestrian, including:
  • the collision risk domain between the vehicle and the pedestrian is determined.
  • the present disclosure provides a method for evaluating the risk of collision between a vehicle and a pedestrian, including:
  • the collision risk between the vehicle and the pedestrian is evaluated.
  • the present disclosure provides a vehicle-pedestrian collision risk evaluation system, including:
  • Detection module used to detect vehicle information and pedestrian information
  • the analysis and judgment module connected to the detection module, is used to judge whether the pedestrian notices the vehicle. If the pedestrian notices the vehicle, the pedestrian takes an active avoidance behavior, and if the pedestrian does not notice the vehicle, the pedestrian walks normally;
  • the calculation module is connected to the analysis and judgment module, and is used to determine the collision risk domain between the vehicle and the pedestrian when the pedestrian takes the active avoidance behavior and the pedestrian does not take the active avoidance behavior;
  • the evaluation module is connected to the calculation module and is used to evaluate the collision risk between the vehicle and the pedestrian according to whether the pedestrian is within the scope of the collision risk domain.
  • the disclosure provides a calculation method and a safety evaluation system for a vehicle-pedestrian collision risk domain.
  • the method for determining the risk domain of collision between a vehicle and a pedestrian includes: detecting and outputting vehicle information and pedestrian information. Determine whether pedestrians notice the vehicle. If the pedestrian notices the vehicle, the pedestrian will take the initiative to avoid behavior, if the pedestrian does not notice the vehicle, the pedestrian walks normally. In the case where pedestrians take active avoidance behavior and pedestrians do not take active avoidance behavior, respectively, it is further assumed whether the vehicle takes immediate response actions. According to the hypothetical results of whether pedestrians take active avoidance behavior and whether the vehicle takes immediate response actions, the collision risk domain between the vehicle and the pedestrian is determined.
  • the pedestrian's active avoidance ability and the immediate response action of the vehicle are taken into consideration at the same time, and the identification of the risk of a collision between the vehicle and the pedestrian is more sufficient. Determining an effective collision risk domain in the present disclosure can effectively improve the safety of pedestrians and the comfort of vehicle driving in the process of vehicle-pedestrian interaction.
  • FIG. 1 is a schematic diagram of a certain operating state of vehicles and pedestrians in road traffic according to an embodiment of the present disclosure
  • FIG. 2 is a flowchart of the steps of a method for determining a collision risk domain between a vehicle and a pedestrian provided in an embodiment of the present disclosure
  • FIG. 3 is a schematic diagram of a generation process of a method for determining a collision risk domain between a vehicle and a pedestrian provided in an embodiment of the present disclosure
  • FIG. 4 is a kinematics characteristic diagram of a pedestrian in the process of emergency stop and back to avoid collision danger according to an embodiment of the present disclosure
  • FIG. 5 is a kinematics characteristic diagram of a pedestrian in a process of advancing acceleration and avoiding collision danger according to an embodiment of the present disclosure
  • FIG. 6 is a schematic diagram of the TTC boundary for pedestrians to avoid backwards and the TTC boundary for pedestrians to avoid forward and the unavoidable area of collision between vehicles and pedestrians provided in an embodiment of the present disclosure
  • Fig. 7 is a schematic diagram of a collision risk domain between a vehicle and a pedestrian under a given working condition provided in an embodiment of the present disclosure
  • FIG. 8 is a schematic diagram of a method for evaluating the risk of collision between a vehicle and a pedestrian provided in an embodiment of the present disclosure
  • Fig. 9 is a schematic diagram of a system for evaluating the risk of collision between a vehicle and a pedestrian provided in an embodiment of the present disclosure.
  • Vehicle and pedestrian collision risk assessment system 100 Vehicle and pedestrian collision risk assessment system 100
  • the present disclosure relates to the field of automobile safety, and proposes a method for identifying the collision risk domain in the process of vehicle-pedestrian interaction based on the active avoidance ability of vehicles and pedestrians.
  • the method calculates and judges in real time according to the driving state of the vehicle and the motion state of the pedestrian Whether the pedestrian is in the dangerous area where the vehicle is driving, predict the risk of vehicle-pedestrian collision.
  • it can obtain a more accurate calculation of the collision risk domain of the vehicle-pedestrian interaction, and provide a basis for the risk determination and decision-making behavior of autonomous vehicles.
  • TTC time to collision
  • the vehicle-mounted perception system includes a variety of sensors, such as visual perception modules, millimeter wave radars, ultrasonic radars, and 360° surround view systems.
  • the synergy of multi-source sensors recognizes obstacles such as road lane lines, pedestrians and vehicles, and escorts safe driving.
  • the advanced vehicle-mounted perception system can identify pedestrian information within a certain range, including pedestrian speed, location, direction of view, etc., and provide information support for vehicle collision risk judgment and control decision-making.
  • Pedestrian avoidance ability refers to the active avoidance ability of pedestrians when they find danger. According to traffic accident investigation and test results, pedestrians' avoidance ability when encountering danger can reduce the risk of dangerous accidents. Therefore, in the following disclosure, based on the test results conducted by the inventor, the speed of pedestrians in the avoidance process is quantified and defined as the pedestrian avoidance ability.
  • FIG. 2 provides a method for determining a collision risk domain between a vehicle and a pedestrian for the present disclosure.
  • the method for determining the risk domain of collision between a vehicle and a pedestrian includes:
  • the vehicle information includes: vehicle position, vehicle speed, vehicle direction, vehicle maximum braking deceleration, and vehicle maximum lateral acceleration.
  • the pedestrian information includes: pedestrian position, pedestrian speed, pedestrian direction, and pedestrian sight direction. Both the vehicle information and the pedestrian information can be obtained by a vehicle-mounted sensing system.
  • S200 Determine whether the pedestrian notices the vehicle.
  • the activation of the pedestrian avoidance ability depends on the direction of the pedestrian's vision. If the direction of the pedestrian's vision is to focus on the approaching vehicle, it is confirmed that the pedestrian avoidance ability is activated. If the pedestrian's view direction does not pay attention to the oncoming vehicle, it is considered that the pedestrian's avoidance ability has not been activated, and the pedestrian continues to maintain normal movement behavior. If the pedestrian notices the vehicle, the pedestrian will take the initiative to avoid behavior. If the pedestrian does not notice the vehicle, the pedestrian walks normally.
  • the active avoidance behaviors taken by pedestrians include: emergency stop and back to avoid or accelerate forward to avoid. Of course, based on the core design ideas of the present disclosure, more pedestrian avoidance behaviors can be designed into this solution to determine more accurate collision risk domains.
  • the immediate response actions taken by the vehicle include but are not limited to normal driving, emergency braking, and emergency steering.
  • the immediate reaction actions taken by the vehicle may also include decelerating and braking with the help of other facilities on the road.
  • S400 Determine the collision risk domain between the vehicle and the pedestrian based on the hypothetical result of whether the pedestrian takes an active avoidance behavior and whether the vehicle takes an immediate response action.
  • the collision risk domain between the vehicle and the pedestrian is different in the case of whether the pedestrian adopts an active avoidance behavior and whether the vehicle takes an immediate response action. It can be understood that when pedestrians adopt active avoidance behavior while vehicles take immediate response actions, the collision risk domain between the vehicle and pedestrians will be smaller.
  • the vehicle-mounted detection system is first used to detect vehicle information and pedestrian information. It is further judged whether pedestrians take active avoidance behavior and whether vehicles take immediate response actions. Calculate the collision risk domains of vehicles and pedestrians in different situations.
  • the method for determining the collision risk domain between a vehicle and a pedestrian takes into account the pedestrian's active avoidance ability and the immediate response action of the vehicle, and the identification of the risk of a collision between the vehicle and the pedestrian is more adequate.
  • determining effective collision risk domains under different conditions can effectively improve the safety of pedestrians and the comfort of vehicle driving in the process of vehicle-pedestrian interaction. By judging whether the pedestrian has noticed the vehicle, the effective avoidance behavior of the pedestrian in the face of dangerous conditions can be classified and quantified.
  • the identification of the collision risk domain in the vehicle-pedestrian interaction process based on the pedestrian's active avoidance behavior in this embodiment is of great significance to the improvement of the safety of the autonomous vehicle.
  • the pedestrian taking an active avoidance behavior includes: emergency stop and back to avoid or accelerate forward to avoid.
  • the steps of determining the collision risk domain between the vehicle and the pedestrian based on the hypothetical result of whether the pedestrian takes active avoidance behavior and whether the vehicle takes immediate response action include:
  • S410 Calculate the time safety boundary at which the vehicle and the pedestrian respectively reach the potential collision point when the pedestrian adopts an active avoidance behavior, and the time safety boundary is the TTC safety envelope.
  • S420 on the basis of the TTC safety envelope, further determine the collision risk domain between the vehicle and the pedestrian.
  • the avoidance capability of the vehicle it is determined that the avoidance capability of the vehicle is not considered, and the TTC safety envelope of the collision between the vehicle and the pedestrian is obtained in the case that the pedestrian actively avoids.
  • the avoidance capability of the vehicle it can be determined that the avoidance capability of the vehicle is considered, and the TTC safety envelope of the collision between the vehicle and the pedestrian can be obtained when the pedestrian actively avoids the situation.
  • the step of determining the TTC safety envelope of the risk of collision between a vehicle and a pedestrian includes:
  • S411 Determine the detection distance range of the vehicle.
  • the vehicle detection distance range can be detected and acquired by the vehicle-mounted detection system.
  • S412 Calculate the shortest distance required for the pedestrian to safely evade the vehicle based on the pedestrian's emergency stop and retreat and avoid speed, the pedestrian's acceleration and forward avoidance speed, and the vehicle width. In this step, the calculation is divided into two situations. The shortest distance required for pedestrians to safely evade the vehicle is calculated according to the pedestrian's emergency stop and retreat speed and vehicle width, which is the first distance. The shortest distance required for pedestrians to safely avoid vehicles is calculated according to pedestrians' acceleration and avoidance speed and vehicle width, which is the second distance.
  • S413 Calculate the safety boundary of the pedestrian collision time according to the pedestrian's initial speed and the shortest distance required for the pedestrian to safely avoid the vehicle. According to the pedestrian's initial speed, the first distance and the second distance calculated in S412 are combined to calculate the first safety margin and the second safety margin of the pedestrian collision time respectively.
