WO2021180035A1 - 一种泊车路径规划方法、装置、车辆和存储介质 - Google Patents

一种泊车路径规划方法、装置、车辆和存储介质 Download PDF

Info

Publication number
WO2021180035A1
WO2021180035A1 PCT/CN2021/079540 CN2021079540W WO2021180035A1 WO 2021180035 A1 WO2021180035 A1 WO 2021180035A1 CN 2021079540 W CN2021079540 W CN 2021079540W WO 2021180035 A1 WO2021180035 A1 WO 2021180035A1
Authority
WO
WIPO (PCT)
Prior art keywords
parking
point
target
vehicle
target vehicle
Prior art date
Application number
PCT/CN2021/079540
Other languages
English (en)
French (fr)
Inventor
李超
杜建宇
刘斌
Original Assignee
中国第一汽车股份有限公司
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 中国第一汽车股份有限公司 filed Critical 中国第一汽车股份有限公司
Publication of WO2021180035A1 publication Critical patent/WO2021180035A1/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
    • 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, or advanced driver assistance systems for ensuring comfort, stability and safety or drive control systems for propelling or retarding the vehicle
    • 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, or advanced driver assistance systems for ensuring comfort, stability and safety or drive control systems for propelling or retarding the vehicle
    • B60W30/06—Automatic manoeuvring for parking
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S15/00—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems
    • G01S15/88—Sonar systems specially adapted for specific applications
    • G01S15/93—Sonar systems specially adapted for specific applications for anti-collision purposes
    • G01S15/931—Sonar systems specially adapted for specific applications for anti-collision purposes of land vehicles
    • 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
    • B60W2520/00—Input parameters relating to overall vehicle dynamics
    • B60W2520/06—Direction of travel
    • 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
    • B60W2530/00—Input parameters relating to vehicle conditions or values, not covered by groups B60W2510/00 or B60W2520/00
    • 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
    • B60W2540/00—Input parameters relating to occupants
    • B60W2540/18—Steering angle
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S15/00—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems
    • G01S15/88—Sonar systems specially adapted for specific applications
    • G01S15/93—Sonar systems specially adapted for specific applications for anti-collision purposes
    • G01S15/931—Sonar systems specially adapted for specific applications for anti-collision purposes of land vehicles
    • G01S2015/932—Sonar systems specially adapted for specific applications for anti-collision purposes of land vehicles for parking operations