  • the first safety boundary and the second safety boundary are safety boundaries at the time of collision.
  • S414 Calculate the TTC safety envelope of the collision risk between the vehicle and the pedestrian according to the safety boundary of the vehicle collision time and the pedestrian collision time.
  • the TTC safety envelope of the vehicle-pedestrian collision risk is determined according to the first safety boundary, the second safety boundary, and the vehicle-pedestrian collision time range.
  • FIG 3 provides the generation process of the vehicle-pedestrian collision risk domain.
  • the collision risk domain derived in Figure 3 is affected by the reaction behavior of both vehicles and pedestrians.
  • the overall method for determining the collision risk domain is shown in Figure 3.
  • the vehicle's own perception detection system detects vehicle information and pedestrian information. Determine whether the pedestrian avoidance ability is activated or not by judging whether the pedestrian notices the vehicle. The activation of the pedestrian avoidance ability may depend on the direction of the pedestrian's view. If the pedestrian's view direction is to focus on the oncoming vehicle, it is confirmed that the pedestrian avoidance ability is activated. If the pedestrian's view direction does not pay attention to the oncoming vehicle, it is considered that the pedestrian's avoidance ability has not been activated, and the pedestrian continues to maintain normal movement behavior.
  • the following formula (1) to formula (7) may be used to calculate the TTC safety envelope of the risk of collision between a vehicle and a pedestrian:
  • TTC pb-ne (TTC vd ) D pb-ne (TTC vd )/v pw (5)
  • TTC pf-fe (TTC vd ) D pf-fe (TTC vd )/v pw (6)
  • Ne in the above formula is the abbreviation of near-end, which means that the side of the vehicle close to the pedestrian is called near end.
  • the fe in the above formula is the abbreviation of far-end, which means the side of the vehicle away from the pedestrian, called the far end.
  • D vr represents the farthest distance detected by the vehicle; v v represents the driving speed of the vehicle; TTC vr represents the time for the vehicle to reach the farthest detection point at the current speed; TTC vd represents the time for the vehicle to reach a potential collision, and TTC vd is a variable; v pb represents the avoidance speed of pedestrians when they stop and retreat; v pf represents the avoidance speed of pedestrians moving forward and accelerating; D pb-ne represents the shortest distance that pedestrians need to reach the near end of the vehicle to safely avoid the vehicle when pedestrians stop and retreat; TTC pf-fe represents the shortest distance from the far end of the vehicle when pedestrians are accelerating and avoiding vehicles; L vw represents the width of the vehicle; v pw represents the initial speed before the pedestrian starts to avoid; TTC pb-ne represents the pedestrian The shortest time required for pedestrians to safely evade vehicles during emergency stop and backing; TTC pf-ne represents the
  • the pedestrian detected by the vehicle is in the right front of the vehicle
  • the direction of movement of the vehicle and the pedestrian is perpendicular
  • the width of the vehicle L vw 2m.
  • the above formula (2) can be expressed as: ⁇ TTC vd :0 ⁇ TTC vd ⁇ 6s ⁇ .
  • the TTC safety envelope of the collision risk between the vehicle and the pedestrian is calculated with reference to the above formula (1) to formula (7), as shown in FIG. 6.
  • the TTC boundary for avoiding pedestrian backward and the TTC boundary for avoiding pedestrian forward show the dangerous unavoidable area between vehicles and pedestrians (shown in the shaded part of FIG. 6 ).
  • the step of determining the collision risk domain between the vehicle and the pedestrian according to the hypothetical result of whether the pedestrian takes the active avoidance behavior and whether the vehicle takes the immediate reaction action includes:
  • the first collision risk domain is determined according to the following formulas (11) to (13): [D v-1 ,D p -1-fe ] ⁇ first collision risk domain ⁇ [D v-1 ,D p-1-ne ];
  • v v represents the current driving speed of the vehicle
  • TTC vd represents the time when the vehicle reaches the potential collision point
  • v pd represents the current speed of the pedestrian detected by the vehicle in the actual motion scene
  • D v-1 represents within the range of TTC vd, The first driving range of the vehicle in the direction of travel
  • L vw represents the width of the vehicle
  • D p-1-ne represents the first shortest distance from the near end of the vehicle when pedestrians are moving at a speed of v pd when pedestrians safely avoid the vehicle
  • D p-1-fe represents the second shortest distance from the far end of the vehicle when the pedestrian is moving at the speed of v pd.
  • the second risk area is determined.
  • the second collision risk domain is determined according to the following formula (14) to formula (16): [D v-2 ,D p- 2-fe ] ⁇ second collision risk domain ⁇ [D v-2 ,D p-2-ne ];
  • v v represents the current speed of the vehicle
  • TTC vd represents the time when the vehicle reaches the potential collision point
  • v pd represents the current speed of the pedestrian detected by the vehicle in the actual motion scene
  • a vx-max represents the maximum braking deceleration of the vehicle
  • D v-2 represents the second driving range of the vehicle in the direction of travel within the range of TTC vd
  • L vw represents the width of the vehicle
  • D p-2-ne represents when pedestrians are moving at a speed of v pd , when pedestrians avoid the vehicle safely Need to be the third shortest distance from the near end of the vehicle
  • D p-2-fe represents the fourth shortest distance from the far end of the vehicle when pedestrians are moving at a speed of v pd.
  • the third risk domain is determined.
  • the third collision risk domain is determined according to the following formula (17) to formula (19): [D v-3 ,D p- 3-BA-ne ] ⁇ third collision risk domain ⁇ [D v-3 ,D p-3-FA-fe ];
  • v v represents the current driving speed of the vehicle
  • TTC vd represents the time when the vehicle arrives at the potential collision point
  • v pd represents the current speed of the pedestrian detected by the vehicle in the actual motion scene
  • TTC pb-ne represents the back and forth of the pedestrian after an emergency stop
  • TTC pf-ne represents the shortest time required for pedestrians to avoid vehicles safely when pedestrians are advancing to accelerate and avoid
  • L vw represents the width of the vehicle
  • D v-3 represents within the range of TTC vd , The third driving range of the vehicle in the direction of travel
  • D p-3-ne represents the fifth shortest distance from the near end of the vehicle when pedestrians are moving at a speed of v pd when pedestrians safely avoid the vehicle
  • D p-3-fe It represents the sixth shortest distance from the far end of the vehicle when the pedestrian is moving at the speed of v pd.
  • the fourth collision risk domain is determined according to the following formula (20) to formula (22): [D v-4 ,D p-4 -ne ] ⁇ fourth collision risk domain ⁇ [D v-4 ,D p-4-fe ];
  • v v represents the current speed of the vehicle
  • TTC vd represents the time when the vehicle reaches the potential collision point
  • v pd represents the current speed of the pedestrian detected by the vehicle in the actual motion scene
  • a vx-max represents the maximum braking deceleration of the vehicle
  • TTC pb-ne represents the shortest time required for pedestrians to safely avoid vehicles when pedestrians stop and back to avoid emergency
  • TTC pf-ne represents the shortest time required for pedestrians to safely avoid vehicles when pedestrians are advancing and accelerate to avoid
  • L vw represents vehicles Width
  • D v-4 represents the fourth driving range of the vehicle in the direction of travel within the range of TTC vd
  • D p-4-ne represents the distance required by the pedestrian to safely avoid the vehicle when the pedestrian is moving at the speed of v pd
  • D p-4-fe represents the eighth shortest distance from the far end of the vehicle when pedestrians are moving at a speed of v
  • the method further includes: determining a fifth risk domain in the case of emergency braking of the vehicle and emergency steering of the vehicle.
  • the immediate response action of the vehicle includes normal running of the vehicle and emergency braking of the vehicle (emergency braking of the vehicle includes straight forward emergency braking and emergency turning).
  • the braking distance of the vehicle can be determined according to the current driving speed of the vehicle and the maximum braking deceleration of the vehicle.
  • the minimum turning radius of the vehicle may be determined according to the current driving speed of the vehicle and the maximum lateral acceleration of the vehicle.
  • the braking deceleration of a vehicle refers to the ability of a vehicle to quickly reduce its travel speed until it stops.
  • the maximum braking deceleration of a vehicle depends on the coefficient of friction between the tires of the vehicle and the ground. In actual working conditions, the coefficient of friction between the ground and tires is usually 0.6 to 0.8, that is, the braking deceleration of the vehicle is usually 6m/s 2 to 8m/s 2 .
  • the maximum lateral acceleration of the vehicle refers to the acceleration in the direction perpendicular to the direction of the vehicle, and the acceleration caused by the centrifugal force generated when the vehicle is turning. This is the tendency for cars to be "flipped off". The greater the acceleration, the theoretically easier the car will be “thrown away” from the driving path. Therefore, the ultimate steering performance of the vehicle during driving depends on the maximum lateral acceleration of the vehicle.
  • the avoidance capability of the vehicle specifically refers to the braking ability and steering ability of the vehicle in the driving state, and the braking ability of the vehicle is the braking distance.
  • D vb depends on the current speed of the vehicle (v v ) and the maximum braking deceleration (a vx-max ).
  • v v represents the current driving speed of the vehicle; a vx-max represents the maximum braking deceleration of the vehicle.
  • the steering ability of a vehicle refers to the minimum turning radius (R vd-min ) that maintains the stability of the vehicle at the current driving speed, which depends on the current driving speed (v v ) and the maximum lateral acceleration (a vy-max ) of the vehicle:
  • R v-min represents the minimum turning radius of the vehicle (which is a vehicle parameter and is less than or equal to the minimum turning radius R vd-min that maintains the stability of the vehicle at the current driving speed).
  • the avoidance capability of the vehicle is taken into consideration, and the immediate response actions of the vehicle include normal running of the vehicle and emergency braking of the vehicle.
  • the emergency braking of a vehicle can include emergency braking in a straight-going situation and emergency braking in a turn.
  • the avoidance ability of the vehicle is taken into consideration, so that the determination of the collision risk domain between the vehicle and the pedestrian can be more accurate.
  • the fifth collision risk domain between the vehicle and the pedestrian is determined according to the following formula (23) to formula (28); [D v-5 ,D vl- 5-fe ] ⁇ fifth collision risk domain ⁇ [D v-5 ,D vl-5-ne ];
  • v v represents the driving speed of the vehicle
  • v vb represents the speed of the vehicle during the braking process
  • a vx-max represents the maximum braking deceleration of the vehicle
  • a vy-max represents the maximum lateral acceleration of the vehicle
  • TTC vd represents the vehicle reaching the potential The time of the collision point
  • a vs represents the accumulated steering angle during the turning process of the vehicle
  • D v-5 represents the fifth driving range of the vehicle in the direction of travel within the range of TTC vd
  • D p-5-ne represents the distance between pedestrians and v v
  • D p-5-fe represents the first distance from the far end of the vehicle required for pedestrians to safely avoid the vehicle when the pedestrian is moving at a speed of v v Ten shortest distance.
  • the present disclosure also clarifies the effectiveness of the shift in risk reduction.
  • the vehicle usually adopts braking behavior for dangerous working conditions, and the risk of braking and steering collision cannot be judged.
  • the immediate response actions of the vehicle (emergency braking in a straight-going situation and emergency braking in a turn) are considered, and the avoidance capability of the vehicle is taken into consideration, so that the collision risk domain between the vehicle and the pedestrian can be determined more accurately.
  • FIG. 7 shows a schematic diagram of a collision risk domain between a vehicle and a pedestrian.
  • the width of the vehicle L vw 2m.
  • the collision risk domain 1 in FIG. 7 is the first collision risk domain calculated in the foregoing embodiment.
  • the collision risk domain 2 in FIG. 7 is the second collision risk domain calculated in the foregoing embodiment.
  • the collision risk domain 3 in FIG. 7 is the third collision risk domain calculated in the foregoing embodiment.
  • the collision risk domain 4 in FIG. 7 is the fourth collision risk domain calculated in the foregoing embodiment.
  • the collision risk domain 5 in FIG. 7 is the fifth collision risk domain calculated in the foregoing embodiment.
  • the following execution strategy can be given to the vehicle. 1.
  • the vehicle can avoid collision by actively avoiding (braking or steering) or alerting pedestrians to pay attention to the vehicle and actively avoiding happen. 2.
  • the vehicle can avoid collision by actively avoiding (turning) or alerting pedestrians to pay attention to the vehicle and actively avoiding. 3.
  • the pedestrian's active avoidance behavior can no longer effectively avoid collisions, and the vehicle can only avoid collisions through active avoidance (braking or steering). 4.
  • the pedestrian's active avoidance behavior can no longer effectively avoid the collision, and if the vehicle only takes braking behavior, the collision cannot be avoided, and the vehicle can only pass through Active emergency steering to avoid collisions. 5.
  • the vehicle cannot avoid collision by taking any measures.
  • the effective avoidance behavior of pedestrians when facing dangerous conditions is classified and quantified.
  • the present disclosure comprehensively considers the avoidance capabilities of people and vehicles to classify risk areas.
  • the collision risk domain determined in the process of identifying the vehicle-pedestrian interaction based on the pedestrian's active avoidance behavior of the present disclosure is of great significance to the safety improvement of the autonomous vehicle.
  • the application scenario of the present disclosure is on a vehicle with active detection capability.
  • the vehicle can detect its own vehicle information including: vehicle speed, braking deceleration, lateral acceleration and other information.
  • the vehicle can identify pedestrians within the detection range, and detect the position, speed, direction of movement and direction of view of the pedestrian.
  • the above-mentioned information obtained by the vehicle is used as input, and the avoidance ability of the vehicle and pedestrian under dangerous working conditions is used as the calculation parameter, and the collision risk existing during the interaction between the vehicle and the pedestrian can be calculated in real time during the driving of the vehicle.
  • the present disclosure also provides a system for determining the risk domain of a collision between a vehicle and a pedestrian.
  • the system for determining the risk domain of collision between a vehicle and a pedestrian includes: a detection module, a first analysis and judgment module, a second analysis and judgment module, and an arithmetic module.
  • the detection module is used to detect and output vehicle information and pedestrian information.
  • the first analysis and judgment module is used to judge whether the pedestrian has noticed the vehicle. If the pedestrian notices the vehicle, the pedestrian will take the initiative to avoid behavior, if the pedestrian does not notice the vehicle, the pedestrian walks normally.
  • the second analysis and judgment module is used to further assume whether the vehicle takes an immediate response action when the pedestrian takes the active avoidance behavior and the pedestrian does not take the active avoidance behavior.
  • the calculation module is used to determine the collision risk domain between the vehicle and the pedestrian based on the hypothetical result of whether the pedestrian takes the initiative to avoid behavior and whether the vehicle takes the immediate reaction action.
  • the system for determining the risk domain of a collision between a vehicle and a pedestrian takes into account the pedestrian's active avoidance ability and the immediate response action of the vehicle, and the identification of the risk of a collision between the vehicle and the pedestrian is more adequate.
  • determining effective collision risk domains under different conditions can effectively improve the safety of pedestrians and the comfort of vehicle driving in the process of vehicle-pedestrian interaction. By judging whether the pedestrian has noticed the vehicle, the effective avoidance behavior of the pedestrian in the face of dangerous conditions can be classified and quantified.
  • the identification of the collision risk domain in the vehicle-pedestrian interaction process based on the pedestrian's active avoidance behavior in this embodiment is of great significance to the improvement of the safety of the autonomous vehicle.
  • the present disclosure provides a method for evaluating the risk of collision between a vehicle and a pedestrian, including:
  • the vehicle information includes: vehicle location, vehicle speed, vehicle direction, vehicle maximum braking deceleration, and vehicle maximum lateral acceleration
  • the pedestrian information includes: pedestrian location, pedestrian speed, pedestrian direction, and pedestrian sight direction.
  • S20 Determine whether the pedestrian notices the vehicle. If the pedestrian notices the vehicle, the pedestrian takes an active avoidance behavior. If the pedestrian does not notice the vehicle, the pedestrian walks normally.
  • the active avoidance behaviors taken by pedestrians include: emergency stop and back to avoid or accelerate forward to avoid.
  • more pedestrian avoidance behaviors can be designed into this solution to determine more accurate collision risk domains.
  • S40 Determine whether the pedestrian is within the scope of the collision risk domain. The judgment is made here based on the current detected pedestrian location, the vehicle location, and the determined collision risk domain range.
  • S50 Evaluate the collision risk between the vehicle and the pedestrian based on the judgment result of whether the pedestrian is within the scope of the collision risk domain. If the pedestrian is not within the scope of the collision risk domain, the risk of collision between the vehicle and the pedestrian is low. If the pedestrian is within the scope of the collision risk domain, the collision risk between the vehicle and the pedestrian is higher.
  • the probability of collision risk can be determined in combination with the five different collision risk domains obtained in the foregoing embodiment. For example, the probability of collision risk in the fifth risk domain is the greatest, because when a pedestrian is in the fifth collision risk domain, no matter what measures the vehicle takes, the collision cannot be avoided.
  • the step of separately determining the collision risk domain between the vehicle and the pedestrian further includes :
  • the vehicle's immediate reaction action includes normal running of the vehicle and emergency braking of the vehicle.
  • the vehicle emergency braking includes: straight forward emergency braking and emergency turning.
  • the specific method for determining the collision risk domain can be determined with reference to the steps in the method for determining the collision risk domain between vehicles and pedestrians, and will not be repeated here.
  • the step of evaluating the risk of a collision between a vehicle and a pedestrian according to a judgment result of whether the pedestrian is within the range of the collision risk domain includes:
  • the collision risk between the vehicle and the pedestrian is a first-level risk.
  • the vehicle can actively avoid (braking or steering) or issue an alarm to remind pedestrians to pay attention to the vehicle and actively avoid the collision to avoid the collision.
  • the collision risk between the vehicle and the pedestrian is a secondary risk.
  • the vehicle can actively avoid (turn) or issue an alarm to remind pedestrians to pay attention to the vehicle and actively avoid the collision to avoid the collision.
  • the collision risk between the vehicle and the pedestrian is a third level risk.
  • the collision risk between a vehicle and a pedestrian is a third-level risk, the pedestrian's active avoidance behavior can no longer effectively avoid the collision, and the vehicle can only avoid the collision through active avoidance (braking or steering).
  • the collision risk between the vehicle and the pedestrian is a fourth level risk.
  • the collision risk between a vehicle and a pedestrian is a level four risk, the pedestrian's active avoidance behavior can no longer effectively avoid the collision, and the vehicle cannot avoid the collision if the vehicle only takes the braking behavior, and the vehicle can only avoid the collision through active emergency steering.
  • the collision risk between the vehicle and the pedestrian is a level five risk, wherein the risk level of the first level risk is the lowest, and the risk level of the fifth level risk is the lowest. The highest level of risk.
  • the risk of collision between a vehicle and a pedestrian is a level five risk, the vehicle cannot avoid the collision by taking any measures.
  • the vehicle-pedestrian collision risk evaluation method provided in the present disclosure clarifies the pedestrian's active avoidance ability under dangerous conditions, and the generated vehicle-pedestrian collision risk domain considers the coupling of factors such as pedestrian position, speed, active avoidance ability, etc. Impact, the identification of the risk of vehicle-pedestrian collisions is more adequate.
  • the vehicle-pedestrian collision risk evaluation method provided in the present disclosure comprehensively considers the braking and steering capabilities of the vehicle and the pedestrian avoidance ability, and proposes a vehicle-pedestrian collision risk domain under multiple working conditions based on the pedestrian's active avoidance ability Generation method. It is of great significance to improve the identification of pedestrian risks by smart vehicles, and can effectively improve the safety of pedestrians and the comfort of vehicle driving in the process of vehicle-pedestrian interaction.
  • the present disclosure also provides a vehicle-pedestrian collision risk assessment system 100, including a detection module 10, an analysis and judgment module 20, a calculation module 30, and an evaluation module 40.
  • the detection module 10 is used to detect vehicle information and pedestrian information.
  • the analysis and judgment module 20 is connected to the detection module 10.
  • the analysis and judgment module 20 is used to judge whether the pedestrian has noticed the vehicle. If the pedestrian notices the vehicle, the pedestrian will take the initiative to avoid behavior, if the pedestrian does not notice the vehicle, the pedestrian walks normally.
  • the calculation module 30 is connected to the analysis and judgment module 20.
  • the calculation module 30 is used to determine the collision risk domains of the vehicle and the pedestrian respectively when the pedestrian adopts an active avoidance behavior and the pedestrian does not adopt an active avoidance behavior.
  • the evaluation module 40 is connected to the calculation module 30.
  • the evaluation module 40 is used to evaluate the collision risk between the vehicle and the pedestrian according to whether the pedestrian is within the scope of the collision risk domain.
  • the above-mentioned modules can be realized by relying on a computer program, and the specific hardware structure of the module is not specifically limited, as long as the above-mentioned functions can be realized.
  • the vehicle-pedestrian collision risk evaluation system 100 provided in this embodiment can execute all steps in the vehicle-pedestrian collision risk evaluation method.
  • the vehicle-pedestrian collision risk assessment system 100 also comprehensively considers the braking and steering capabilities of the vehicle and the pedestrian avoidance ability, and proposes a method for generating a vehicle-pedestrian collision risk domain under multiple working conditions based on the pedestrian's active avoidance ability. It is of great significance to improve the identification of pedestrian risks by smart vehicles, and can effectively improve the safety of pedestrians and the comfort of vehicle driving in the process of vehicle-pedestrian interaction.