Definitions

  • the embodiments of the present application relate to parking technologies, for example, to a parking route planning method, device, vehicle, and storage medium.
  • the present application provides a parking route planning method, device, vehicle and storage medium, which improve the calculation speed of route search and ensure the real-time nature of route planning.
  • an embodiment of the present application provides a parking route planning method, including:
  • collision detection and cost calculation are performed on each parking path planned by the RS geometric algorithm
  • Collision detection and cost calculation are performed on each candidate state point in the set of candidate state points until the output cost is the lowest, no collision, and the state point of the target point is reached.
  • an embodiment of the present application also provides a parking route planning device, including:
  • the first determining module is set to determine the starting point and the target point of the parking path on the pre-established parking scene coordinate system
  • a second determining module configured to determine whether there is a parking path from the starting point to the target point at the current state point of the target vehicle using an RS geometric algorithm when the target point is valid;
  • the first detection calculation module is configured to perform collision detection and cost calculation on each parking path planned by the RS geometric algorithm when there is a parking path from the starting point to the target point;
  • the third determining module is configured to use the hybrid A-star algorithm to determine the set of candidate state points of the target vehicle when all parking paths planned by the RS geometric algorithm are parking paths with collisions;
  • the second detection calculation module is configured to perform collision detection and cost calculation on each candidate state point in the set of candidate state points until the output cost is the lowest, no collision, and the state point of the target point is reached.
  • an embodiment of the present application also provides a vehicle, including: a memory, and at least one processor;
  • Memory set to store at least one program
  • the at least one processor When the at least one program is executed by the at least one processor, the at least one processor implements the parking route planning method according to the first aspect.
  • a computer-readable storage medium has a computer program stored on the computer-readable storage medium, and when the computer program is executed by a processor, the parking route planning method as described in the first aspect is implemented.
  • FIG. 1 is a flowchart of a parking route planning method provided by an embodiment of the present application
  • FIG. 2 is a schematic diagram of a target point corresponding to a valid target parking space provided by an embodiment of the present application
  • FIG. 3 is a schematic diagram of a display of obstacle map information provided by an embodiment of the present application.
  • FIG. 4 is a schematic diagram of establishing a parking scene coordinate system according to an embodiment of the present application.
  • FIG. 5 is a schematic diagram of collision detection provided by an embodiment of the present application.
  • Fig. 6 is a schematic diagram of a state update of a hybrid A-star algorithm provided by an embodiment of the present application.
  • FIG. 7 is a schematic diagram showing a result of parking route planning provided by an embodiment of the present application.
  • FIG. 8 is a flowchart of another parking route planning method provided by an embodiment of the present application.
  • FIG. 9 is a schematic display diagram of a parking route planning process provided by an embodiment of the present application.
  • FIG. 10 is a structural block diagram of a parking route planning device provided by an embodiment of the present application.
  • FIG. 11 is a schematic diagram of the hardware structure of a vehicle provided by an embodiment of the present application.
  • FIG. 1 is a flowchart of a parking route planning method provided by an embodiment of the present application. This embodiment can be applied to a situation in which a parking route is automatically planned for a vehicle.
  • the method can be executed by a parking route planning device, where The method can be implemented in hardware and/or software, and generally can be integrated in a vehicle.
  • the method includes steps S110 to S150.
  • S110 Determine the starting point and the target point of the parking path on the pre-established parking scene coordinate system.
  • the parking scene coordinate system is used to determine the starting point and the target point of the parking path.
  • the starting point of the parking route can be determined according to the current position automatically positioned by the Global Positioning System (GPS) positioning module of the target vehicle, that is, the starting point of the parking route is the current position point of the target vehicle;
  • the target point of the vehicle path can be determined according to the target location to be reached by the target vehicle, that is, the target point of the parking path is the location point where the target parking space of the target vehicle is located.
  • GPS Global Positioning System
  • the midpoint of the rear axle corresponding to the current position of the target vehicle may be used as the starting point, and the target parking space to be reached by the target vehicle The geometric center point of the location point is used as the target point.
  • the target point can also be determined in other ways.
  • the user can directly click on the parking scene coordinate system to manually confirm the coordinate point, and use the coordinate point as the coordinate of the target point.
  • the target point effectively means that the target vehicle can be parked normally.
  • the method of judging whether the target point is valid includes: determining whether the target parking space corresponding to the target point can park the target vehicle. For example, suppose that the length and width of the target vehicle are 5m and 2m, respectively, and the standard length and standard width of the target parking space corresponding to the target point are 6m and 2.5m, respectively, but because there are parking spaces on both sides of the target parking space. And, if the parking is somewhat inclined, resulting in the current length and width of the target parking space being 4.5m and 1.9m, respectively, it can be determined that the target parking space cannot be parked in the target vehicle normally, that is, the target point is invalid.
  • Fig. 2 is a schematic diagram of a target point corresponding to a valid target parking space provided by an embodiment of the present application.
  • the parking spaces can be divided into vertical parking spaces, horizontal parking spaces and inclined parking spaces.
  • the standard length and width of parking spaces in different directions are both 6m and 2.5m. It can be understood that when the length and width of the target parking space satisfy 6m and 2.5m, the target parking space can be determined to be valid, that is, the target point is valid. Of course, in actual operation, it is also possible to determine whether the target parking space is valid according to the actual length and width of the target vehicle.
  • the target vehicle can park in the target parking space, that is, it can be determined that the target point is valid.
  • the current state point of the target vehicle refers to the state of the target vehicle at the current moment.
  • the current state point may include: the current position point, the current steering wheel angle, the current vehicle speed, and so on.
  • the RS geometric algorithm is used to determine whether a path can be planned between the starting point and the target point of the target vehicle.
  • the RS geometric algorithm is a route planning method based on the geometric algorithm, which can quickly plan the path from the starting point to the ending point (that is, the target point).
  • the path planned by the RS geometric algorithm does not consider the obstacle information. Therefore, collision detection and cost calculation need to be performed on the path planned by the RS geometric algorithm. It can be understood that when the parking path from the starting point to the target point can be planned by using the RS geometric algorithm, collision detection and cost calculation need to be performed on each planned parking path. If there are at least two collision-free paths, the path with the least cost is output as the parking path.
  • S140 When all parking paths planned by the RS geometric algorithm are parking paths with collisions, use the hybrid A-star algorithm to determine a set of candidate state points of the target vehicle.
  • the hybrid A-star algorithm is used to update the parking paths.
  • the set of candidate state points of the target vehicle may be determined according to the Akaman steering equation.
  • the set of candidate state points refers to a combination of multiple candidate state points determined according to the Akaman steering equation. At least two candidate state points are included in the set of candidate state points.
  • S150 Perform collision detection and cost calculation on each candidate state point in the set of candidate state points until the state point with the lowest cost, no collision, and reaching the target point is output.
  • the cost of the candidate state point is calculated, and the cost of selecting the transition from the current state point to the candidate state point and the cost from the candidate state point to the target point are calculated (that is, the cost of each candidate state point includes the current state point jump Go to the cost of the candidate state point and the cost from the candidate state point to the target point).
  • the current state point of the target vehicle is updated and the state point with the lowest cost is selected As the next state, to update the state. It should be noted that the cost from the candidate state point to the target point is calculated based on the Euclidean distance between the candidate state point and the target point.
  • the calculation is terminated, and the result of the hybrid A-star algorithm search is directly output; if the target point is not reached, the S120 is repeated, that is, the RS geometric algorithm is used Determine whether the updated current state point of the target vehicle has a collision-free parking path.
  • the updated current state point of the target vehicle has a collision-free parking path
  • terminate the hybrid A-star algorithm and calculate the cost from the RS geometric algorithm
  • the lowest path and the results of the hybrid A-star algorithm search are spliced together and output; if all the parking paths corresponding to the updated current state point of the target vehicle are parking paths with collisions, continue to perform the hybrid A-star algorithm search to determine the target The set of new candidate state points of the vehicle and S120 until a state point that has no collision and can reach the target point is obtained.
  • the starting point and target point of the parking path are determined on the pre-established parking scene coordinate system; then based on the starting point and target point of the target vehicle, the hybrid A-star algorithm is used to search for the parking path, and The parking path performs collision detection to ensure that the planned parking path can achieve collision avoidance while meeting vehicle constraints.
  • the RS geometric algorithm is spliced, which improves the calculation speed of the path search and ensures the real-time nature of the parking path search.
  • before establishing the parking scene coordinate system in advance it further includes: determining the parking type of the target vehicle and the location of the target parking space when the parking function of the target vehicle is turned on.
  • the parking type refers to the type of the target parking space corresponding to the target point to be reached by the target vehicle.
  • the parking types may include: vertical parking spaces, horizontal parking spaces, and inclined parking spaces.
  • the location of the target parking space refers to the location of the target parking space corresponding to the target point to be reached by the target vehicle.
  • the target parking space position includes the target point, that is, the target point may be the geometric center point of the target parking space position.
  • the RS geometric algorithm is used to determine whether there is parking from the starting point to the target point at the current state point of the target vehicle Before the route, it also includes: establishing obstacle map information according to the ultrasonic radar information in the parking space search process (that is, the process of determining the location of the target parking space); and determining the parkingable area and obstacle coordinate points of the target vehicle according to the obstacle map information.
  • the obstacle map information refers to the location information containing all obstacles from the starting point to the target point.
  • the ultrasonic radar information can be used to determine the location information of all obstacles in the process of parking space search.
  • other positioning methods can also be used to determine the location information of the obstacles in the actual operation process, which is not limited. .
  • the user can see the location information of the obstacle on the obstacle map, and the processor of the vehicle can automatically detect and determine the parkable area and the parking area of the target vehicle based on the obstacle map information.
  • Obstacle coordinate point Among them, the parkable area refers to the area where the target vehicle can drive but cannot park. FIG.
  • FIG. 3 is a schematic diagram of displaying obstacle map information provided by an embodiment of the present application.
  • an obstacle map can be established based on the ultrasonic radar information, and the parking area and the location of the obstacle can be displayed on the obstacle map, so as to facilitate the subsequent determination of the target point of the target vehicle and the Impact checking.
  • the process of establishing the parking scene coordinate system includes: obtaining the current parking parameters of the target vehicle.
  • the current parking parameters include: the minimum turning radius of the vehicle, the distance from the midpoint of the rear axle to the front overhang, and the middle of the rear axle. The distance from the point to the rear overhang, the width of the vehicle and the wheelbase; according to the current parking parameters, the geometric center point of the target vehicle and the direction of the front of the vehicle, the corresponding parking scene coordinate system is established.
  • the current parking parameters of the target vehicle may include the following items: the minimum turning radius R of the vehicle, the distance Lf from the midpoint of the rear axle to the front overhang, the distance Lr from the midpoint of the rear axle to the rear overhang, and the vehicle width Lw. And the wheelbase L.
  • FIG. 4 is a schematic diagram of establishing a parking scene coordinate system provided by an embodiment of the present application. As shown in Figure 4, when the parking function is turned on, the geometric center point of the target vehicle is used as the coordinate origin, the longitudinal axis of the front of the car is the x-axis, and the left and perpendicular to the x-axis is the y-axis to establish a right-handed coordinate system. Then, the geometric center point of the target parking space is taken as the end point, and the coordinates of the target point (that is, the coordinates of the end point) are determined.
  • the coordinates of the midpoint of the rear axle of the target vehicle are used to plan the parking path. That is, when the parking function of the target vehicle is turned on, the midpoint of the rear axle of the target vehicle As the starting position (that is, the starting point), the position designated by the driver is used as the end position (that is, the target point).
  • the process of performing collision detection on each parking path planned by the RS geometric algorithm includes: determining four boundary points of the target vehicle; according to the distance between the four boundary points and the obstacle coordinate point, Determine whether there is a path point that collides with an obstacle in each parking path.
  • collision detection refers to sequentially performing collision detection on the planned waypoints, and checking whether each waypoint collides with obstacles.
  • FIG. 5 is a schematic diagram of collision detection provided by an embodiment of the present application.
  • the target vehicle is simplified into a rectangle, A, B, C, and D are the coordinates of the four vertices of the target vehicle (that is, four boundary points), and Ob is the obstacle point in the obstacle map, and its coordinates Is (Ob i , Ob j ), i, j ⁇ [1,2,...n], if the Ob point satisfies It indicates that the Ob point collides with the target vehicle, that is, the target vehicle collides with an obstacle.
  • the process of performing cost calculation for each planned parking path or alternative state point includes: obtaining the steering wheel angle of the target vehicle, the amount of steering wheel angle change, and the current distance between the obstacle and the target vehicle; Determine according to the steering wheel angle of the target vehicle, the amount of steering wheel angle change, the current distance between the obstacle and the target vehicle, the preset steering wheel weight coefficient, the preset steering wheel change weight system, and the preset weight coefficient of distance to the obstacle.
  • the cost of the parking route or the cost of the alternate state point includes: obtaining the steering wheel angle of the target vehicle, the amount of steering wheel angle change, and the current distance between the obstacle and the target vehicle; Determine according to the steering wheel angle of the target vehicle, the amount of steering wheel angle change, the current distance between the obstacle and the target vehicle, the preset steering wheel weight coefficient, the preset steering wheel change weight system, and the preset weight coefficient of distance to the obstacle.
  • the cost calculation is used to select and evaluate multiple alternative collision-free paths, that is, the lower the cost, the more reasonable the path.
  • set different weight coefficients for the planned path points such as shift coefficient weight, left and right turn coefficient weight, weight close to obstacles, calculate the cost of each path point, and select the path point with the lowest cost as the optimal path.
  • the cost calculation formula for each path point is as follows:
  • cst represents the cost of the waypoint
  • ⁇ 1 represents the weight coefficient of adjusting the steering wheel
  • ⁇ 2 represents the weight coefficient of the steering wheel change
  • ⁇ 3 represents the weight coefficient of the distance to the obstacle
  • ⁇ represents the steering wheel angle
  • ⁇ represents the steering wheel angle.
  • Dis represents the distance between the obstacle and the target vehicle.
  • using the hybrid A-star algorithm to determine the set of candidate state points of the target vehicle includes: obtaining the control value and the state value of the target vehicle during the search and update process using the hybrid A-star algorithm; the control value includes: the direction of travel , Travel distance and steering wheel angle; state variables include: the coordinates and heading angle of the midpoint of the rear axle; determine the corresponding at least two candidate state points according to the control variables and state variables; combine the at least two candidate state points into corresponding A collection of alternative status points.
  • the hybrid A-star algorithm is used to search and update the waypoints.
  • the selected control variables are the direction of travel, the distance of travel, and the steering wheel angle;
  • the state parameter selects the coordinates and heading angle of the midpoint of the target vehicle's rear axle, and updates the vehicle state according to different control input and Ackerman steering equations, and calculates the next state of the vehicle.
  • it is restricted. Due to the limitation of the steering angle of the car (that is, the steering wheel angle), the next state of the target vehicle is limited. Exemplarily, FIG.
  • FIG. 6 is a schematic diagram of a state update of a hybrid A-star algorithm provided by an embodiment of the present application.
  • the state of the target vehicle is updated using the Ackerman steering principle to ensure that the next updated state of the target vehicle meets the constraints of the target vehicle. It can be understood that the target vehicle can go straight (i.e. A'), turn left (A) or turn right (i.e. A").
  • L represents the wheelbase of the vehicle
  • D represents the distance traveled by the vehicle in each route search
  • x lst represents the abscissa of the vehicle after the previous state update
  • y lst represents the ordinate of the vehicle after the previous state update
  • ⁇ lst Indicates the heading angle of the vehicle after the previous state update
  • x and y respectively represent the horizontal and vertical coordinates of the midpoint of the rear axle of the current state point after the vehicle state update