Landscapes

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

Abstract

本公开涉及一种车辆-行人碰撞风险域的计算方法及安全评价系统。所述车辆与行人碰撞风险域的确定方法，包括：探测并输出车辆信息和行人信息。判断行人是否注意到车辆。在行人采取主动避让行为和行人不采取主动避让行为的情况下，分别进一步假设车辆是否采取即时反应动作。根据行人是否采取主动避让行为和车辆是否采取即时反应动作的假设结果，确定车辆与行人的碰撞风险域。本公开中，车辆与行人碰撞风险域的确定方法中，同时考虑到了行人的主动避让能力和车辆的即时反应动作，对于车辆与行人碰撞风险的识别更加充分。本公开中确定有效的碰撞风险域，可以有效提高车辆与行人交互过程中行人的安全性及车辆行驶的舒适性。

Description

车辆-行人碰撞风险域的计算方法及安全评价系统
相关申请
本公开要求2020年04月27日申请的，申请号为202010344141.0，名称为“车辆-行人碰撞风险域的计算方法及安全评价系统”的中国专利申请的优先权，在此将其全文引入作为参考。
技术领域
本公开涉及汽车安全技术领域，特别是涉及一种车辆-行人碰撞风险域的计算方法及安全评价系统。
背景技术
随着汽车自动驾驶技术的发展，针对车辆安全的技术越来越成熟，比如目前比较成熟的汽车主动安全技术。目前，存在一种基于汽车主动安全技术构建的自动紧急刹车系统(英文全称Autonomous Emergency Braking，简称AEB系统)。AEB系统主要包括三大模块：控制模块(ECU)，测距模块和制动模块。其中测距模块的核心包括微波雷达、人脸识别技术和视频系统，它可以提供前方道路安全、准确、实时的图像和路况信息。AEB系统可对危险工况进行探测并自动启动紧急制动功能。在AEB系统的行人探测模块中，车辆首先识别到行人，然后默认该行人处于静止或匀速运动状态对行人未来的运动行为进行预测。但实际的工况是，当行人发现车辆并存在危险时，行人会采取主动的后退避让或加速前进的方式躲避车辆。由于目前的AEB系统中没有充分考虑行人的主动避让行为，因此对于车辆与行人的碰撞风险的识别不充分。
发明内容
基于此，有必要针对传统的车辆与行人参与交通运行过程中对于车辆与行人的碰撞风险的识别不充分的问题，提供一种车辆-行人碰撞风险域的计算方法及安全评价系统。
本公开提供一种车辆与行人碰撞风险域的确定方法，包括：
探测并输出车辆信息和行人信息；
判断行人是否注意到车辆，若行人注意到车辆，则行人采取主动避让行为，若行人没有注意到车辆，则行人正常行走；
在行人采取主动避让行为和行人不采取主动避让行为的情况下，分别进一步假设车辆 是否采取即时反应动作；
根据行人是否采取主动避让行为的判断结果和车辆是否采取即时反应动作的假设结果，确定车辆与行人的碰撞风险域。
本公开提供一种车辆与行人碰撞风险的评价方法，包括：
探测并输出车辆信息和行人信息；
判断行人是否注意到车辆，若行人注意到车辆，则行人采取主动避让行为，若行人没有注意到车辆，则行人正常行走；
在行人采取主动避让行为和行人不采取主动避让行为的情况下，分别确定车辆与行人的碰撞风险域；
判断行人是否在所述碰撞风险域的范围内；
根据行人是否在所述碰撞风险域的范围内的判断结果，评价车辆与行人的碰撞风险。
本公开提供一种车辆与行人碰撞风险的评价系统，包括：
探测模块，用于探测车辆信息和行人信息；
分析判断模块，与所述探测模块连接，用于判断行人是否注意到车辆，若行人注意到车辆，则行人采取主动避让行为，若行人没有注意到车辆，则行人正常行走；
运算模块，与所述分析判断模块连接，用于在行人采取主动避让行为和行人不采取主动避让行为的情况下，分别确定车辆与行人的碰撞风险域；以及
评价模块，与所述运算模块连接，用于根据行人是否在所述碰撞风险域的范围内，来评价车辆与行人的碰撞风险。
本公开中提供一种车辆-行人碰撞风险域的计算方法及安全评价系统。所述车辆与行人碰撞风险域的确定方法，包括：探测并输出车辆信息和行人信息。判断行人是否注意到车辆。若行人注意到车辆，则行人采取主动避让行为，若行人没有注意到车辆，则行人正常行走。在行人采取主动避让行为和行人不采取主动避让行为的情况下，分别进一步假设车辆是否采取即时反应动作。根据行人是否采取主动避让行为和车辆是否采取即时反应动作的假设结果，确定车辆与行人的碰撞风险域。本公开中，车辆与行人碰撞风险域的确定方法中，同时考虑到了行人的主动避让能力和车辆的即时反应动作，对于车辆与行人碰撞风险的识别更加充分。本公开中确定有效的碰撞风险域，可以有效提高车辆与行人交互过程中行人的安全性及车辆行驶的舒适性。
附图说明
为了更清楚地说明本公开实施例或传统技术中的技术方案，下面将对实施例或传统技 术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本公开的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为本公开一个实施例中提供的车辆与行人在道路交通中某一运行状态的示意图；
图2为本公开一个实施例中提供的车辆与行人碰撞风险域的确定方法步骤流程图；
图3为本公开一个实施例中提供的车辆与行人碰撞风险域的确定方法的生成流程示意图；
图4为本公开一个实施例中提供的行人急停后退避让碰撞危险过程中的运动学特征图；
图5为本公开一个实施例中提供的行人前进加速避让碰撞危险过程中的运动学特征图；
图6为本公开一个实施例中提供的行人后退避让的TTC边界和行人前进避让的TTC边界以及车辆与行人的碰撞不可避免区域的示意图；
图7为本公开一个实施例中提供的给定工况下，车辆与行人碰撞风险域示意图；
图8为本公开一个实施例中提供的车辆与行人碰撞风险的评价方法示意图；
图9为本公开一个实施例中提供的车辆与行人碰撞风险的评价系统示意图。
附图标号：
车辆与行人碰撞风险的评价系统100
探测模块10
分析判断模块20
运算模块30
评价模块40
具体实施方式
为了便于理解本公开，下面将参照相关附图对本公开进行更全面的描述。附图中给出了本公开的较佳实施方式。但是，本公开可以以许多不同的形式来实现，并不限于本文所描述的实施方式。相反地，提供这些实施方式的目的是使对本公开的公开内容理解的更加透彻全面。
需要说明的是，当元件被称为“固定于”另一个元件，它可以直接在另一个元件上或者也可以存在居中的元件。当一个元件被认为是“连接”另一个元件，它可以是直接连接到另一个元件或者可能同时存在居中元件。本文所使用的术语“垂直的”、“水平的”、“左”、“右”以及类似的表述只是为了说明的目的，并不表示是唯一的实施方式。
除非另有定义，本文所使用的所有的技术和科学术语与属于本公开的技术领域的技术 人员通常理解的含义相同。本文中在本公开的说明书中所使用的术语只是为了描述具体的实施方式的目的，不是旨在于限制本公开。
本公开涉及汽车安全领域，提出了一种基于车辆和行人的主动避让能力，识别车辆-行人交互过程中碰撞风险域的确定方法，该方法根据车辆行驶状态和行人的运动状态，实时计算并判断行人是否处于车辆行驶的危险区域，预测车辆-行人的碰撞风险。相对于传统的碰撞估计/评价方法，可以获取更精确的计算车辆-行人交互的碰撞风险域，为自动驾驶车辆的风险判定、决策行为提供依据。
车辆和行人作为道路交通参与者，在道路使用过程中两者会发生交互和冲突，受车辆和行人运动过程中惯性影响，在危险工况出现的时候，其运动状态不能在瞬间被控制以避免碰撞。因此，在一定的人车交互的时间和空间范围内，车辆与行人直接的碰撞是无法避免的，存在碰撞风险的车辆-行人相对位置的区域，定义为车辆-行人碰撞风险域。如图1所示，图1中的潜在碰撞区域可以被定义为车辆-行人碰撞风险域。在理解本公开时还需要了解另一个概念，碰撞发生时间(time to collision，简称TTC)。TTC是一个被广泛应用于车辆碰撞风险评估方面的参数，通常指车辆距离碰撞发生的时间，由车辆与危险处的相对距离与当前车速的比值获得。具体在如图1所示的车辆与行人交互场景下计算得出。其中，车辆到达潜在碰撞区域的时间为TTC v，计算过程TTC v＝D 1/v v，其中v v为车辆横向运动速度。行人避免碰撞所需的最短时间为TTC p，计算过程TTC p＝D 2/v p，其中v p为行人纵向(垂直于车辆行驶方向)运动速度。
设定车辆安装有车载感知系统。车载感知系统包括多种传感器，比如通常包括：视觉感知模块、毫米波雷达、超声波雷达、360°环视系统等。多源传感器的协同作用识别道路车道线、行人车辆等障碍物，为安全驾驶保驾护航。先进的车载感知系统可以在一定范围内识别行人信息，包括行人速度、位置、视野方向等，为车辆的碰撞风险判断及控制决策提供信息支撑。
本公开考虑到在交通运行过程中，行人的避让能力。行人避让能力指行人在发现危险时所采取的主动避让能力。根据交通事故调查及试验结果表明，行人遇到危险时其避让能力能够降低危险事故发生的风险。因此下述本公开中基于发明人进行的试验结果，对行人在避让过程中的速度进行了量化，定义为行人的避让能力。
请参阅图2，图2为本公开提供一种车辆与行人碰撞风险域的确定方法。
所述车辆与行人碰撞风险域的确定方法，包括：
S100，探测并输出车辆信息和行人信息。所述车辆信息包括：车辆位置、车辆速度、 车辆方向、车辆最大制动减速度和车辆最大侧向加速度。所述行人信息包括：行人位置、行人速度、行人方向和行人视线方向。所述车辆信息和所述行人信息均可由车载感知系统获得。
S200，判断行人是否注意到车辆。本步骤中，行人避让能力的激活取决于行人视野的方向，若行人视野的方向为关注到驶来的车辆，即确认行人避让能力被激活。若行人的视野方向未关注到驶来的车辆，则认为行人的避让能力没有被激活，行人继续保持正常的运动行为。若行人注意到车辆，则行人采取主动避让行为。若行人没有注意到车辆，则行人正常行走。在一个实施例中，行人采取的主动避让行为包括：急停后退避让或者加速前进避让。当然基于本公开的核心设计思路还可以将更多的行人避让行为设计到本方案中，以确定更加精确的碰撞风险域。
S300，在行人采取主动避让行为和行人不采取主动避让行为的情况下，分别进一步假设车辆是否采取即时反应动作。本步骤中，车辆采取的即时反应动作包括但不限于正常行驶、紧急制动和紧急转向。比如车辆采取的即时反应动作还可以包括借助道路上的其他设施进行减速制动。
S400，根据行人是否采取主动避让行为和车辆是否采取即时反应动作的假设结果，确定车辆与行人的碰撞风险域。本步骤中，在行人是否采用主动避让行为及车辆是否采取即时反应动作的情况下，车辆与行人的碰撞风险域是不同的。可以理解，在行人采用主动避让行为同时车辆采取即时反应动作的情况下，车辆与行人的碰撞风险域会更小。
本实施例中，首先采用车载探测系统探测车辆信息和行人信息。进一步判断行人是否采取主动避让行为以及车辆是否采取即时反应动作。在不同的情况下分别计算车辆与行人的碰撞风险域。本实施例中，车辆与行人碰撞风险域的确定方法，同时考虑到了行人的主动避让能力和车辆的即时反应动作，对于车辆与行人碰撞风险的识别更加充分。本公开中，在不同的情况下确定有效的碰撞风险域，可以有效提高车辆与行人交互过程中行人的安全性及车辆行驶的舒适性。通过判断行人是否注意到车辆，可以对行人面对危险工况时的有效避让行为进行分类和量化。另外，本实施例基于行人主动避让行为识别车辆-行人交互过程中的碰撞风险域对自动驾驶车辆的安全性提升具有重要的意义。
在一个实施例中所述车辆与行人碰撞风险域的确定方法中，所述行人采取主动避让行为包括：急停后退避让或者加速前进避让。
所述S400，根据行人是否采取主动避让行为和车辆是否采取即时反应动作的假设结果，确定车辆与行人的碰撞风险域的步骤包括：
S410，在行人采取主动避让行为的情况下，计算车辆与行人分别到达潜在碰撞点的时 间安全边界，所述时间安全边界即为TTC安全包络线。S420，在所述TTC安全包络线的基础上，进一步确定车辆与行人的碰撞风险域。
本实施例中，确定的是未考虑车辆的避让能力，在行人主动避让的情况下，得出车辆与行人碰撞的TTC安全包络线。本公开中，如果有需要还可以确定在考虑车辆的避让能力，并且在行人主动避让的情况下，得出车辆与行人碰撞的TTC安全包络线。