  • ⁇ represents the heading angle of the current state point after the vehicle state update. Round corner.
  • S120 is repeated at all times to detect whether there is a collision-free path at the updated position.
  • FIG. 7 is a schematic diagram showing a result of parking route planning provided by an embodiment of the present application. As shown in Figure 7, the line is the result of the hybrid A-star algorithm search, and the dotted line is the path calculated by the RS geometric algorithm.
  • the hybrid A-star algorithm is used to search until the end point (that is, the target point).
  • the state update selects the state with no collision and the least cost as the next state of the hybrid A star algorithm.
  • FIG. 8 is a flowchart of another parking route planning method provided by an embodiment of the present application. As shown in FIG. 8, this embodiment includes steps S210 to S2130.
  • S220 Determine the parking type and search for parking spaces.
  • S260 Whether the target point is reached, if the target point is reached, S2110 is executed; if the target point is not reached, S270 is executed.
  • judging whether to reach the target point is to determine whether the current state point of the target vehicle can reach the target point, that is, to determine whether the path from the current state point to the target point can be planned through the hybrid A-star algorithm and the RS geometric algorithm .
  • S290 Determine whether there are multiple parking paths that have not collided with the obstacle, and if there are multiple parking paths that have not collided with the obstacle, execute S2100, and if there is only one parking path that has not collided with the obstacle, In the case of a parking route, S2120 is executed.
  • Path splicing optimization refers to splicing the path planned by the hybrid A star algorithm and the path planned by the RS geometric algorithm, and removes the unreasonable commutation in the path planned by the hybrid A star algorithm and the path planned by the RS geometric algorithm.
  • S2130 Use the hybrid A-star algorithm to update the status, and then execute S260.
  • the parking function of the target vehicle is turned on, the parking type and the location of the target parking space are determined; the parking scene coordinate system is established according to the current parking parameters of the target vehicle, and the starting point and target point of the parking path planning are determined according to Obstacle map information determines the parking area and obstacle information of the vehicle; judge whether the target point is valid, if the target point is invalid, terminate the parking path planning, if the target point is valid, continue; use the RS geometric algorithm to determine the current target vehicle Whether there is a parking path from the starting point to the target point at the state point; if there is a parking path from the starting point to the target point, collision detection and cost calculation are required for each parking path planned by the RS geometry algorithm.
  • the calculation is terminated and the parking path with the lowest cost is directly output. If all parking paths planned by the RS geometric algorithm are parking paths with collisions, then Continue; use the hybrid A-star algorithm to update the path state points, determine the candidate state points of the vehicle according to the Akaman steering equation, and then perform collision detection and cost calculation on the set of candidate state points in the search process. If the candidate state point has a collision, skip the state point.
  • Cost calculation calculate the cost of selecting the jump from the current state point to the candidate state point and the cost from the candidate state point to the target point. After the detection of all candidate state points is completed, the current state is updated, and the selection cost is the lowest The status point is used as the next status for status update. Determine whether the current state point has reached the target point. If the target point is reached, the calculation is terminated, and the result of the hybrid A-star algorithm search is directly output.
  • the RS geometric algorithm is reused to determine the process based on the current state, that is, to determine whether the current position can be planned by using the RS geometric algorithm.
  • the lowest cost parking path and the results of the hybrid A-star algorithm search are spliced together and output; if all the parking paths from the current position to the target point planned by the RS geometry algorithm are parking paths with collisions, continue to use the RS geometry
  • the process of algorithm judgment. This embodiment can ensure that under various working conditions, a path that realizes collision avoidance while meeting vehicle constraints can be planned.
  • splicing RS geometric algorithms improves the calculation speed of path search and ensures the real-time nature of path planning.
  • FIG. 9 is a schematic diagram of a parking route planning process provided by an embodiment of the present application.
  • the starting point of the parking path is A
  • the target point is B.
  • the RS geometric algorithm is used to determine that there is a parking path from the starting point to the target point at the current state point of the target vehicle, that is, the dashed line shown in Figure 9, and collision detection is performed on the planned path.
  • the hybrid A-star algorithm is used to determine the set of candidate state points of the target vehicle, that is, the candidate state points C, C′, C′′, C′′ as shown in Fig.
  • Figure 9 shows the candidate state points D, D′, D′′, D′′′, and then judge whether the RS geometric algorithm can be used to reach each candidate state point (D, D′, D′′, D′′′) Plan a path between target points B.
  • a path cannot be planned from any candidate state point to target point B, select the optimal state point from the four candidate state points (for example, D′) , And so on, until it is updated to the point E′′, the RS geometric algorithm can be obtained to plan the path to the target point B, then the A, C, D′, E′′ are combined with the path output by the RS algorithm to obtain the final parking path, Thereby, it can be ensured that a path that can avoid collisions while meeting vehicle constraints can be planned under various working conditions, and the path points obtained by the RS geometric algorithm are spliced, which improves the calculation speed of path search and ensures the real-time nature of path planning.
  • Figure 10 is a structural block diagram of a parking route planning device provided by an embodiment of the present application.
  • the device is suitable for automatically planning a parking route for a vehicle.
  • the device can be implemented by hardware/software, and can generally be integrated in a vehicle. middle.
  • the device includes: a first determination module 310, a second determination module 320, a first detection calculation module 330, a third determination module 340, and a second detection calculation module 350.
  • the first determining module 310 is configured to determine the starting point and the target point of the parking path on the pre-established parking scene coordinate system
  • the second determining module 320 is configured to use the RS geometric algorithm to determine whether there is a parking path from the starting point to the target point at the current state point of the target vehicle when the target point is valid;
  • the first detection calculation module 330 is configured to perform collision detection and cost calculation on each parking path planned by the RS geometric algorithm when there is a parking path from the starting point to the target point;
  • the third determining module 340 is configured to use the hybrid A-star algorithm to determine the set of candidate state points of the target vehicle when all parking paths planned by the RS geometric algorithm are parking paths with collisions;
  • the second detection calculation module 350 is configured to perform collision detection and cost calculation on each candidate state point in the set of candidate state points until the state point with the lowest cost, no collision, and reaching the target point is output.
  • the starting point and target point of the parking path are determined on the pre-established parking scene coordinate system; then based on the starting point and target point of the target vehicle, the hybrid A-star algorithm is used to search for the parking path, and Carry out collision detection on the parking path to ensure that the planned parking path can achieve collision avoidance while meeting vehicle constraints.
  • the RS geometric algorithm is spliced, which improves the calculation speed of the path search and ensures the real-time nature of the parking path search.
  • the parking route planning device further includes:
  • the fourth determining module is configured to determine the parking type of the target vehicle and the location of the target parking space when the parking function of the target vehicle is turned on before the pre-established parking scene coordinate system.
  • the parking path planning device further includes:
  • the establishment module is set to determine whether there is a parking path from the starting point to the target point at the current state point of the target vehicle by using the RS geometric algorithm after the starting point and the target point of the parking path are determined, when the target point is valid Previously, the obstacle map information was established based on the ultrasonic radar information during the parking search process;
  • the fifth determining module is configured to determine the parkingable area of the target vehicle and the obstacle coordinate point according to the obstacle map information.
  • the process of establishing the parking scene coordinate system includes: obtaining the current parking parameters of the target vehicle.
  • the current parking parameters include the following: the minimum turning radius of the vehicle, the distance from the midpoint of the rear axle to the front overhang, and the rear The distance from the midpoint of the axle to the rear overhang, the width of the vehicle and the wheelbase; according to the current parking parameters, the geometric center point of the target vehicle and the direction of the front of the vehicle, the corresponding parking scene coordinate system is established.
  • the process of performing collision detection on each parking path planned by the RS geometric algorithm includes: determining four boundary points of the target vehicle; according to the distance between the four boundary points and the obstacle coordinate point, Determine whether there is a path point that collides with an obstacle in each parking path.
  • the process of performing cost calculation for each planned parking path or alternative state point includes: obtaining the steering wheel angle of the target vehicle, the amount of steering wheel angle change, and the current distance between the obstacle and the target vehicle; Determine according to the steering wheel angle of the target vehicle, the amount of steering wheel angle change, the current distance between the obstacle and the target vehicle, the preset steering wheel weight coefficient, the preset steering wheel change weight system, and the preset weight coefficient of distance to the obstacle.
  • the cost of the parking route or the cost of the alternate state point includes: obtaining the steering wheel angle of the target vehicle, the amount of steering wheel angle change, and the current distance between the obstacle and the target vehicle; Determine according to the steering wheel angle of the target vehicle, the amount of steering wheel angle change, the current distance between the obstacle and the target vehicle, the preset steering wheel weight coefficient, the preset steering wheel change weight system, and the preset weight coefficient of distance to the obstacle.
  • using the hybrid A-star algorithm to determine the set of candidate state points of the target vehicle includes: obtaining the control value and the state value of the target vehicle during the search and update process using the hybrid A-star algorithm; the control value includes: the direction of travel , Travel distance and steering wheel angle; state variables include: the coordinates and heading angle of the midpoint of the rear axle; determine the corresponding at least two candidate state points according to the control variables and state variables; combine the at least two candidate state points into corresponding A collection of alternative status points.
  • the above parking path planning device can execute the parking path planning method provided by any embodiment of the present application, and has functional modules corresponding to the execution method.
  • FIG. 11 is a schematic diagram of the hardware structure of a vehicle provided by an embodiment of the present application.
  • the vehicle provided by the embodiment of the present application includes a memory 410 and at least one processor 420.
  • the processor 420 in the vehicle may be at least one.
  • one processor 420 is taken as an example.
  • the processor 420 and the memory 410 in the vehicle may be connected by a bus or other methods. In FIG. 11, a bus connection is taken as an example.
  • the memory 410 in the vehicle can be configured to store at least one program.
  • the program can be a software program, a computer-executable program, and a module, as corresponding to the parking route planning method provided in the embodiment of the present application.
  • Program instructions/modules (for example, the modules in the parking route planning device shown in FIG. 10 include: a first determination module 310, a second determination module 320, a first detection calculation module 330, a third determination module 340, and a second determination module Detection calculation module 350).
  • the processor 410 executes various functional applications and data processing of the device by running the software programs, instructions, and modules stored in the memory 410, that is, realizes the parking route planning method in the foregoing method embodiment.
  • the memory 410 may include a storage program area and a storage data area.
  • the storage program area may store an operating system and an application program required by at least one function; the storage data area may store data created according to the use of a device configured in the device, and the like.
  • the memory 410 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage devices.
  • the memory 410 may include a memory remotely provided with respect to the processor 420, and these remote memories may be connected to a device configured in the device through a network. Examples of the aforementioned networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.
  • a vehicle including a memory 410 and a processor 420, the memory 410 stores a computer program, and the processor 420 implements the following steps when the computer program is executed:
  • the starting point and target point of the parking path determine the starting point and target point of the parking path; when the target point is valid, use the RS geometric algorithm to determine whether there is a parking from the starting point to the target point in the current state of the target vehicle Vehicle path; in the case of a parking path from the starting point to the target point, collision detection and cost calculation are performed on each parking path planned by the RS geometric algorithm; all parking paths planned by the RS geometric algorithm are In the case of a collision parking path, the hybrid A-star algorithm is used to determine the candidate state point set of the target vehicle; collision detection and cost calculation are performed on each candidate state point in the candidate state point set until the output cost is the lowest , No collision, and until reaching the state point of the target point.
  • the above-mentioned vehicle can execute the parking route planning method provided by any embodiment of the present application, and has functional modules corresponding to the execution method.
  • the embodiment of the present application also provides a computer-readable storage medium on which a computer program is stored.
  • the parking route planning method provided by the embodiment of the present application is implemented.
  • the method includes: On the established parking scene coordinate system, determine the starting point and target point of the parking path; when the target point is valid, use the RS geometric algorithm to determine whether there is a parking path from the starting point to the target point in the current state of the target vehicle ; In the case that there is a parking path from the starting point to the target point, collision detection and cost calculation are performed on each parking path planned by the RS geometric algorithm; all parking paths planned by the RS geometric algorithm are collisions
  • the hybrid A-star algorithm is used to determine the set of candidate state points of the target vehicle; collision detection and cost calculation are performed on each candidate state point in the set of candidate state points, until the output cost is the lowest and no Until the collision and reaching the state point of the target point.
  • the computer storage medium of the embodiment of the present application may adopt any combination of at least one computer-readable medium.
  • the computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium.
  • the computer-readable storage medium may be, for example, but is not limited to an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the above.
  • computer-readable storage media include: electrical connections with at least one wire, portable computer disks, hard disks, random access memory (RAM), read-only memory (Read -Only Memory, ROM), Erasable Programmable Read-Only Memory (EPROM or flash memory), optical fiber, and Portable Compact Disc Read-Only Memory (CD-ROM) , Optical storage devices, magnetic storage devices, or any suitable combination of the above.
  • the computer-readable storage medium can be any tangible medium that contains or stores a program, and the program can be used by or in combination with an instruction execution system, apparatus, or device.
  • the computer-readable signal medium may include a data signal propagated in baseband or as a part of a carrier wave, and computer-readable program code is carried therein. This propagated data signal can take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing.
  • the computer-readable signal medium may also be any computer-readable medium other than the computer-readable storage medium.
  • the computer-readable medium may send, propagate, or transmit the program for use by or in combination with the instruction execution system, apparatus, or device .
  • the program code contained on the computer-readable medium can be transmitted by any suitable medium, including, but not limited to, wireless, wire, optical cable, radio frequency (RF), etc., or any suitable combination of the foregoing.
  • suitable medium including, but not limited to, wireless, wire, optical cable, radio frequency (RF), etc., or any suitable combination of the foregoing.
  • RF radio frequency
  • the computer program code used to perform the operations of this application can be written in one or more programming languages or a combination thereof.
  • the programming languages include object-oriented programming languages, such as Java, Smalltalk, and C++, as well as conventional procedural programming languages.
  • Programming language such as "C" language or similar programming language.
  • the program code can be executed entirely on the user's computer, partly on the user's computer, executed as an independent software package, partly on the user's computer and partly executed on a remote computer, or entirely executed on the remote computer or server.
  • the remote computer can be connected to the user's computer through any kind of network, including Local Area Network (LAN) or Wide Area Network (WAN), or it can be connected to an external computer (for example, using Internet service provider to connect via the Internet).
  • LAN Local Area Network
  • WAN Wide Area Network
  • This application determines the starting point and the target point of the parking path on the pre-established parking scene coordinate system; when the target point is valid, the RS geometric algorithm is used to determine whether the current state point of the target vehicle exists from the starting point to the target point In the case of a parking path from the starting point to the target point, collision detection and cost calculation are performed on each parking path planned by the RS geometric algorithm; all parking paths planned by the RS geometric algorithm In the case of parking paths with collisions, the hybrid A-star algorithm is used to determine the set of candidate state points of the target vehicle; collision detection and cost calculation are performed on each candidate state point in the set of candidate state points, until output The lowest cost, no collision, and until reaching the state point of the target point. In the embodiment of the present application, the hybrid A-star algorithm is used to splice the RS geometric algorithm in the parking route search process, which improves the calculation speed of the route search and ensures the real-time nature of the parking route search.