在一个实施例中，确定车辆与行人碰撞风险的TTC安全包络线的步骤，包括：
S411，确定车辆探测距离范围。本步骤中，车辆探测距离范围可以通过车载探测系统进行探测获取。
S412，根据行人急停后退避让速度、行人加速前进避让速度和车辆宽度计算行人安全避让车辆所需的最短距离。本步骤中，分为两种情况进行运算。根据行人急停后退避让速度和车辆宽度计算行人安全避让车辆所需的最短距离，为第一距离。在根据行人加速前进避让速度和车辆宽度计算行人安全避让车辆所需的最短距离，为第二距离。
S413，根据行人初始速度和行人安全避让车辆所需的最短距离计算行人碰撞时间的安全边界。根据行人初始速度，结合上述S412中计算的得出的所述第一距离和所述第二距离，分别计算行人碰撞时间的第一安全边界和第二安全边界。所述第一安全边界和所述第二安全边界为碰撞时间的安全边界。
S414，根据车辆碰撞时间和行人碰撞时间的安全边界，计算车辆与行人碰撞风险的TTC安全包络线。本步骤中，根据所述第一安全边界、所述第二安全边界和车辆行人碰撞时间范围，确定车辆与行人碰撞风险的TTC安全包络线。
本实施例中，给出了在不考虑车辆的避让能力时，确定或者计算行人主动避让能力下的TTC安全包络范围的具体步骤。当然，还可以在考虑车辆的避让能力时，确定或者计算行人主动避让能力下的TTC安全包络范围的具体步骤。
请参阅图3，提供了车辆-行人碰撞风险域的生成流程。图3中得出的碰撞风险域受车辆和行人两者的反应行为的影响。确定碰撞风险域的整体方法如图3所示，车辆自带的感知探测系统对车辆信息和行人信息进行探测。通过判断行人是否注意到车辆来确定行人避让能力的激活与否。行人避让能力的激活可以取决于行人视野的方向，若行人的视野方向为关注到驶来的车辆，即确认行人避让能力被激活。若行人的视野方向未关注到驶来的车辆，则认为行人的避让能力没有被激活，行人为继续保持正常的运动行为。根据目前的行人交通行为研究发现，行人面对危险时的有效避让行为为及时的后退避让或加速前进避让，因此本公开的发明人基于准真实交通场景下的行人志愿者试验，测得了行人面对危险工况采取避让行为下的行人急停后退避让速度(v pb)和行人前进加速避让速度(v pf)。具体的试 验数据请参阅图4和图5，图4为行人急停后退避让碰撞危险过程中的运动学特征图，t＝0时刻表示行人开始避让的时刻。图5为行人前进加速避让碰撞危险过程中的运动学特征图，t＝0时刻表示行人开始避让的时刻。
在一个实施例中，所述步骤S411-S414中，可以采用以下公式(1)至公式(7)计算车辆与行人碰撞风险的TTC安全包络线：
Figure PCTCN2020091061-appb-000001
{0≤TTC vd≤TTC vr}     (2)
Figure PCTCN2020091061-appb-000002
Figure PCTCN2020091061-appb-000003
TTC pb-ne(TTC vd)＝D pb-ne(TTC vd)/v pw    (5)
TTC pf-fe(TTC vd)＝D pf-fe(TTC vd)/v pw     (6)
[TTC vd,TTC pf-fe]≤TTCdangerous-area≤[TTC vd,TTC pb-ne]   (7)上述公式中出现的ne，是near-end的缩写，表示车辆靠近行人的一侧，叫近端。上述公式中出现的fe，是far-end的缩写，表示车辆远离行人的一侧，叫远端。D vr代表车辆探测到的最远距离；v v代表车辆的行驶速度；TTC vr代表当前车速下车辆到达最远探测处的时间；TTC vd代表车辆到达潜在碰撞的时间，TTC vd为一个变量；v pb代表行人急停后退的避让速度；v pf代表行人前进加速的避让速度；D pb-ne代表在行人急停后退避让时，行人安全避让车辆时所需要到达车辆近端的最短距离；TTC pf-fe代表在行人前进加速避让时，行人安全避让车辆时所需要距离车辆远端的最短距离；L vw代表车辆宽度；v pw代表行人开始避让前的初始速度；TTC pb-ne代表在行人急停后退避让时，行人安全避让车辆所需的最短时间；TTC pf-ne代表在行人前进加速避让时，行人安全避让车辆所需的最短时间；TTCdangerous-area代表车辆与行人碰撞风险的TTC安全包络线。
在考虑行人和车辆在险态工况下，其主动避让措施的是否触发来计算在实际交通环境中车辆与行人的相对安全距离。由于车辆与行人之间为动态的交互过程，车辆与行人的碰撞风险取决于车辆和行人当时的行驶速度、行驶方向、相对位置等信息。因此，本节为简 要说明车辆-行人二维碰撞风险域的生成方法，以某一已知的人车交互工况作为示例，计算该工况下的车辆-行人二维碰撞风险域。假设该工况为：车辆的行驶速度v v＝60km/h,车辆探测到的行人在车辆的右前方，步行速度为v pd＝1m/s，车辆和行人的运动方向垂直，车辆的最大制动减速度为a vx-max＝-7m/s 2，最大侧向加速度为a vy-max＝-6.5m/s 2，车辆的宽度L vw＝2m。车辆传感器探测最远距离范围为D vr＝100m，则该车辆探测的TTC范围为TTC vr＝6s，上述公式(2)可以表达为：{TTC vd:0≤TTC vd≤6s}。
本实施例中，参照上述公式(1)至公式(7)计算得出车辆与行人碰撞风险的TTC安全包络线，如图6所示。图6中通过行人后退避让的TTC边界和行人前进避让的TTC边界示出了车辆与行人的危险不可避免区域(如图6的阴影部分所示)。
在一个实施例中，所述S400，根据行人是否采取主动避让行为和车辆是否采取即时反应动作的假设结果，确定车辆与行人的碰撞风险域的步骤包括：
第一、在行人不采取主动避让行为同时车辆不采取即时反应动作的情况下，确定第一风险域。
在一个实施例中，在行人不采取主动避让行为，车辆不采取即时反应动作的情况下，根据以下公式(11)至公式(13)确定第一碰撞风险域：[D v-1,D p-1-fe]≤第一碰撞风险域≤[D v-1,D p-1-ne]；
Figure PCTCN2020091061-appb-000004
Figure PCTCN2020091061-appb-000005
Figure PCTCN2020091061-appb-000006
其中，v v代表车辆当前的行驶速度；TTC vd代表车辆到达潜在碰撞点的时间；v pd代表实际运动场景中车辆探测到的行人当前的运动速度；D v-1代表在TTC vd范围内，车辆在行驶方向的第一行驶范围；L vw代表车辆宽度；D p-1-ne代表行人以v pd的速度运动时，行人安全避让车辆时所需要距离车辆近端的第一最短距离；D p-1-fe代表行人以v pd的速度运动时，行人安全避让车辆时所需要距离车辆远端的第二最短距离。
第二、在行人不采取主动避让行为同时车辆采取紧急制动的情况下，确定第二风险域。
在一个实施例中，在行人不采取主动避让行为，车辆采取紧急制动的情况下，根据以 下公式(14)至公式(16)确定第二碰撞风险域：[D v-2,D p-2-fe]≤第二碰撞风险域≤[D v-2,D p-2-ne]；
Figure PCTCN2020091061-appb-000007
Figure PCTCN2020091061-appb-000008
Figure PCTCN2020091061-appb-000009
其中，v v代表车辆当前的行驶速度；TTC vd代表车辆到达潜在碰撞点的时间；v pd代表实际运动场景中车辆探测到的行人当前的运动速度；a vx-max代表车辆最大制动减速度；D v-2代表在TTC vd范围内，车辆在行驶方向的第二行驶范围；L vw代表车辆宽度；D p-2-ne代表行人以v pd的速度运动时，行人安全避让车辆时所需要距离车辆近端的第三最短距离；D p-2-fe代表行人以v pd的速度运动时，行人安全避让车辆时所需要距离车辆远端的第四最短距离。
第三、在行人采取主动避让行为同时车辆不采取即时反应动作的情况下，确定第三风险域。
在一个实施例中，在行人采取主动避让行为，车辆不采取即时反应动作的情况下，根据以下公式(17)至公式(19)确定第三碰撞风险域：[D v-3,D p-3-BA-ne]≤第三碰撞风险域≤[D v-3,D p-3-FA-fe]；
Figure PCTCN2020091061-appb-000010
Figure PCTCN2020091061-appb-000011
Figure PCTCN2020091061-appb-000012
其中，v v代表车辆当前的行驶速度；TTC vd代表车辆到达潜在碰撞点的时间；v pd代表实际运动场景中车辆探测到的行人当前的运动速度；TTC pb-ne代表在行人急停后退避让时，行人安全避让车辆所需的最短时间；TTC pf-ne代表在行人前进加速避让时，行人安全避让车辆所需的最短时间；L vw代表车辆宽度；D v-3代表在TTC vd范围内，车辆在行驶方向的第三行驶范围；D p-3-ne代表行人以v pd的速度运动时，行人安全避让车辆时所需要距离车辆 近端的第五最短距离；D p-3-fe代表行人以v pd的速度运动时，行人安全避让车辆时所需要距离车辆远端的第六最短距离。
第四、在行人采取主动避让行为同时车辆采取紧急制动的情况下，确定第四风险域。
在一个实施例中，在行人采取主动避让行为，车辆采取紧急制动的情况下，根据以下公式(20)至公式(22)确定第四碰撞风险域：[D v-4,D p-4-ne]≤第四碰撞风险域≤[D v-4,D p-4-fe]；
Figure PCTCN2020091061-appb-000013
Figure PCTCN2020091061-appb-000014
Figure PCTCN2020091061-appb-000015
其中，v v代表车辆当前的行驶速度；TTC vd代表车辆到达潜在碰撞点的时间；v pd代表实际运动场景中车辆探测到的行人当前的运动速度；a vx-max代表车辆最大制动减速度；TTC pb-ne代表在行人急停后退避让时，行人安全避让车辆所需的最短时间；TTC pf-ne代表在行人前进加速避让时，行人安全避让车辆所需的最短时间；L vw代表车辆宽度；D v-4代表在TTC vd范围内，车辆在行驶方向的第四行驶范围；D p-4-ne代表行人以v pd的速度运动时，行人安全避让车辆时所需要距离车辆近端的第七最短距离；D p-4-fe代表行人以v pd的速度运动时，行人安全避让车辆时所需要距离车辆远端的第八最短距离。
在一个实施例中，在探测并输出车辆信息和行人信息的步骤之后，还包括：在车辆紧急制动以及车辆紧急转向的情况下，确定第五风险域。
在一个实施例中，当考虑车辆的避让能力时，所述车辆即时反应动作包括车辆正常行驶和车辆紧急制动(车辆紧急制动包括直行紧急制动和紧急转弯)。可以根据车辆当前行驶速度和车辆最大制动减速度，确定车辆的制动距离。或者可以根据所述车辆当前行驶速度和车辆最大侧向加速度，确定车辆的最小转弯半径。
具体的，车辆制动减速度是指车辆在行驶中迅速降低行驶速度直至停车的能力。车辆的最大制动减速度取决于车辆轮带与地面之间的摩擦系数，在实际工况中，地面与轮胎的摩擦系数通常为0.6至0.8，即车辆的制动减速度通常为6m/s 2至8m/s 2。
车辆最大侧向加速度即汽车横向加速度，指的是与汽车行驶方向垂直的方向的加速度，在车辆进行转弯行驶时产生的离心力所带来的加速度。也就是车被“甩飞”的趋势。这个 加速度越大车子理论上就容易被“甩”离行驶路径。因此，车辆在行驶过程中的极限转向性能却决于车辆的最大侧向加速度。
在一个实施例中，参考公式(8)至公式(10)考虑车辆的避让能力，车辆的避让能力具体指车辆在行驶状态下的制动能力和转向能力，车辆的制动能力即制动距离(D vb)取决于车辆当前的行驶速度(v v)和最大制动减速度(a vx-max)。
D vb＝(v v) 2/2a vx-max     (8)
v v代表车辆当前的行驶速度；a vx-max代表车辆最大制动减速度。
车辆的转向能力指当前行驶速度下保持车辆稳定性的最小转弯半径(R vd-min)，取决于车辆当前的行驶速度(v v)和最大侧向加速度(a vy-max)：
R v-min＝v v/w v          (9)