Abstract

一种泊车路径规划方法，包括：在预先建立的泊车场景坐标系上，确定泊车路径的起始点和目标点（S110）；在目标点有效的情况下，利用RS几何算法确定目标车辆的当前状态点是否存在起始点到目标点的泊车路径（S120）；在存在起始点到目标点的泊车路径时，对RS几何算法规划出的每个泊车路径进行碰撞检测及成本计算（S130）；在RS几何算法规划出的所有泊车路径均为存在碰撞的泊车路径时，利用混合A星算法确定目标车辆的备选状态点集合（S140）；对备选状态点集合中每个备选状态点进行碰撞检测及成本计算，直至输出成本最低、不碰撞、以及到达目标点的状态点为止（S150）。还公开了一种泊车路径规划装置、一种车辆和一种计算机可读存储介质。

Description

一种泊车路径规划方法、装置、车辆和存储介质
本申请要求在2020年3月11日提交中国专利局、申请号为202010166385.4的中国专利申请的优先权，该申请的全部内容通过引用结合在本申请中。
技术领域
本申请实施例涉及泊车技术，例如涉及一种泊车路径规划方法、装置、车辆和存储介质。
背景技术
随着车辆数量的不断增加，自动泊车可以有效避免泊车难的情况，而路径规划是自动泊车的重要步骤。
目前路径规划方法有很多种，一部分方法是基于几何的方式，以RS(Reeds-Shepp)算法为例，这种方式大多是针对特定泊车场景设计的泊车路径，存在场景限制，无法适用于所有泊车类型，部分场景可能无法规划泊车路径，导致泊车失败；另外一类方法是基于搜索的方式，以混合A星算法为例，这种方法针对各种泊车场景均能搜索需要的泊车路径，但是受限于计算时间与资源的限制导致实时性较差，无法移植到产品控制器。
发明内容
本申请提供一种泊车路径规划方法、装置、车辆和存储介质，提高了路径搜索的计算速度，保证路径规划的实时性。
第一方面，本申请实施例提供了一种泊车路径规划方法，包括：
在预先建立的泊车场景坐标系上，确定泊车路径的起始点和目标点；
响应于确定所述目标点有效，利用RS几何算法确定所述目标车辆的当前状态点是否存在所述起始点到所述目标点的泊车路径；
响应于存在所述起始点到所述目标点的泊车路径的确定结果，对RS几何算 法规划出的每个泊车路径进行碰撞检测以及成本计算；
响应于确定RS几何算法规划出的所有泊车路径均为存在碰撞的泊车路径的情况下，利用混合A星算法确定所述目标车辆的备选状态点集合；
对所述备选状态点集合中的每个备选状态点进行碰撞检测以及成本计算，直至输出成本最低、不碰撞、以及到达所述目标点的状态点为止。
第二方面，本申请实施例还提供了一种泊车路径规划装置，包括：
第一确定模块，设置为在预先建立的泊车场景坐标系上，确定泊车路径的起始点和目标点；
第二确定模块，设置为在所述目标点有效的情况下，利用RS几何算法确定所述目标车辆的当前状态点是否存在所述起始点到所述目标点的泊车路径；
第一检测计算模块，设置为在存在所述起始点到所述目标点的泊车路径的情况下，对RS几何算法规划出的每个泊车路径进行碰撞检测以及成本计算；
第三确定模块，设置为在RS几何算法规划出的所有泊车路径均为存在碰撞的泊车路径的情况下，利用混合A星算法确定所述目标车辆的备选状态点集合；
第二检测计算模块，设置为对所述备选状态点集合中的每个备选状态点进行碰撞检测以及成本计算，直至输出成本最低、不碰撞、以及到达所述目标点的状态点为止。
第三方面，本申请实施例还提供了一种车辆，包括：存储器，以及至少一个处理器；
存储器，设置为存储至少一个程序；
当所述至少一个程序被所述至少一个处理器执行，使得所述至少一个处理器实现如第一方面所述的泊车路径规划方法。
第四方面，一种计算机可读存储介质，计算机可读存储介质上存储有计算机程序，所述计算机程序被处理器执行时，实现如第一方面所述的泊车路径规划方法。
附图说明
图1是本申请一实施例提供的一种泊车路径规划方法的流程图；
图2是本申请一实施例提供的一种目标点对应目标车位有效的示意图；
图3是本申请一实施例提供的一种障碍物地图信息的显示示意图；
图4是本申请一实施例提供的一种泊车场景坐标系的建立示意图；
图5是本申请一实施例提供的一种碰撞检测示意图；
图6是本申请一实施例提供的一种混合A星算法状态更新的示意图；
图7是本申请一实施例提供的一种泊车路径规划结果的显示示意图；
图8是本申请一实施例提供的另一种泊车路径规划方法的流程图；
图9是本申请一实施例提供的一种泊车路径规划过程的显示示意图；
图10是本申请一实施例提供的一种泊车路径规划装置的结构框图；
图11是本申请一实施例提供的一种车辆的硬件结构示意图。
具体实施方式
图1是本申请一实施例提供的一种泊车路径规划方法的流程图，本实施例可适用于对车辆自动规划泊车路径的情况，该方法可以由泊车路径规划装置来执行，其中，该方法可由硬件和/或软件的方式实现，并一般可集成在车辆中。
如图1所示，该方法包括步骤S110至S150。
S110、在预先建立的泊车场景坐标系上，确定泊车路径的起始点和目标点。
在实施例中，泊车场景坐标系用于确定泊车路径的起始点和目标点。其中，泊车路径的起始点，可以根据目标车辆中全球定位系统(Global Positioning System，GPS)定位模块自动定位的当前位置进行确定，即泊车路径的起始点为目标车辆的当前位置点；泊车路径的目标点，可根据目标车辆所要到达的目标地位置进行确定，即泊车路径的目标点为目标车辆的目标车位所在的位置点。当然，为了更准确地确定泊车路径的起始点和目标点，在实施例中，可将目标车辆的当前位置点所对应的后轴中点作为起始点，将目标车辆所要到达的目标 车位的位置点的几何中心点作为目标点。
在实际操作过程中，目标点，也可以采用其它方式进行确定。比如，用户可直接在泊车场景坐标系上，手动点击确定坐标点，并将该坐标点作为目标点的坐标。
S120、在目标点有效的情况下，利用RS几何算法确定目标车辆的当前状态点是否存在起始点到目标点的泊车路径。
在实施例中，目标点有效指的是目标车辆可以正常泊入。判断目标点是否有效的方式，包括：确定目标点所对应的目标车位是否能够泊入目标车辆。示例性地，假设目标车辆的长度和宽度分别为5m和2m，而目标点所对应目标车位的标准长度和标准宽度分别为6m和2.5m，但由于目标车位两侧的车位均有车辆泊入，并且，停靠的有些倾斜，导致目标车位的当前长度和宽度分别为4.5m和1.9m，则可确定目标车位无法正常泊入目标车辆，即目标点是无效的。
图2是本申请一实施例提供的一种目标点对应目标车位有效的示意图。在实施例中，可以将车位分别垂直车位、水平车位和倾斜车位。如图2所示，不同方向的车位所对应的标准长度和宽度均为6m和2.5m。可以理解为，在目标车位的长度和宽度满足6m和2.5m的情况下，可认定该目标车位是有效的，即目标点是有效的。当然，在实际操作过程中，也可以根据目标车辆的实际长度和宽度来确定目标车位是否有效。比如，目标车辆为迷你型的QQ车，即使在目标车位的长度和宽度不满足6m和2.5m的情况下，目标车辆也可以泊入该目标车位，即也可以确定该目标点是有效的。
在实施例中，目标车辆的当前状态点，指的是目标车辆在当前时刻所处的状态。比如，当前状态点可以包括：当前所处的位置点、当前的转向盘转角、当前车速等等。在实施例中，利用RS几何算法判断目标车辆的起始点和目标点之间是否可以规划路径。其中，RS几何算法是一种基于几何算法的路线规划方法，能够快速的规划处起始点到终止点(即目标点)的路径。
S130、在存在起始点到目标点的泊车路径的情况下，对RS几何算法规划出 的每个泊车路径进行碰撞检测以及成本计算。
在实际操作过程中，由于RS几何算法规划的路径未考虑障碍物信息。因此，需对RS几何算法规划出的路径进行碰撞检测以及成本计算。可以理解为，在利用RS几何算法可以规划出起始点到目标点的泊车路径的情况下，需对规划出的每个泊车路径进行碰撞检测以及成本计算。若存在至少两条无碰撞路径，则输出成本最小的路径作为泊车路径。
S140、在RS几何算法规划出的所有泊车路径均为存在碰撞的泊车路径的情况下，利用混合A星算法确定目标车辆的备选状态点集合。
在实施例中，在利用RS几何算法规划出的所有路径中均为存在碰撞的泊车路径的情况下，则利用混合A星算法进行泊车路径更新。示例性的，可根据阿卡曼转向方程确定目标车辆的备选状态点集合。其中，备选状态点集合指的是根据阿卡曼转向方程确定的多个备选状态点的组合。在备选状态点集合中至少包括两个备选状态点。
S150、对备选状态点集合中的每个备选状态点进行碰撞检测以及成本计算，直至输出成本最低、不碰撞、以及到达目标点的状态点为止。
在实施例中，在得到多个备选状态点之后，需依次对每个备选状态点进行碰撞检测(即对由当前状态点跳转到备选状态点的路径进行碰撞检测)以及成本计算，若备选状态点有碰撞，则跳过该备选状态点；若所有备选状态点均碰撞，则终止计算，直接输出无解；若备选状态点没有碰撞，则依次对没有碰撞的备选状态点进行成本计算，计算选择由当前状态点跳转到该备选状态点的成本以及由备选状态点到目标点的成本(即每个备选状态点的成本包括当前状态点跳转到该备选状态点的成本以及由备选状态点到目标点的成本两部分)，在完成所有备选状态点的检测之后，则更新目标车辆的当前状态点，选择成本最低的状态点作为下一步的状态，以进行状态更新。需要说明的是，由备选状态点到目标点的成本是以备选状态点和目标点之间的欧式距离来计算的。
然后，判断目标车辆更新后的当前状态点是否到达目标点，若到达目标点 则终止计算，直接输出混合A星算法搜索的结果；若没有到达目标点，则重复执行S120，即利用RS几何算法确定目标车辆更新后的当前状态点是否存在无碰撞的泊车路径，若目标车辆更新后的当前状态点存在无碰撞的泊车路径，则终止混合A星算法，并将RS几何算法计算出成本最低的路径与混合A星算法搜索的结果拼接在一起输出；若目标车辆更新后的当前状态点对应的所有泊车路径均为存在碰撞的泊车路径，则继续执行混合A星算法搜索确定目标车辆的新的备选状态点集合以及S120，直至得到无碰撞且可以到达目标点的状态点为止。
本实施的技术方案，在预先建立的泊车场景坐标系上，确定泊车路径的起始点和目标点；然后基于目标车辆的起始点和目标点利用混合A星算法搜索泊车路径，并对泊车路径进行碰撞检测，保证规划得到的泊车路径在满足车辆约束的情况下能够实现避撞。同时，在利用混合A星算法进行路径搜索过程中，拼接了RS几何算法，提高了路径搜索的计算速度，保证了泊车路径搜索的实时性。
在一个实施例中，在预先建立泊车场景坐标系之前，还包括：在目标车辆的泊车功能开启的情况下，确定目标车辆的泊车类型和目标车位位置。
在实施例中，泊车类型指的是目标车辆所要达到的目标点对应的目标车位的类型。示例性地，泊车类型可以包括：垂直车位、水平车位和倾斜车位。目标车位位置，指的是目标车辆所要到达的目标点所对应的目标车位的位置。在实施例中，目标车位位置包含目标点，即目标点可以为目标车位位置的几何中心点。
当然，为了能够实现目标车辆的自动泊车，需在目标车辆上安装并开启泊车功能。
在一实施例中，在确定泊车路径的起始点与目标点之后，在所述目标点有效的情况下，利用RS几何算法确定目标车辆的当前状态点是否存在起始点到目标点的泊车路径之前，还包括：根据车位搜索过程(即确定目标车位位置的过 程)中超声波雷达信息建立障碍物地图信息；根据障碍物地图信息确定目标车辆的可泊区域与障碍物坐标点。
在实施例中，障碍物地图信息，指的是包含从起始点到目标点之间所有障碍物的位置信息。在实施例中，可在车位搜索过程中利用超声波雷达信息确定所有障碍物的位置信息，当然，在实际操作过程中，也可以采用其它定位方式确定障碍物的位置信息，对此并不进行限定。在实施例中，在建立障碍物地图信息之后，用户可在障碍物地图上看到障碍物的位置信息，并且车辆的处理器可根据障碍物地图信息自动检测并确定目标车辆的可泊区域和障碍物坐标点。其中，可泊区域指的是目标车辆可以行驶，但不能停靠的区域。图3是本申请一实施例提供的一种障碍物地图信息的显示示意图。如图3所示，在车位搜索过程中，可以根据超声波雷达信息建立障碍物地图，并在障碍物地图上显示可泊区域和障碍物所在位置，从而便于后续确定目标车辆的目标点，以及进行碰撞检测。
在一实施例中，泊车场景坐标系的建立过程，包括：获取目标车辆的当前泊车参数，当前泊车参数包括：车辆最小转弯半径、后轴中点到前悬的距离、后轴中点到后悬的距离、车辆宽度和轴距；根据当前泊车参数、目标车辆的几何中心点和车头所指方向，建立对应的泊车场景坐标系。
在实施例中，目标车辆的当前泊车参数可以包括下述几项：车辆最小转弯半径R、后轴中点到前悬的距离Lf、后轴中点到后悬的距离Lr、车辆宽度Lw和轴距L。示例性地，图4是本申请一实施例提供的一种泊车场景坐标系的建立示意图。如图4所示，以泊车功能开启时，目标车辆的几何中心点作为坐标原点，车头所指的纵轴方向为x轴，向左垂直于x轴的为y轴，建立右手坐标系。然后，将目标车位的几何中心点作为终点，确定目标点的坐标(即终点的坐标)。
在实施例中，在对泊车路径进行规划的过程中，以目标车辆后轴中点的坐标进行泊车路径规划，即在目标车辆的泊车功能开启时刻，可以将目标车辆后轴中点作为起始位置(即起始点)，驾驶员指定的位置作为终止位置(即目标点)。
在一实施例中，对RS几何算法规划出的每个泊车路径进行碰撞检测的过程，包括：确定目标车辆的四个边界点；根据四个边界点和障碍物坐标点之间的距离，确定每个泊车路径中是否存在与障碍物发生碰撞的路径点。
在实施例中，碰撞检测指的是对规划得到的路径点依次进行碰撞检测，检查每个路径点是否与障碍物有碰撞。在实际操作过程中，可以采用多种方式进行碰撞检测。示例性地，图5是本申请一实施例提供的一种碰撞检测示意图。如图5所示，将目标车辆简化为一个长方形，A，B，C，D分别为目标车辆的四个顶点坐标(即四个边界点)，Ob为障碍物地图中障碍物点，其坐标为(Ob i，Ob j)，i，j∈[1,2，......n]，若Ob点满足
Figure PCTCN2021079540-appb-000001
则表明Ob点与目标车辆有碰撞，即目标车辆与障碍物有碰撞。
在一实施例中，对规划的每个泊车路径或者备选状态点进行成本计算的过程，包括：获取目标车辆的方向盘转角、方向盘转角变化量以及障碍物与目标车辆之间的当前距离；根据目标车辆的方向盘转角、方向盘转角变化量、障碍物与目标车辆之间的当前距离以及预先设置的方向盘权重系数、预先设置的方向盘变化权重系统和预先设置的距离障碍物远近的权重系数，确定泊车路径的成本或备选状态点的成本。
在实施例中，成本计算用于对备选的多条无碰撞路径进行选择评价，即成本越低，意味着路径越合理。示例性地，对规划得到的路径点设置不同的权重系数，譬如换挡系数权重，左右转系数权重，靠近障碍物的权重，计算每条路径点的成本，选择成本最低的路径点作为最优路径。示例性地，每个路径点的成本计算公式如下：
cst t＝cst t-1+κ 1*δ+κ 2*Δδ+κ 3/Dis
式中，cst表示路径点的成本，κ 1表示调整方向盘的权重系数，κ 2表示方向盘变化的权重系数，κ 3表示距离障碍物远近的权重系数，δ表示方向盘的转角，Δδ表示方向盘转角的变化量，Dis表示障碍物与目标车辆的距离。利用上 述公式计算路径点的成本，从备选路径中选取成本最优的路径。
在一实施例中，利用混合A星算法确定目标车辆的备选状态点集合，包括：在利用混合A星算法搜索更新过程中，获取目标车辆的控制量和状态量；控制量包括：行进方向、行进距离和方向盘转角；状态量包括：后轴中点的坐标和航向角；根据控制量以及状态量确定对应的至少两个备选状态点；将至少两个备选状态点组合成对应的备选状态点集合。
在实施例中，如果当前位置没有不碰撞的路径，则利用混合A星算法进行路径点搜索更新，利用混合A星算法搜索更新过程中，选择的控制量为行进方向、行进距离与方向盘转角；状态量选择目标车辆后轴中点的坐标和航向角，根据不同的控制量输入结合阿克曼转向方程进行车辆状态更新，计算车辆下一步的状态，在利用混合A星算法更新过程中，受制于汽车转向角(即方向盘转角)的限制，目标车辆下一步的状态有限。示例性地，图6是本申请一实施例提供的一种混合A星算法状态更新的示意图。如图6所示，利用阿克曼转向原理对目标车辆的状态进行更新，保证目标车辆下一步更新的状态满足目标车辆的约束。可以理解为，目标车辆可以直行(即A’)、左转(A)或右转(即A”)。
Figure PCTCN2021079540-appb-000002
Figure PCTCN2021079540-appb-000003
Figure PCTCN2021079540-appb-000004
在式中，L表示车辆轴距，D表示每次路径搜索中车辆行进的距离，x lst表示上一步状态更新后车辆的横坐标，y lst表示上一步状态更新后车辆的纵坐标，θ lst表示上一步状态更新后车辆的航向角，x,y分别表示车辆状态更新后当前状态点的后轴中点的横纵坐标θ表示车辆状态更新后当前状态点的航向角，δ表示车辆的前轮转角。在混合A星算法的搜索过程中时刻重复S120，检测更 新后的位置是否存在无碰撞的路径，在更新后的位置存在无碰撞的路径的情况下，终止混合A星算法，并将混合A星算法的搜索的结果和RS几何算法计算出成本最低的路径拼接在一起输出。图7是本申请一实施例提供的一种泊车路径规划结果的显示示意图。如图7所示，线条为混合A星算法搜索的结果，点划线为RS几何算法计算的路径。在更新后的位置对应的所有泊车路径均为存在碰撞的路径的情况下，利用混合A星算法一直搜索到终止点(即为目标点)为止。在混合A星算法搜索过程中，每更新一次状态就需要对状态进行碰撞检测并计算可选状态的成本，状态更新选择无碰撞且成本最小的状态作为混合A星算法下一步的状态。
图8是本申请一实施例提供的另一种泊车路径规划方法的流程图。如图8所示，本实施例包括步骤S210至S2130。
S210、开启泊车功能。
S220、确定泊车类型，并搜索车位。
S230、确定目标车位位置。
S240、建立泊车场景坐标系，确定目标点坐标以及障碍物地图信息。
S250、目标点是否有效，在目标点有效的情况下，执行S260；在目标点无效的情况下，执行S2140。
S260、是否到达目标点，在到达目标点的情况下，执行S2110；在未到达目标点的情况下，执行S270。
需要说明的是，判断是否到达目标点，即为判断目标车辆的当前状态点是否可以到达目标点，也即判断通过混合A星算法和RS几何算法能否规划出当前状态点到目标点的路径。
S270、RS几何算法是否有解，在RS几何算法有解的情况下，执行S280；在RS几何算法无解的情况下，执行S2130。
S280、检测RS几何算法规划出的每个泊车路径是否与障碍物发生碰撞，在检测到所有规划出的泊车路径均与障碍物发生碰撞的情况下，执行S2130；在检 测到存在规划出的泊车路径未与障碍物发生碰撞的情况下，执行S290。
S290、判断是否存在多条未与障碍物发生碰撞的泊车路径，在存在多条未与障碍物发生碰撞的泊车路径的情况下，执行S2100，在只存在一条未与障碍物发生碰撞的泊车路径的情况下，执行S2120。
S2100、输出成本最低的泊车路径。
S2110、路径拼接优化。
路径拼接优化是指对混合A星算法规划的路径和RS几何算法规划的路径进行拼接，并去掉混合A星算法规划的路径和RS几何算法规划的路径中不合理的换向。
S2120、输出泊车路径。
S2130、利用混合A星算法更新状态，然后执行S260。
S2140、终止泊车路径规划。
在实施例中，开启目标车辆的泊车功能，确定泊车类型以及目标车位位置；根据目标车辆的当前泊车参数建立泊车场景坐标系，确定泊车路径规划的起始点与目标点，根据障碍物地图信息确定车辆的可泊区域与障碍物信息；判断目标点是否有效，如果目标点无效，则终止泊车路径规划，如果目标点有效，则继续；利用RS几何算法判断目标车辆的当前状态点是否存在起始点到目标点的泊车路径；如果存在起始点到目标点的泊车路径，则需要对RS几何算法规划出的每个泊车路径进行碰撞检测以及成本计算，如果RS几何算法规划出的泊车路径中存在不碰撞的泊车路径，则终止计算，直接输出成本最低的泊车路径，如果RS几何算法规划出的所有泊车路径均为存在碰撞的泊车路径，则继续；利用混合A星算法进行路径状态点更新，根据阿卡曼转向方程确定车辆的备选状态点，然后在搜索过程中对备选的状态点集合依次进行碰撞检测以及成本计算。如果备选状态点有碰撞则跳过该状态点，如果所有备选状态点均碰撞，则终止计算，直接输出无解；如果备选状态点没有碰撞则依次对没有碰撞的备选状态点进行成本计算，计算选择由当前状态点跳转到该备选状态点的成本以及由备 选状态点到目标点的成本，在完成所有备选状态点的检测后，则更新当前状态，选择成本最低的状态点作为下一步的状态，进行状态更新。判断当前状态点是否到达目标点。如果到达目标点则终止计算，直接输出混合A星算法搜索的结果，如果没有到达目标点则基于当前状态再重复利用RS几何算法进行判断的过程，即判断一次当前位置利用RS几何算法是否可以规划当前位置到目标点的无碰撞泊车路径，如果RS几何算法规划出的当前位置到目标点的泊车路径中存在无碰撞泊车路径，则终止混合A星算法，并将RS几何算法计算的成本最低的泊车路径与混合A星算法搜索的结果拼接在一起输出；如果RS几何算法规划出的当前位置到目标点的所有泊车路径均为存在碰撞的泊车路径，则继续利用RS几何算法进行判断的过程。本实施例能够保证各种工况下，均能规划在满足车辆约束的情况下实现避撞的路径，此外拼接RS几何算法，提高了路径搜索的计算速度，保证了路径规划的实时性。