w v＝v v/a vy-max,(R vd-min≥R v-min)       (10)
w v代表车辆横摆角速度，R v-min代表车辆最小转弯半径(属于车辆参数，小于或等于当前行驶速度下保持车辆稳定性的最小转弯半径R vd-min)。
本实施例中，考虑了车辆的避让能力，在所述车辆即时反应动作包括车辆正常行驶和车辆紧急制动。而车辆紧急制动有可以包括直行情况下的紧急制动和转弯时的紧急制动。本实施例中，将车辆的避让能力考虑进来，使得车辆与行人的碰撞风险域的确定可以更准确。
在一个实施例中，在车辆紧急制动和车辆紧急转弯的情况下，根据以下公式(23)至公式(28)确定车辆与行人的第五碰撞风险域；[D v-5,D vl-5-fe]≤第五碰撞风险域≤[D v-5,D vl-5-ne]；
Figure PCTCN2020091061-appb-000016
v vb(TTC vd)＝v v+a vx-max×TTC vd        (24)
R vb(TTC vd)＝(v vb) 2/a vy-max       (25)
Figure PCTCN2020091061-appb-000017
Figure PCTCN2020091061-appb-000018
Figure PCTCN2020091061-appb-000019
其中，v v代表车辆的行驶速度；v vb代表制动过程中车辆的速度；a vx-max代表车辆最大制动减速度；a vy-max代表车辆最大侧向加速度；TTC vd代表车辆到达潜在碰撞点的时间；A vs代表车辆转向过程中的累计转向角度；D v-5代表在TTC vd范围内，车辆在行驶方向的第五行驶范围；D p-5-ne代表行人以v v的速度运动时，行人安全避让车辆时所需要距离车辆近端的第九最短距离；D p-5-fe代表行人以v v的速度运动时，行人安全避让车辆时所需要距离车辆远端的第十最短距离。
本实施例中，在计算所述第五风险域的过程中，本公开还明确了转向对风险降低的有效性。其中，车辆对危险工况通常采取制动行为，不能判断其制动和转向碰撞的风险。本实施例中，考虑了车辆即时反应动作(直行情况下的紧急制动和转弯时的紧急制动)，将车辆的避让能力考虑进来，使得车辆与行人的碰撞风险域的确定可以更准确。
请参阅图7，图7中示出了车辆与行人碰撞风险域示意图。图7中车辆当前的运行工况为：车辆的行驶速度v v＝60km/h,车辆探测到的行人在车辆的右前方，步行速度为v pd＝1m/s，车辆和行人的运动方向垂直，车辆的最大制动减速度为a vx-max＝-7m/s 2，最大侧向加速度为a vy-max＝-6.5m/s 2，车辆的宽度L vw＝2m。车辆传感器探测最远距离范围为D vr＝100m，则该车辆探测的TTC范围为TTC vr＝6s。在结合上述实施例中得出的所述第一碰撞风险域、所述第二碰撞风险域、所述第三碰撞风险域、所述第四碰撞风险域和所述第五碰撞风险域之后，可以对车辆给出碰撞风险更小的、更加正确合理的执行策略。
图7中的碰撞风险域1为上述实施例中计算得出的所述第一碰撞风险域。图7中的碰撞风险域2为上述实施例中计算得出的所述第二碰撞风险域。图7中的碰撞风险域3为上述实施例中计算得出的所述第三碰撞风险域。图7中的碰撞风险域4为上述实施例中计算得出的所述第四碰撞风险域。图7中的碰撞风险域5为上述实施例中计算得出的所述第五碰撞风险域。比如，按照图7中示意的碰撞风险域，可以对车辆给出如下的执行策略。1、对位于碰撞风险域1减去碰撞风险域2、碰撞风险域3所剩区域内的行人，车辆可通过主动避让(制动或转向)或发出警报提醒行人注意车辆并主动避让来避免碰撞发生。2、对位于碰撞风险域2减去碰撞风险域3、碰撞风险域4所剩区域内的行人，车辆可通过主动避让(转向)或发出警报提醒行人注意车辆并主动避让来避免碰撞发生。3、对位于碰撞风险域3减去碰撞风险域4所剩区域内的行人，行人的主动避让行为已经不能有效避免碰撞发生，车辆只能通过主动避让(制动或转向)来避免碰撞发生。4、对位于碰撞风险域4减去 碰撞风险域5所剩区域内的行人，行人的主动避让行为已经不能有效避免碰撞发生，并且车辆若只采取制动行为依然不能避免碰撞，车辆只能通过主动紧急转向来避免碰撞发生。5、对位于碰撞风险域4与碰撞风险域5交汇区域内的行人，车辆采取任何措施都不能避免碰撞。
在本公开上述记载中，对行人面对危险工况时的有效避让行为进行了分类和量化。并且，本公开综合考虑人-车的避让能力对风险区域进行划分。本公开的基于行人主动避让行为识别车辆-行人交互过程中确定的碰撞风险域对自动驾驶车辆的安全性提升具有重要的意义。
在一个具体的实施例中，本公开的应用场景为具有主动探测能力的车辆上。该车辆可以检测自身车辆的信息包括：车辆的速度、制动减速度、侧向加速度等信息。同时该车辆可以识别探测范围内的行人，并探测行人的位置、速度、移动方向及视野方向。上述车辆获得的信息作为输入，以车辆和行人的在危险态工况下的避让能力作为计算参数，可在车辆行驶过程中实时计算车辆与行人交互过程中的存在的碰撞风险。
在一个实施例中，本公开还提供一种车辆与行人碰撞风险域的确定系统。所述车辆与行人碰撞风险域的确定系统包括：探测模块、第一分析判断模块、第二分析判断模块和运算模块。
探测模块，用于探测并输出车辆信息和行人信息。
第一分析判断模块，用于判断行人是否注意到车辆。若行人注意到车辆，则行人采取主动避让行为，若行人没有注意到车辆，则行人正常行走。
第二分析判断模块，用于在行人采取主动避让行为和行人不采取主动避让行为的情况下，分别进一步假设车辆是否采取即时反应动作。
运算模块，用于根据行人是否采取主动避让行为和车辆是否采取即时反应动作的假设结果，确定车辆与行人的碰撞风险域。
本实施例中，车辆与行人碰撞风险域的确定系统，同时考虑到了行人的主动避让能力和车辆的即时反应动作，对于车辆与行人碰撞风险的识别更加充分。本公开中，在不同的情况下确定有效的碰撞风险域，可以有效提高车辆与行人交互过程中行人的安全性及车辆行驶的舒适性。通过判断行人是否注意到车辆，可以对行人面对危险工况时的有效避让行为进行分类和量化。另外，本实施例基于行人主动避让行为识别车辆-行人交互过程中的碰撞风险域对自动驾驶车辆的安全性提升具有重要的意义。
请参阅图8，本公开提供一种车辆与行人碰撞风险的评价方法，包括：
S10，探测并输出车辆信息和行人信息。所述车辆信息包括：车辆位置、车辆速度、车 辆方向、车辆最大制动减速度和车辆最大侧向加速度；所述行人信息包括：行人位置、行人速度、行人方向和行人视线方向。
S20，判断行人是否注意到车辆，若行人注意到车辆，则行人采取主动避让行为，若行人没有注意到车辆，则行人正常行走。
S30，在行人采取主动避让行为和行人不采取主动避让行为的情况下，分别确定车辆与行人的碰撞风险域。在一个实施例中，行人采取的主动避让行为包括：急停后退避让或者加速前进避让。当然基于本公开的核心设计思路还可以将更多的行人避让行为设计到本方案中，以确定更加精确的碰撞风险域。
S40，判断行人是否在所述碰撞风险域的范围内。这里根据当前探测到的行人所在位置、车辆所在位置以及确定的所述碰撞风险域范围来进行判断。
S50，根据行人是否在所述碰撞风险域的范围内的判断结果，评价车辆与行人的碰撞风险。若行人不在所述碰撞风险域的范围内，则车辆与行人的碰撞风险较低。若行人在所述碰撞风险域的范围内，则车辆与行人的碰撞风险较高。具体的，碰撞风险的概率可以结合上述实施例中得出的五种不同的碰撞风险域进行确定。比如，所述第五风险域的碰撞风险的概率最大，因为当行人处于所述第五碰撞风险域中时，无论车辆采取任何措施都不能避免碰撞。
在一个实施例中，在所述车辆与行人碰撞风险的评价方法中，在行人采取主动避让行为和行人不采取主动避让行为的情况下，分别确定车辆与行人的碰撞风险域的步骤，还包括：
假设车辆是否采取即时反应动作，所述车辆即时反应动作包括车辆正常行驶和车辆紧急制动。其中，车辆紧急制动又包括：直行紧急制动和紧急转弯。
结合行人是否采取主动避让行为和车辆是否采取即时反应动作，确定车辆与行人的碰撞风险域。
本实施例中，确定所述碰撞风险域的具体方法可以参考上述车辆与行人碰撞风险域的确定方法中的步骤进行确定，在此不再赘述。
在一个实施例中，在所述车辆与行人碰撞风险的评价方法中，所述根据行人是否在所述碰撞风险域的范围内的判断结果，评价车辆与行人的碰撞风险的步骤包括：
位于所述第一碰撞风险域减去所述第二碰撞风险域、减去所述第三碰撞风险域所剩区域内的行人，车辆与行人的碰撞风险为一级风险。当车辆与行人的碰撞风险为一级风险时，可通过车辆主动避让(制动或转向)或发出警报提醒行人注意车辆并主动避让来避免碰撞发生。
位于所述第二碰撞风险域减去所述第三碰撞风险域、减去所述第四碰撞风险域所剩区域内的行人，车辆与行人的碰撞风险为二级风险。当车辆与行人的碰撞风险为二级风险时，可通过车辆主动避让(转向)或发出警报提醒行人注意车辆并主动避让来避免碰撞发生。
位于所述第三碰撞风险域减去所述碰撞风险域所剩区域内的行人，车辆与行人的碰撞风险为三级风险。当车辆与行人的碰撞风险为三级风险时，行人的主动避让行为已经不能有效避免碰撞发生，车辆只能通过主动避让(制动或转向)来避免碰撞发生。
位于所述第四碰撞风险域减去所述第五碰撞风险域所剩区域内的行人，车辆与行人的碰撞风险为四级风险。当车辆与行人的碰撞风险为四级风险时，行人的主动避让行为已经不能有效避免碰撞发生，并且车辆若只采取制动行为依然不能避免碰撞，车辆只能通过主动紧急转向来避免碰撞发生。
位于所述第四碰撞风险域与所述第五碰撞风险域交汇区域内的行人，车辆与行人的碰撞风险为五级风险，其中所述一级风险的风险等级最低，所述五级风险的风险等级最高。当车辆与行人的碰撞风险为五级风险时，车辆采取任何措施都不能避免碰撞。
本公开提供的所述车辆与行人碰撞风险的评价方法，明确了行人在危险工况下的主动避让能力，生成的车辆-行人碰撞风险域考虑了行人位置、速度、主动避让能力等因素的耦合影响，对车辆-行人碰撞风险的识别更加充分。
本公开提供的所述车辆与行人碰撞风险的评价方法，综合考虑车辆的制动和转向能力以及行人的避让能力，提出了基于行人主动避让能力的，多工况下的车辆-行人碰撞风险域生成方法。对于提高智能车辆对行人风险的识别具有重要意义，可以有效提高车辆-行人交互过程中行人的安全性及车辆行驶的舒适性。
请参阅图9，本公开还提供一种车辆与行人碰撞风险的评价系统100，包括：探测模块10、分析判断模块20、运算模块30和评价模块40。
所述探测模块10用于探测车辆信息和行人信息。
所述分析判断模块20与所述探测模块10连接。所述分析判断模块20用于判断行人是否注意到车辆。若行人注意到车辆，则行人采取主动避让行为，若行人没有注意到车辆，则行人正常行走。
所述运算模块30与所述分析判断模块20连接。所述运算模块30用于在行人采取主动避让行为和行人不采取主动避让行为的情况下，分别确定车辆与行人的碰撞风险域。
所述评价模块40与所述运算模块30连接。所述评价模块40用于根据行人是否在所述碰撞风险域的范围内，来评价车辆与行人的碰撞风险。
本实施例中，上述模块可以依托计算机程序实现，模块的具体硬件结构并不做具体的 限定，能实现上述功能即可。本实施例中提供的所述车辆与行人碰撞风险的评价系统100，可以执行所述车辆与行人碰撞风险的评价方法中的所有步骤。所述车辆与行人碰撞风险的评价系统100也综合考虑车辆的制动和转向能力以及行人的避让能力，提出了基于行人主动避让能力的，多工况下的车辆-行人碰撞风险域生成方法。对于提高智能车辆对行人风险的识别具有重要意义，可以有效提高车辆-行人交互过程中行人的安全性及车辆行驶的舒适性。
以上所述实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的各个技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。
以上所述实施例仅表达了本公开的几种实施方式，其描述较为具体和详细，但并不能因此而理解为对发明专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本公开构思的前提下，还可以做出若干变形和改进，这些都属于本公开的保护范围。因此，本公开专利的保护范围应以所附权利要求为准。