图9是本申请一实施例提供的一种泊车路径规划过程的显示示意图。如图9所示，假设在预先建立的泊车场景坐标系上，确定泊车路径的起始点为A，目标点为B。在目标点B有效的情况下，利用RS几何算法确定目标车辆的当前状态点存在一条起始点到目标点的泊车路径，即图9所示的虚线，对规划的该条路径进行碰撞检测，若该条路径存在与障碍物发生碰撞的情况，则利用混合A星算法确定目标车辆的备选状态点集合，即如图9所示的备选状态点C、C′、C″、C″′，然后判断利用RS几何算法是否能在每个备选状态点到目标点B之间规划出路径，若在任意一个备选状态点到目标点B之间不能规划出路径，则从这四个备选状态点中选出最优状态点(即备选状态点中的成本最低的点)(比如，C)；然后基于C点，再次采用混合A星算法进行状态更新，比如，得到如图9所示的备选状态点D、D′、D″、D″′，然后判断利用RS几何算法是否能在每个备选状态点(D、D′、D″、D″′)到目标点B之间规划 出路径，若不能在任意一个备选状态点到目标点B之间规划出路径，则从这四个备选状态点中选出最优状态点(比如，D′)，依次类推，直至更新到E″点，得到RS几何算法可以规划出到达目标点B的路径，则把A、C、D′、E″拼接RS算法输出的路径，得到最终的泊车路径，从而能够保证各种工况下均能规划在满足车辆约束的情况下实现避撞的路径，并拼接RS几何算法得到的路径点，提高了路径搜索的计算速度，保证了路径规划的实时性。
图10是本申请一实施例提供的一种泊车路径规划装置的结构框图，该装置适用于对车辆自动规划泊车路径的情况，该装置可以由硬件/软件实现，并一般可集成在车辆中。如图10所示，该装置包括：第一确定模块310、第二确定模块320、第一检测计算模块330、第三确定模块340和第二检测计算模块350。
其中，第一确定模块310，设置为在预先建立的泊车场景坐标系上，确定泊车路径的起始点和目标点；
第二确定模块320，设置为在目标点有效的情况下，利用RS几何算法确定目标车辆的当前状态点是否存在起始点到目标点的泊车路径；
第一检测计算模块330，设置为在存在起始点到目标点的泊车路径的情况下，对RS几何算法规划出的每个泊车路径进行碰撞检测以及成本计算；
第三确定模块340，设置为在RS几何算法规划出的所有泊车路径均为存在碰撞的泊车路径的情况下，利用混合A星算法确定目标车辆的备选状态点集合；
第二检测计算模块350，设置为对备选状态点集合中的每个备选状态点进行碰撞检测以及成本计算，直至输出成本最低、不碰撞、以及到达目标点的状态点为止。
本实施例的技术方案，在预先建立的泊车场景坐标系上，确定泊车路径的起始点和目标点；然后基于目标车辆的起始点和目标点利用混合A星算法搜索泊车路径，并对泊车路径进行碰撞检测，保证规划得到的泊车路径在满足车辆约束的情况下能够实现避撞。同时，在利用混合A星算法进行路径搜索过程中， 拼接了RS几何算法，提高了路径搜索的计算速度，保证了泊车路径搜索的实时性。
在一个实施例中，泊车路径规划装置，还包括：
第四确定模块，设置为在预先建立的泊车场景坐标系之前，在目标车辆的泊车功能开启的情况下，确定目标车辆的泊车类型和目标车位位置。
在一实施例中，泊车路径规划装置，还包括：
建立模块，设置为在确定泊车路径的起始点与目标点之后，在所述目标点有效的情况下，利用RS几何算法确定目标车辆的当前状态点是否存在起始点到目标点的泊车路径之前，根据车位搜索过程中超声波雷达信息建立障碍物地图信息；
第五确定模块，设置为根据障碍物地图信息确定目标车辆的可泊区域与障碍物坐标点。
在一实施例中，泊车场景坐标系的建立过程，包括：获取目标车辆的当前泊车参数，当前泊车参数包括下述：车辆最小转弯半径、后轴中点到前悬的距离、后轴中点到后悬的距离、车辆宽度和轴距；根据当前泊车参数、目标车辆的几何中心点和车头所指方向，建立对应的泊车场景坐标系。
在一实施例中，对RS几何算法规划出的每个泊车路径进行碰撞检测的过程，包括：确定目标车辆的四个边界点；根据四个边界点和障碍物坐标点之间的距离，确定每个泊车路径中是否存在与障碍物发生碰撞的路径点。
在一实施例中，对规划的每个泊车路径或者备选状态点进行成本计算的过程，包括：获取目标车辆的方向盘转角、方向盘转角变化量以及障碍物与目标车辆之间的当前距离；根据目标车辆的方向盘转角、方向盘转角变化量、障碍物与目标车辆之间的当前距离以及预先设置的方向盘权重系数、预先设置的方向盘变化权重系统和预先设置的距离障碍物远近的权重系数，确定泊车路径的成本或备选状态点的成本。
在一实施例中，利用混合A星算法确定目标车辆的备选状态点集合，包括： 在利用混合A星算法搜索更新过程中，获取目标车辆的控制量和状态量；控制量包括：行进方向、行进距离和方向盘转角；状态量包括：后轴中点的坐标和航向角；根据控制量以及状态量确定对应的至少两个备选状态点；将至少两个备选状态点组合成对应的备选状态点集合。
上述泊车路径规划装置可执行本申请任意实施例所提供的泊车路径规划方法，具备执行方法相应的功能模块。
图11是本申请一实施例提供的一种车辆的硬件结构示意图。如图11所示，本申请实施例提供的车辆，包括：存储器410，以及至少一个处理器420。该车辆中的处理器420可以是至少一个，图11中以一个处理器420为例，车辆中的处理器420和存储器410可以通过总线或其他方式连接，图11中以通过总线连接为例。
该车辆中的存储器410作为一种计算机可读存储介质，可设置为存储至少一个程序，程序可以是软件程序、计算机可执行程序以及模块，如本申请实施例所提供泊车路径规划方法对应的程序指令/模块(例如，图10所示的泊车路径规划装置中的模块，包括：第一确定模块310、第二确定模块320、第一检测计算模块330、第三确定模块340和第二检测计算模块350)。处理器410通过运行存储在存储器410中的软件程序、指令以及模块，从而执行设备的各种功能应用以及数据处理，即实现上述方法实施例中的泊车路径规划方法。
存储器410可包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需的应用程序；存储数据区可存储根据设备中所配置设备的使用所创建的数据等。此外，存储器410可以包括高速随机存取存储器，还可以包括非易失性存储器，例如至少一个磁盘存储器件、闪存器件、或其他非易失性固态存储器件。在一些实例中，存储器410可包括相对于处理器420远程设置的存储器，这些远程存储器可以通过网络连接至设备中所配置的设备。上述网络的实例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。
在一个实施例中，提供了一种车辆，包括存储器410和处理器420，该存储器410存储有计算机程序，该处理器420执行计算机程序时实现以下步骤：
在预先建立的泊车场景坐标系上，确定泊车路径的起始点和目标点；在目标点有效的情况下，利用RS几何算法确定目标车辆的当前状态点是否存在起始点到目标点的泊车路径；在存在起始点到目标点的泊车路径的情况下，对RS几何算法规划出的每个泊车路径进行碰撞检测以及成本计算；在RS几何算法规划出的所有泊车路径均为存在碰撞的泊车路径的情况下，利用混合A星算法确定目标车辆的备选状态点集合；对备选状态点集合中的每个备选状态点进行碰撞检测以及成本计算，直至输出成本最低、不碰撞、以及到达目标点的状态点为止。
上述车辆可执行本申请任意实施例所提供的泊车路径规划方法，具备执行方法相应的功能模块。
本申请实施例还提供了一种计算机可读存储介质，其上存储有计算机程序，该计算机程序被处理器执行时，实现本申请实施例提供的泊车路径规划方法，该方法包括：在预先建立的泊车场景坐标系上，确定泊车路径的起始点和目标点；在目标点有效的情况下，利用RS几何算法确定目标车辆的当前状态点是否存在起始点到目标点的泊车路径；在存在起始点到目标点的泊车路径的情况下，对RS几何算法规划出的每个泊车路径进行碰撞检测以及成本计算；在RS几何算法规划出的所有泊车路径均为存在碰撞的泊车路径的情况下，利用混合A星算法确定目标车辆的备选状态点集合；对备选状态点集合中的每个备选状态点进行碰撞检测以及成本计算，直至输出成本最低、不碰撞、以及到达目标点的状态点为止。
本申请实施例的计算机存储介质，可以采用至少一个计算机可读的介质的任意组合。计算机可读介质可以是计算机可读信号介质或者计算机可读存储介质。计算机可读存储介质例如可以是，但不限于电、磁、光、电磁、红外线、或半导体的系统、装置或器件，或者任意以上的组合。计算机可读存储介质的 更具体的例子(非穷举的列表)包括：具有至少一个导线的电连接、便携式计算机磁盘、硬盘、随机存取存储器(Random Access Memory，RAM)、只读存储器(Read-Only Memory，ROM)、可擦式可编程只读存储器((Erasable Programmable Read-Only Memory，EPROM)或闪存)、光纤、便携式紧凑磁盘只读存储器(Compact Disc Read-Only Memory，CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。在本文件中，计算机可读存储介质可以是任何包含或存储程序的有形介质，该程序可以被指令执行系统、装置或者器件使用或者与其结合使用。
计算机可读的信号介质可以包括在基带中或者作为载波一部分传播的数据信号，其中承载了计算机可读的程序代码。这种传播的数据信号可以采用多种形式，包括但不限于电磁信号、光信号或上述的任意合适的组合。计算机可读的信号介质还可以是计算机可读存储介质以外的任何计算机可读介质，该计算机可读介质可以发送、传播或者传输用于由指令执行系统、装置或者器件使用或者与其结合使用的程序。
计算机可读介质上包含的程序代码可以用任何适当的介质传输，包括，但不限于无线、电线、光缆、射频(Radio Frequency，RF)等等，或者上述的任意合适的组合。
可以以一种或多种程序设计语言或其组合来编写用于执行本申请操作的计算机程序代码，程序设计语言包括面向对象的程序设计语言，诸如Java、Smalltalk、C++，还包括常规的过程式程序设计语言，诸如“C”语言或类似的程序设计语言。程序代码可以完全地在用户计算机上执行、部分地在用户计算机上执行、作为一个独立的软件包执行、部分在用户计算机上部分在远程计算机上执行、或者完全在远程计算机或服务器上执行。在涉及远程计算机的情形中，远程计算机可以通过任意种类的网络，包括局域网(Local Area Network，LAN)或广域网(Wide Area Network，WAN)连接到用户计算机，或者，可以连接到外部计算机(例如利用因特网服务提供商来通过因特网连接)。
本申请在预先建立的泊车场景坐标系上，确定泊车路径的起始点和目标点；在目标点有效的情况下，利用RS几何算法确定目标车辆的当前状态点是否存在起始点到目标点的泊车路径；在存在起始点到目标点的泊车路径的情况下，对RS几何算法规划出的每个泊车路径进行碰撞检测以及成本计算；在RS几何算法规划出的所有泊车路径均为存在碰撞的泊车路径的情况下，利用混合A星算法确定目标车辆的备选状态点集合；对备选状态点集合中的每个备选状态点进行碰撞检测以及成本计算，直至输出成本最低、不碰撞、以及到达目标点的状态点为止。本申请实施例利用混合A星算法在泊车路径搜索过程中，拼接RS几何算法，提高了路径搜索的计算速度，保证了泊车路径搜索的实时性。

Claims (10)

  1. 一种泊车路径规划方法，包括：
    在预先建立的泊车场景坐标系上，确定泊车路径的起始点和目标点；
    响应于确定所述目标点有效，利用RS几何算法确定目标车辆的当前状态点是否存在所述起始点到所述目标点的泊车路径；
    响应于存在所述起始点到所述目标点的泊车路径的确定结果，对RS几何算法规划出的每个泊车路径进行碰撞检测以及成本计算；
    响应于确定RS几何算法规划出的所有泊车路径均为存在碰撞的泊车路径，利用混合A星算法确定所述目标车辆的备选状态点集合；
    对所述备选状态点集合中的每个备选状态点进行碰撞检测以及成本计算，直至输出成本最低、不碰撞、以及到达所述目标点的状态点为止。
  2. 根据权利要求1所述的方法，在所述预先建立泊车场景坐标系之前，还包括：响应于目标车辆的泊车功能开启，确定所述目标车辆的泊车类型和目标车位位置。
  3. 根据权利要求1所述的方法，在所述确定泊车路径的起始点与目标点之后，所述响应于所述目标点有效，利用RS几何算法确定所述目标车辆的当前状态点是否存在所述起始点到所述目标点的泊车路径之前，还包括：
    根据车位搜索过程中超声波雷达信息建立障碍物地图信息；
    根据所述障碍物地图信息确定所述目标车辆的可泊区域与障碍物坐标点。
  4. 根据权利要求1所述的方法，其中，所述泊车场景坐标系的建立过程，包括：
    获取所述目标车辆的当前泊车参数，所述当前泊车参数包括：车辆最小转弯半径、后轴中点到前悬的距离、后轴中点到后悬的距离、车辆宽度和轴距；
    根据所述当前泊车参数、目标车辆的几何中心点和车头所指方向，建立对应的泊车场景坐标系。
  5. 根据权利要求3所述的方法，其中，对RS几何算法规划出的每个泊车路径进行碰撞检测的过程，包括：
    确定所述目标车辆的四个边界点；
    根据所述四个边界点和障碍物坐标点之间的距离，确定每个所述泊车路径中是否存在与障碍物发生碰撞的路径点。
  6. 根据权利要求1所述的方法，其中，对规划的每个泊车路径或者备选状态点进行成本计算的过程，包括：
    获取所述目标车辆的方向盘转角、方向盘转角变化量以及障碍物与所述目标车辆之间的当前距离；
    根据所述目标车辆的方向盘转角、方向盘转角变化量、障碍物与所述目标车辆之间的当前距离以及预先设置的方向盘权重系数、预先设置的方向盘变化权重系统和预先设置的距离障碍物远近的权重系数，确定所述泊车路径的成本或备选状态点的成本。
  7. 根据权利要求1所述的方法，其中，所述利用混合A星算法确定所述目标车辆的备选状态点集合，包括：
    在利用混合A星算法搜索更新过程中，获取所述目标车辆的控制量和状态量；所述控制量包括：行进方向、行进距离和方向盘转角；所述状态量包括：后轴中点的坐标和航向角；
    根据所述控制量以及所述状态量确定对应的至少两个备选状态点；
    将所述至少两个备选状态点组合成对应的备选状态点集合。
  8. 一种泊车路径规划装置，包括：
    第一确定模块，设置为在预先建立的泊车场景坐标系上，确定泊车路径的起始点和目标点；
    第二确定模块，设置为在所述目标点有效的情况下，利用RS几何算法确定目标车辆的当前状态点是否存在所述起始点到所述目标点的泊车路径；
    第一检测计算模块，设置为在存在所述起始点到所述目标点的泊车路径的情况下，对RS几何算法规划出的每个泊车路径进行碰撞检测以及成本计算；
    第三确定模块，设置为在RS几何算法规划出的所有泊车路径均为存在碰撞 的泊车路径的情况下，利用混合A星算法确定所述目标车辆的备选状态点集合；
    第二检测计算模块，设置为对所述备选状态点集合中的每个备选状态点进行碰撞检测以及成本计算，直至输出成本最低、不碰撞、以及到达所述目标点的状态点为止。
  9. 一种车辆，包括：存储器，以及至少一个处理器；
    存储器，设置为存储至少一个程序；
    当所述至少一个程序被所述至少一个处理器执行，使得所述至少一个处理器实现如权利要求1-7中任一所述的泊车路径规划方法。
  10. 一种计算机可读存储介质，所述计算机可读存储介质上存储有计算机程序，所述计算机程序被处理器执行时，实现如权利要求1-7中任一所述的泊车路径规划方法。
PCT/CN2021/079540 2020-03-11 2021-03-08 一种泊车路径规划方法、装置、车辆和存储介质 WO2021180035A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010166385.4 2020-03-11
CN202010166385.4A CN111301409A (zh) 2020-03-11 2020-03-11 一种泊车路径规划方法、装置、车辆和存储介质

Publications (1)

Publication Number Publication Date
WO2021180035A1 true WO2021180035A1 (zh) 2021-09-16

Family

ID=71155163

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/079540 WO2021180035A1 (zh) 2020-03-11 2021-03-08 一种泊车路径规划方法、装置、车辆和存储介质

Country Status (2)

Country Link
CN (1) CN111301409A (zh)
WO (1) WO2021180035A1 (zh)

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113830079A (zh) * 2021-10-19 2021-12-24 同济大学 任意起始位姿的连续曲率泊车路径在线规划方法及系统
CN113887060A (zh) * 2021-10-13 2022-01-04 英博超算(南京)科技有限公司 一种新型的自动泊车系统车辆定位算法
CN114001745A (zh) * 2021-10-25 2022-02-01 东南大学 一种电动汽车泊车引导系统及其位置检测评价方法
CN114030463A (zh) * 2021-11-23 2022-02-11 上海汽车集团股份有限公司 一种自动泊车系统的路径规划方法及装置
CN114103934A (zh) * 2021-12-23 2022-03-01 岚图汽车科技有限公司 自动泊车路径规划方法、装置、设备及可读存储介质
CN114274952A (zh) * 2021-12-24 2022-04-05 重庆长安汽车股份有限公司 垂直车位自主泊车方法、系统、存储介质及电子设备
CN114355886A (zh) * 2021-12-06 2022-04-15 深圳优地科技有限公司 泊车位的选择方法、装置、设备以及可读存储介质
CN114516322A (zh) * 2022-02-24 2022-05-20 重庆长安汽车股份有限公司 一种车辆泊出的系统、方法、车辆及计算机存储介质
CN114613180A (zh) * 2022-02-22 2022-06-10 恒大新能源汽车投资控股集团有限公司 自主泊车方法、设备、车辆及停车场端服务器
CN114852060A (zh) * 2022-05-23 2022-08-05 广州小鹏自动驾驶科技有限公司 泊车控制方法、泊车控制装置、车辆和存储介质
CN115014380A (zh) * 2022-06-30 2022-09-06 智道网联科技(北京)有限公司 泊车路径规划方法以及装置、电子设备、存储介质
CN115609594A (zh) * 2022-12-15 2023-01-17 国网瑞嘉(天津)智能机器人有限公司 机械臂路径的规划方法、装置、上位控制端及存储介质
CN113887060B (zh) * 2021-10-13 2024-05-10 英博超算(南京)科技有限公司 一种新型的自动泊车系统车辆定位方法