Claims (18)

  1. 一种车辆与行人碰撞风险域的确定方法，其特征在于，包括：
    探测并输出车辆信息和行人信息；
    判断行人是否注意到车辆，若行人注意到车辆，则行人采取主动避让行为，若行人没有注意到车辆，则行人正常行走；
    在行人采取主动避让行为和行人不采取主动避让行为的情况下，分别进一步假设车辆是否采取即时反应动作；
    根据行人是否采取主动避让行为的判断结果和车辆是否采取即时反应动作的假设结果，确定车辆与行人的碰撞风险域。
  2. 根据权利要求1所述的车辆与行人碰撞风险域的确定方法，其特征在于，
    所述行人采取主动避让行为包括：急停后退避让或者加速前进避让；
    所述根据行人是否采取主动避让行为的判断结果和车辆是否采取即时反应动作的假设结果，确定车辆与行人的碰撞风险域的步骤包括：
    在行人采取主动避让行为的情况下，确定车辆与行人碰撞风险的TTC安全包络线；
    在所述TTC安全包络线的基础上，进一步确定车辆与行人的碰撞风险域；
    其中，确定车辆与行人碰撞风险的TTC安全包络线的步骤，包括：
    确定车辆探测距离范围；
    根据行人急停后退避让速度、行人加速前进避让速度和车辆宽度计算行人安全避让车辆所需的最短距离；
    根据行人初始速度和行人安全避让车辆所需的最短距离计算行人避让碰撞所需的时间，即行人避免碰撞所需最短时间；
    根据车辆到达潜在碰撞点的时间和行人避免碰撞所需的最短时间，计算车辆与行人分别到达潜在碰撞点的时间安全边界，所述时间安全边界即为TTC安全包络线。
  3. 根据权利要求2所述的车辆与行人碰撞风险域的确定方法，其特征在于，根据行人是否采取主动避让行为和车辆是否采取即时反应动作的假设结果，确定车辆与行人的碰撞风险域的步骤包括：
    在行人不采取主动避让行为同时车辆不采取即时反应动作的情况下，确定第一风险域；
    在行人不采取主动避让行为同时车辆采取紧急制动的情况下，确定第二风险域；
    在行人采取主动避让行为同时车辆不采取即时反应动作的情况下，确定第三风险域；以及
    在行人采取主动避让行为同时车辆采取紧急制动的情况下，确定第四风险域。
  4. 根据权利要求3所述的车辆与行人碰撞风险域的确定方法，其特征在于，在探测并输出车辆信息和行人信息的步骤之后，还包括：
    在车辆紧急制动以及车辆紧急转向的情况下，确定第五风险域。
  5. 根据权利要求4所述的车辆与行人碰撞风险域的确定方法，其特征在于，采用以下公式计算车辆与行人碰撞风险的TTC安全包络线：
    Figure PCTCN2020091061-appb-100001
    {0≤TTC vd≤TTC vr}
    Figure PCTCN2020091061-appb-100002
    Figure PCTCN2020091061-appb-100003
    TTC pb-ne(TTC vd)＝D pb-ne(TTC vd)/v pw
    TTC pf-fe(TTC vd)＝D pf-fe(TTC vd)/v pw
    [TTC vd,TTC pf-fe]≤TTCdangerous-area≤[TTC vd,TTC pb-ne]
    D vr代表车辆探测到的最远距离；v v代表车辆的行驶速度；TTC vr代表当前车速下车辆到达最远探测处的时间；TTC vd代表车辆到达潜在碰撞的时间，TTC vd是一个变量；v pb代表行人急停后退的避让速度；v pf代表行人前进加速的避让速度；D pb-ne代表在行人急停后退避让时，行人安全避让车辆时所需要到达车辆近端的最短距离；TTC pf-fe代表在行人前进加速避让时，行人安全避让车辆时所需要距离车辆远端的最短距离；L vw代表车辆宽度；v pw代表行人开始避让前的初始速度；TTC pb-ne代表在行人急停后退避让时，行人安全避让车辆所需的最短时间；TTC pf-ne代表在行人前进加速避让时，行人安全避让车辆所需的最短时间；TTCdangerous-area代表车辆与行人碰撞风险的TTC安全包络线。
  6. 根据权利要求5所述的车辆与行人碰撞风险域的确定方法，其特征在于，在行人不采取主动避让行为，车辆不采取即时反应动作的情况下，根据以下公式确定第一碰撞风险域：[D v-1,D p-1-fe]≤第一碰撞风险域≤[D v-1,D p-1-ne]；
    Figure PCTCN2020091061-appb-100004
    Figure PCTCN2020091061-appb-100005
    Figure PCTCN2020091061-appb-100006
    其中，v v代表车辆当前的行驶速度；TTC vd代表车辆到达潜在碰撞点的时间；v pd代表实际运动场景中车辆探测到的行人当前的运动速度；D v-1代表在TTC vd范围内，车辆在行驶方向的第一行驶范围；L vw代表车辆宽度；D p-1-ne代表行人以v pd的速度运动时，行人安全避让车辆时所需要距离车辆近端的第一最短距离；D p-1-fe代表行人以v pd的速度运动时，行人安全避让车辆时所需要距离车辆远端的第二最短距离。
  7. 根据权利要求5所述的车辆与行人碰撞风险域的确定方法，其特征在于，在行人不采取主动避让行为，车辆采取紧急制动的情况下，根据以下公式确定第二碰撞风险域：[D v-2,D p-2-fe]≤第二碰撞风险域≤[D v-2,D p-2-ne]；
    Figure PCTCN2020091061-appb-100007
    Figure PCTCN2020091061-appb-100008
    Figure PCTCN2020091061-appb-100009
    其中，v v代表车辆当前的行驶速度；TTC vd代表车辆到达潜在碰撞点的时间；v pd代表实际运动场景中车辆探测到的行人当前的运动速度；a vx-max代表车辆最大制动减速度；D v-2代表在TTC vd范围内，车辆在行驶方向的第二行驶范围；L vw代表车辆宽度；D p-2-ne代表行人以v pd的速度运动时，行人安全避让车辆时所需要距离车辆近端的第三最短距离；D p-2-fe代表行人以v pd的速度运动时，行人安全避让车辆时所需要距离车辆远端的第四最短距离。
  8. 根据权利要求5所述的车辆与行人碰撞风险域的确定方法，其特征在于，在行人采取主动避让行为，车辆不采取即时反应动作的情况下，根据以下公式确定第三碰撞风险域：[D v-3,D p-3-BA-ne]≤第三碰撞风险域≤[D v-3,D p-3-FA-fe]；
    Figure PCTCN2020091061-appb-100010
    Figure PCTCN2020091061-appb-100011
    Figure PCTCN2020091061-appb-100012
    其中，v v代表车辆当前的行驶速度；TTC vd代表车辆到达潜在碰撞点的时间；v pd代表实际运动场景中车辆探测到的行人当前的运动速度；TTC pb-ne代表在行人急停后退避让时，行人安全避让车辆所需的最短时间；TTC pf-ne代表在行人前进加速避让时，行人安全避让车辆所需的最短时间；L vw代表车辆宽度；D v-3代表在TTC vd范围内，车辆在行驶方向的第三行驶范围；D p-3-ne代表行人以v pd的速度运动时，行人安全避让车辆时所需要距离车辆近端的第五最短距离；D p-3-fe代表行人以v pd的速度运动时，行人安全避让车辆时所需要距离车辆远端的第六最短距离。
  9. 根据权利要求3所述的车辆与行人碰撞风险域的确定方法，其特征在于，在行人采取主动避让行为，车辆采取紧急制动的情况下，根据以下公式确定第四碰撞风险域：[D v-4,D p-4-ne]≤第四碰撞风险域≤[D v-4,D p-4-fe]；
    Figure PCTCN2020091061-appb-100013
    Figure PCTCN2020091061-appb-100014
    Figure PCTCN2020091061-appb-100015
    其中，v v代表车辆当前的行驶速度；TTC vd代表车辆到达潜在碰撞点的时间；v pd代表实际运动场景中车辆探测到的行人当前的运动速度；a vx-max代表车辆最大制动减速度；TTC pb-ne代表在行人急停后退避让时，行人安全避让车辆所需的最短时间；TTC pf-ne代表在行人前进加速避让时，行人安全避让车辆所需的最短时间；L vw代表车辆宽度；D v-4代表在TTC vd范围内，车辆在行驶方向的第四行驶范围；D p-4-ne代表行人以v pd的速度运动时，行人安全避让车辆时所需要距离车辆近端的第七最短距离；D p-4-fe代表行人以v pd的速度运动时，行人安全避让车辆时所需要距离车辆远端的第八最短距离。
  10. 根据权利要求3所述的车辆与行人碰撞风险域的确定方法，其特征在于，当考虑车辆采取即时反应动作的情况下，所述车辆的即时反应动作包括车辆正常行驶、车辆直行紧急制动和车辆紧急转弯；
    当所述车辆采取的即时反应动作为车辆直行紧急制动时，根据所述车辆的当前行驶速度和所述车辆的最大制动减速度，确定所述车辆的制动距离；
    当所述车辆采取的即时反应动作为车辆紧急转弯时，根据所述车辆的当前行驶速度和所述车辆的最大侧向加速度，确定所述车辆的最小转弯半径。
  11. 根据权利要求10所述的车辆与行人碰撞风险域的确定方法，其特征在于，当考虑车辆采取即时反应动作的情况下，参考以下公式计算所述车辆的制动距离D vb和所述车辆稳定性的最小转弯半径R vd-min：
    D vb＝(v v) 2/2a vx-max
    v v代表车辆当前的行驶速度；a vx-max代表车辆最大制动减速度；
    R v-min＝v v/w v   (9)
    w v＝v v/a vy-max,(R vd-min≥R v-min)  (10)
    w v代表车辆横摆角速度，R v-min代表车辆最小转弯半径，a vy-max代表最大侧向加速度。
  12. 根据权利要求11所述的车辆与行人碰撞风险域的确定方法，其特征在于，在车辆紧急制动和车辆紧急转弯的情况下，根据以下公式确定车辆与行人的第五碰撞风险域；[D v-5,D v1-5-fe]≤第五碰撞风险域≤[D V-5,D v1-5-ne]；
    Figure PCTCN2020091061-appb-100016
    v vb(TTC vd)＝v v+a vx-max×TTC vd
    R vb(TTC vd)＝(v vb(TTC vd)) 2/a vy-max
    Figure PCTCN2020091061-appb-100017
    Figure PCTCN2020091061-appb-100018
    Figure PCTCN2020091061-appb-100019
    其中，v v代表车辆的行驶速度；v vb代表制动过程中车辆的速度；a vx-max代表车辆最大制动减速度；a vy-max代表车辆最大侧向加速度；TTC vd代表车辆到达潜在碰撞点的时间；A vs代表车辆转向过程中的累计转向角度；D v-5代表在TTC vd范围内，车辆在行驶方向的第五行驶范围；D p-5-ne代表行人以v v的速度运动时，行人安全避让车辆时所需要距离车辆近端的第九最短距离；D p-5-fe代表行人以v v的速度运动时，行人安全避让车辆时所需要距离车 辆远端的第十最短距离。
  13. 一种车辆与行人碰撞风险的评价方法，其特征在于，包括：
    探测并输出车辆信息和行人信息；
    判断行人是否注意到车辆，若行人注意到车辆，则行人采取主动避让行为，若行人没有注意到车辆，则行人正常行走；
    在行人采取主动避让行为和行人不采取主动避让行为的情况下，分别确定车辆与行人的碰撞风险域；
    判断行人是否在所述碰撞风险域的范围内；
    根据行人是否在所述碰撞风险域的范围内的判断结果，评价车辆与行人的碰撞风险。
  14. 根据权利要求13所述的车辆与行人碰撞风险的评价方法，其特征在于，在行人采取主动避让行为和行人不采取主动避让行为的情况下，分别确定车辆与行人的碰撞风险域的步骤，还包括：
    假设车辆是否采取即时反应动作；
    结合行人是否采取主动避让行为和车辆是否采取即时反应动作，确定车辆与行人的碰撞风险域。
  15. 根据权利要求14所述的车辆与行人碰撞风险的评价方法，其特征在于，
    所述行人采取主动避让行为包括：急停后退避让或者加速前进避让；
    所述根据行人是否采取主动避让行为的判断结果和车辆是否采取即时反应动作的假设结果，确定车辆与行人的碰撞风险域的步骤包括：
    在行人采取主动避让行为的情况下，确定车辆与行人碰撞风险的TTC安全包络线；
    在所述TTC安全包络线的基础上，进一步确定车辆与行人的碰撞风险域；
    其中，确定车辆与行人碰撞风险的TTC安全包络线的步骤，包括：
    确定车辆探测距离范围；
    根据行人急停后退避让速度、行人加速前进避让速度和车辆宽度计算行人安全避让车辆所需的最短距离；
    根据行人初始速度和行人安全避让车辆所需的最短距离计算行人避让碰撞所需的时间，即行人避免碰撞所需最短时间；
    根据车辆到达潜在碰撞点的时间和行人避免碰撞所需的最短时间，计算车辆与行人分别到达潜在碰撞点的时间安全边界，所述时间安全边界即为TTC安全包络线。
  16. 根据权利要求15所述的车辆与行人碰撞风险的评价方法，其特征在于，根据行人是否采取主动避让行为和车辆是否采取即时反应动作的假设结果，确定车辆与行人的碰撞 风险域的步骤包括：
    在行人不采取主动避让行为同时车辆不采取即时反应动作的情况下，确定第一风险域；
    在行人不采取主动避让行为同时车辆采取紧急制动的情况下，确定第二风险域；
    在行人采取主动避让行为同时车辆不采取即时反应动作的情况下，确定第三风险域；
    在行人采取主动避让行为同时车辆采取紧急制动的情况下，确定第四风险域；以及
    在车辆紧急制动以及车辆紧急转向的情况下，确定第五风险域。
  17. 根据权利要求16所述的车辆与行人碰撞风险的评价方法，其特征在于，所述根据行人是否在所述碰撞风险域的范围内的判断结果，评价车辆与行人的碰撞风险的步骤包括：
    位于所述第一碰撞风险域减去所述第二碰撞风险域、减去所述第三碰撞风险域所剩区域内的行人，车辆与行人的碰撞风险为一级风险；
    位于所述第二碰撞风险域减去所述第三碰撞风险域、减去所述第四碰撞风险域所剩区域内的行人，车辆与行人的碰撞风险为二级风险；
    位于所述第三碰撞风险域减去所述碰撞风险域所剩区域内的行人，车辆与行人的碰撞风险为三级风险；
    位于所述第四碰撞风险域减去所述第五碰撞风险域所剩区域内的行人，车辆与行人的碰撞风险为四级风险；
    位于所述第四碰撞风险域与所述第五碰撞风险域交汇区域内的行人，车辆与行人的碰撞风险为五级风险，其中所述一级风险的风险等级最低，所述五级风险的风险等级最高。
  18. 一种车辆与行人碰撞风险的评价系统，其特征在于，包括：
    探测模块，用于探测车辆信息和行人信息；
    分析判断模块，与所述探测模块连接，用于判断行人是否注意到车辆，若行人注意到车辆，则行人采取主动避让行为，若行人没有注意到车辆，则行人正常行走；
    运算模块，与所述分析判断模块连接，用于在行人采取主动避让行为和行人不采取主动避让行为的情况下，分别确定车辆与行人的碰撞风险域；以及
    评价模块，与所述运算模块连接，用于根据行人是否在所述碰撞风险域的范围内，来评价车辆与行人的碰撞风险。
PCT/CN2020/091061 2020-04-27 2020-05-19 车辆-行人碰撞风险域的计算方法及安全评价系统 WO2021217752A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/920,805 US20230192078A1 (en) 2020-04-27 2020-05-19 Method for calculating risk domain of vehicle-pedestrian collision and safety evaluation system