Families Citing this family (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111301409A (zh) * 2020-03-11 2020-06-19 中国第一汽车股份有限公司 一种泊车路径规划方法、装置、车辆和存储介质
CN111746523A (zh) * 2020-06-30 2020-10-09 三一专用汽车有限责任公司 车辆泊车路径规划方法、装置、车辆和存储介质
CN111897341A (zh) * 2020-08-05 2020-11-06 三一专用汽车有限责任公司 泊车路径规划方法、装置和计算机可读存储介质
CN111923902B (zh) * 2020-08-10 2022-03-01 华人运通(上海)自动驾驶科技有限公司 一种泊车控制方法、装置、电子设备及存储介质
CN112092810B (zh) * 2020-09-24 2022-06-14 上海汽车集团股份有限公司 一种车辆泊出方法、装置及电子设备
CN112046469B (zh) * 2020-09-29 2022-07-05 上海汽车集团股份有限公司 一种自动泊车方法、系统、服务器及存储介质
CN112414422A (zh) * 2020-11-01 2021-02-26 北京航空航天大学 一种自动泊车的路径规划方法、装置及存储介质
CN112477850A (zh) * 2020-11-27 2021-03-12 北京罗克维尔斯科技有限公司 泊车路径规划方法、装置、车载设备和存储介质
CN112595337B (zh) * 2020-12-01 2023-08-15 苏州欧菲光科技有限公司 避障路径规划方法、装置、电子装置、车辆及存储介质
CN112379679B (zh) * 2021-01-15 2021-04-23 北京理工大学 一种无人车辆局部路径规划方法
CN112558617B (zh) * 2021-02-19 2022-01-18 深圳佑驾创新科技有限公司 泊车控制方法、装置、计算机设备和存储介质
CN112937557B (zh) * 2021-03-09 2022-08-12 东风汽车集团股份有限公司 一种基于曲率控制的代客泊车路径规划方法及系统
CN113548042A (zh) * 2021-05-17 2021-10-26 中国第一汽车股份有限公司 汽车防倾砸紧急避让控制方法、控制系统和汽车
CN113734200A (zh) * 2021-09-14 2021-12-03 北京超星未来科技有限公司 自动驾驶车辆无线充电混合式路径规划的方法及系统
CN117002480A (zh) * 2022-04-27 2023-11-07 广州汽车集团股份有限公司 一种泊车路径规划方法与系统
CN115214626B (zh) * 2022-04-28 2024-04-26 广州汽车集团股份有限公司 泊车控制方法、泊车控制装置、车辆及存储介质
CN114995440A (zh) * 2022-06-15 2022-09-02 中国第一汽车股份有限公司 车辆控制方法、装置、车辆及存储介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2015081022A (ja) * 2013-10-23 2015-04-27 クラリオン株式会社 自動駐車制御装置および駐車支援装置
US20160129907A1 (en) * 2014-11-12 2016-05-12 Hyundai Motor Company Driving path planning apparatus and method for autonomous vehicle
CN108121205A (zh) * 2017-12-13 2018-06-05 深圳市航盛电子股份有限公司 一种用于多种泊车场景的路径规划方法、系统及介质
CN109606354A (zh) * 2018-10-18 2019-04-12 同济大学 一种基于分层规划的自动泊车方法及辅助系统
CN110806218A (zh) * 2019-11-29 2020-02-18 北京京东乾石科技有限公司 泊车路径规划方法、装置和系统
CN111301409A (zh) * 2020-03-11 2020-06-19 中国第一汽车股份有限公司 一种泊车路径规划方法、装置、车辆和存储介质

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101912453B1 (ko) * 2012-02-17 2018-10-26 현대모비스 주식회사 장애물 검출 및 충돌 경보 장치 및 방법
US11066067B2 (en) * 2018-06-29 2021-07-20 Baidu Usa Llc Planning parking trajectory for self-driving vehicles
CN110040132A (zh) * 2019-04-19 2019-07-23 宝能汽车有限公司 智能泊车方法、装置和车辆
CN110834630A (zh) * 2019-10-22 2020-02-25 中国第一汽车股份有限公司 一种车辆的驾驶控制方法、装置、车辆及存储介质

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2015081022A (ja) * 2013-10-23 2015-04-27 クラリオン株式会社 自動駐車制御装置および駐車支援装置
US20160129907A1 (en) * 2014-11-12 2016-05-12 Hyundai Motor Company Driving path planning apparatus and method for autonomous vehicle
CN108121205A (zh) * 2017-12-13 2018-06-05 深圳市航盛电子股份有限公司 一种用于多种泊车场景的路径规划方法、系统及介质
CN109606354A (zh) * 2018-10-18 2019-04-12 同济大学 一种基于分层规划的自动泊车方法及辅助系统
CN110806218A (zh) * 2019-11-29 2020-02-18 北京京东乾石科技有限公司 泊车路径规划方法、装置和系统
CN111301409A (zh) * 2020-03-11 2020-06-19 中国第一汽车股份有限公司 一种泊车路径规划方法、装置、车辆和存储介质

Cited By (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113887060A (zh) * 2021-10-13 2022-01-04 英博超算(南京)科技有限公司 一种新型的自动泊车系统车辆定位算法
CN113887060B (zh) * 2021-10-13 2024-05-10 英博超算(南京)科技有限公司 一种新型的自动泊车系统车辆定位方法
CN113830079B (zh) * 2021-10-19 2023-09-01 同济大学 任意起始位姿的连续曲率泊车路径在线规划方法及系统
CN113830079A (zh) * 2021-10-19 2021-12-24 同济大学 任意起始位姿的连续曲率泊车路径在线规划方法及系统
CN114001745A (zh) * 2021-10-25 2022-02-01 东南大学 一种电动汽车泊车引导系统及其位置检测评价方法
CN114001745B (zh) * 2021-10-25 2023-11-24 东南大学 一种电动汽车泊车引导系统及其位置检测评价方法
CN114030463A (zh) * 2021-11-23 2022-02-11 上海汽车集团股份有限公司 一种自动泊车系统的路径规划方法及装置
CN114030463B (zh) * 2021-11-23 2024-05-14 上海汽车集团股份有限公司 一种自动泊车系统的路径规划方法及装置
CN114355886A (zh) * 2021-12-06 2022-04-15 深圳优地科技有限公司 泊车位的选择方法、装置、设备以及可读存储介质
CN114103934A (zh) * 2021-12-23 2022-03-01 岚图汽车科技有限公司 自动泊车路径规划方法、装置、设备及可读存储介质
CN114274952A (zh) * 2021-12-24 2022-04-05 重庆长安汽车股份有限公司 垂直车位自主泊车方法、系统、存储介质及电子设备
CN114274952B (zh) * 2021-12-24 2023-09-22 重庆长安汽车股份有限公司 垂直车位自主泊车方法、系统、存储介质及电子设备
CN114613180A (zh) * 2022-02-22 2022-06-10 恒大新能源汽车投资控股集团有限公司 自主泊车方法、设备、车辆及停车场端服务器
CN114516322B (zh) * 2022-02-24 2024-04-26 重庆长安汽车股份有限公司 一种车辆泊出的系统、方法、车辆及计算机存储介质
CN114516322A (zh) * 2022-02-24 2022-05-20 重庆长安汽车股份有限公司 一种车辆泊出的系统、方法、车辆及计算机存储介质
CN114852060A (zh) * 2022-05-23 2022-08-05 广州小鹏自动驾驶科技有限公司 泊车控制方法、泊车控制装置、车辆和存储介质
CN114852060B (zh) * 2022-05-23 2024-04-09 广州小鹏汽车科技有限公司 泊车控制方法、泊车控制装置、车辆和存储介质
CN115014380A (zh) * 2022-06-30 2022-09-06 智道网联科技(北京)有限公司 泊车路径规划方法以及装置、电子设备、存储介质
CN115609594A (zh) * 2022-12-15 2023-01-17 国网瑞嘉(天津)智能机器人有限公司 机械臂路径的规划方法、装置、上位控制端及存储介质
CN115609594B (zh) * 2022-12-15 2023-03-28 国网瑞嘉(天津)智能机器人有限公司 机械臂路径的规划方法、装置、上位控制端及存储介质

Also Published As

Publication number Publication date
CN111301409A (zh) 2020-06-19

Similar Documents

Publication Publication Date Title
WO2021180035A1 (zh) 一种泊车路径规划方法、装置、车辆和存储介质
CN108062094B (zh) 基于处理器实现车辆行驶轨迹规划的自主系统和方法
US11718318B2 (en) Method and apparatus for planning speed of autonomous vehicle, and storage medium
CN109521764B (zh) 车辆远程辅助模式
US10515321B2 (en) Cost based path planning for autonomous driving vehicles
US10948919B2 (en) Dynamic programming and gradient descent based decision and planning for autonomous driving vehicles
US10754339B2 (en) Dynamic programming and quadratic programming based decision and planning for autonomous driving vehicles
US10459441B2 (en) Method and system for operating autonomous driving vehicles based on motion plans
JP6269552B2 (ja) 車両走行制御装置
WO2021120200A1 (en) Spline curve and spiral curve based reference line smoothing method
CN113916246A (zh) 一种无人驾驶避障路径规划方法和系统
WO2019225268A1 (ja) 走行計画生成装置、走行計画生成方法、及び制御プログラム
US11584248B2 (en) Method of parking an autonomous driving vehicle for autonomous charging
US11353878B2 (en) Soft-boundary based path optimization for complex scenes for autonomous driving vehicles
US20220297718A1 (en) Navigation route planning method for autonomous vehicles
US20210229690A1 (en) Method for operating a motor vehicle for improving working conditions of evaluation units in the motor vehicle, control system for performing a method of this kind, and motor vehicle having a control system of this kind
US20200310451A1 (en) Path planning for complex scenes with self-adjusting path length for autonomous driving vehicles
US20220412755A1 (en) Autonomous vehicle routing with local and general routes
US20210300333A1 (en) A parking-trajectory generation method combined with offline and online solutions
JP6658968B2 (ja) 運転支援方法及び運転支援装置
US11414096B2 (en) QP spline path and spiral path based reference line smoothing method for autonomous driving
CN116476840B (zh) 变道行驶方法、装置、设备及存储介质
US11878712B2 (en) Trajectory planning with obstacle avoidance for autonomous driving vehicles
CN114084132A (zh) 控制方法与装置、终端设备
CN114701495B (zh) 车道偏离抑制方法、电子设备和存储介质

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: 21766950

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: 21766950

Country of ref document: EP

Kind code of ref document: A1