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010344141.0 2020-04-27
CN202010344141.0A CN111497840B (zh) 2020-04-27 2020-04-27 车辆-行人碰撞风险域的计算方法及安全评价系统

Publications (1)

Publication Number Publication Date
WO2021217752A1 true WO2021217752A1 (zh) 2021-11-04

Family

ID=71868062

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/091061 WO2021217752A1 (zh) 2020-04-27 2020-05-19 车辆-行人碰撞风险域的计算方法及安全评价系统

Country Status (3)

Country Link
US (1) US20230192078A1 (zh)
CN (1) CN111497840B (zh)
WO (1) WO2021217752A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113895460A (zh) * 2021-11-11 2022-01-07 中国第一汽车股份有限公司 行人轨迹预测方法、装置及存储介质
CN114170805A (zh) * 2021-12-23 2022-03-11 南京理工大学 基于aeb碰撞速度的车辆路径规划方法
CN114379587A (zh) * 2021-12-28 2022-04-22 阿波罗智联(北京)科技有限公司 自动驾驶中避让行人的方法与装置

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114426030B (zh) * 2020-10-29 2024-04-09 华为技术有限公司 一种行人穿行意图估计方法、装置、设备和汽车
CN112650243B (zh) 2020-12-22 2023-10-10 北京百度网讯科技有限公司 车辆控制方法、装置、电子设备和自动驾驶车辆
CN114023107A (zh) * 2021-11-01 2022-02-08 深圳智慧车联科技有限公司 预防城市渣土车右转过程与非机动车发生冲突的方法、系统及其存储介质
CN115273539B (zh) * 2022-06-16 2024-01-30 中国第一汽车股份有限公司 一种基于v2x通信的车辆危险预警方法、装置和计算机可读存储介质
CN116223056B (zh) * 2022-12-14 2024-03-12 清华大学 虚拟碰撞测试方法、装置、设备、存储介质和程序产品
CN116187748B (zh) * 2022-12-16 2023-12-29 清华大学 风险域确定方法、装置、计算机设备、介质和程序产品
CN116872881A (zh) * 2023-08-09 2023-10-13 清华大学 车辆行人主被动安全协同控制方法、装置和计算机设备

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103303307A (zh) * 2012-03-13 2013-09-18 通用汽车环球科技运作有限责任公司 驾驶员辅助系统
CN106428000A (zh) * 2016-09-07 2017-02-22 清华大学 一种车辆速度控制装置和方法
CN108263380A (zh) * 2016-12-30 2018-07-10 现代自动车株式会社 考虑行人注视方式的行人碰撞预防装置和方法
CN110834631A (zh) * 2019-11-01 2020-02-25 中国第一汽车股份有限公司 一种行人避让方法、装置、车辆及存储介质

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9809219B2 (en) * 2014-01-29 2017-11-07 Continental Automotive Systems, Inc. System for accommodating a pedestrian during autonomous vehicle operation
JP6354746B2 (ja) * 2015-12-24 2018-07-11 マツダ株式会社 運転支援装置
JP2017114405A (ja) * 2015-12-25 2017-06-29 マツダ株式会社 運転支援装置
FR3066731B1 (fr) * 2017-05-29 2020-10-30 Peugeot Citroen Automobiles Sa Dispositif et procede d’assistance par eclairage orientable en fonction de la position d’un etre vivant situe devant un vehicule
JP2020003832A (ja) * 2018-06-23 2020-01-09 株式会社Ｍｏｍｏ 自動運転する車両の走行を制御するためのコンピュータ、車両、プログラムおよび方法
CN110544377A (zh) * 2019-08-31 2019-12-06 武汉理工大学 基于车路协同的交叉口行人避撞方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103303307A (zh) * 2012-03-13 2013-09-18 通用汽车环球科技运作有限责任公司 驾驶员辅助系统
CN106428000A (zh) * 2016-09-07 2017-02-22 清华大学 一种车辆速度控制装置和方法
CN108263380A (zh) * 2016-12-30 2018-07-10 现代自动车株式会社 考虑行人注视方式的行人碰撞预防装置和方法
CN110834631A (zh) * 2019-11-01 2020-02-25 中国第一汽车股份有限公司 一种行人避让方法、装置、车辆及存储介质

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
LU SIWEN, FANG SHOUEN, WANG JUNHUA: "Mechanism Analysis on Conflicts and Collisions Caused by Aggressive Pedestrian Crossing Single Lane", JOURNAL OF CHONGQING JIAOTONG UNIVERSITY (NATURAL SCIENCE), vol. 28, no. 1, 28 February 2009 (2009-02-28), pages 108 - 110, XP055861979, ISSN: 1674-0696 *
ZHIJUN CHEN; CHAOZHONG WU; NENGCHAO LYU; GANG LIU; YI HE: "Pedestrian-vehicular collision avoidance based on vision system", 17TH INTERNATIONAL IEEE CONFERENCE ON INTELLIGENT TRANSPORTATION SYSTEMS (ITSC), IEEE, 8 October 2014 (2014-10-08), pages 11 - 15, XP032685905, DOI: 10.1109/ITSC.2014.6957658 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113895460A (zh) * 2021-11-11 2022-01-07 中国第一汽车股份有限公司 行人轨迹预测方法、装置及存储介质
CN113895460B (zh) * 2021-11-11 2023-01-13 中国第一汽车股份有限公司 行人轨迹预测方法、装置及存储介质
CN114170805A (zh) * 2021-12-23 2022-03-11 南京理工大学 基于aeb碰撞速度的车辆路径规划方法
CN114170805B (zh) * 2021-12-23 2024-04-23 南京理工大学 基于aeb碰撞速度的车辆路径规划方法
CN114379587A (zh) * 2021-12-28 2022-04-22 阿波罗智联(北京)科技有限公司 自动驾驶中避让行人的方法与装置
CN114379587B (zh) * 2021-12-28 2024-05-24 阿波罗智联(北京)科技有限公司 自动驾驶中避让行人的方法与装置

Also Published As

Publication number Publication date
CN111497840A (zh) 2020-08-07
CN111497840B (zh) 2021-04-16
US20230192078A1 (en) 2023-06-22

Similar Documents

Publication Publication Date Title
WO2021217752A1 (zh) 车辆-行人碰撞风险域的计算方法及安全评价系统
CN106985780B (zh) 车辆安全辅助系统
EP2388756B1 (en) Forward collision risk reduction
EP2302412B1 (en) System and method for evaluation of an automotive vehicle forward collision threat
US9566959B2 (en) Method for determining an emergency braking situation of a vehicle
US8543309B2 (en) ACC and AM braking range variable based on lateral and longitudinal position of forward vehicle and curvature of road
CN110217226B (zh) 车辆控制装置
KR20140057583A (ko) 자동차용 안전 장치
US20130261916A1 (en) Driving support apparatus for vehicle
JP6828429B2 (ja) 車両用衝突回避支援装置および車両の衝突回避支援方法
JP4961592B2 (ja) 車両の走行支援装置
JP4952938B2 (ja) 車両の走行支援装置
CN110712645B (zh) 一种预测盲区目标车辆相对位置的方法及系统
CN111768651A (zh) 一种预防车辆碰撞的预警方法及装置
JP2020523709A (ja) 後方プリクラッシュセーフティシステム
US11599117B2 (en) Systems and methods for obstacle proximity detection
JP2019191882A (ja) 隊列走行制御装置
CN113053165A (zh) 车辆及其碰撞识别方法、装置和设备
JP5129657B2 (ja) 車両用衝突回避支援装置
EP2172920B1 (en) Threat assessment for unexpected events
KR101511860B1 (ko) 운전보조시스템 및 그 제어방법
JP6331233B2 (ja) 車両制御装置
CN108238102A (zh) 一种车载超声波传感器系统
Chen et al. Safe technology with a novel rear collision avoidance system of vehicles
CN115167367A (zh) 基于主动安全和v2x技术的倒车避撞系统

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: 20933062

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: 20933062

Country of ref document: EP

Kind code of ref document: A1