CN112606830B - Two-section type autonomous parking path planning method based on mixed A-algorithm - Google Patents
Two-section type autonomous parking path planning method based on mixed A-algorithm Download PDFInfo
- Publication number
- CN112606830B CN112606830B CN202011594837.5A CN202011594837A CN112606830B CN 112606830 B CN112606830 B CN 112606830B CN 202011594837 A CN202011594837 A CN 202011594837A CN 112606830 B CN112606830 B CN 112606830B
- Authority
- CN
- China
- Prior art keywords
- parking
- vehicle
- path
- minimum
- path planning
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 43
- 230000008859 change Effects 0.000 claims description 9
- 230000007613 environmental effect Effects 0.000 claims description 3
- 230000008569 process Effects 0.000 abstract description 17
- 238000004364 calculation method Methods 0.000 description 11
- 238000010586 diagram Methods 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 230000006399 behavior Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 206010019233 Headaches Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 231100000869 headache Toxicity 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000028838 turning behavior Effects 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
- B60W30/06—Automatic manoeuvring for parking
-
- 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
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
-
- 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
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
- B60W2050/0001—Details of the control system
- B60W2050/0019—Control system elements or transfer functions
- B60W2050/0028—Mathematical models, e.g. for simulation
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Human Computer Interaction (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention discloses a two-section autonomous parking path planning method based on a mixed A-algorithm, which comprises the following steps: dividing the parking path into a first section and a second section; the first section is a path from the vehicle entering the parking lot to the vehicle running to the minimum parking distance point, and the second section is a path from the minimum parking distance point to the parking termination point; when the distance between the vehicle and the parking ending point is the minimum parking distance, judging that the vehicle reaches the minimum parking distance point; and carrying out path planning on the first section by adopting a first heuristic function and carrying out path planning on the second section by adopting a second heuristic function through a mixed A algorithm. According to the two-section type autonomous parking path planning method based on the mixed A-algorithm, the parking path is divided into two parts, and different heuristic functions are respectively adopted for path planning for the two parts by combining the characteristics of the two parts of paths, so that the optimal path can be obtained in the path searching process with the minimum iteration times, and the path planning efficiency is improved.
Description
Technical Field
The invention belongs to the technical field of autonomous parking path planning, and particularly relates to a two-section autonomous parking path planning method based on a mixed A-algorithm.
Background
With the development of automobile technology and the improvement of living standard of people, automobile travel becomes necessary for people to live, and along with the automobile travel, roads and parking lots are crowded. Meanwhile, the requirements of people on the traveling quality of automobiles are higher and higher, and the realization of parking is always a task of headache of driver groups. In recent years, research on autonomous parking systems has been conducted, and how to plan a path from entering a parking lot to achieving a specific parking is a serious consideration. Autonomous parking path planning itself includes a forward driving process from entering a parking lot or cell to near a parking point and a reverse entering process from near the parking point to effect parking. And because the requirements of the two paths on the automobile are different, different node expansion modes should be adopted for the two paths. However, the existing researches generally only study one of the paths, and do not consider the two paths on the same problem. Two-stage autonomous parking path planning research is necessary.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, and provides a two-section type autonomous parking path planning method based on a mixed A algorithm, which divides a parking path into two parts according to the minimum parking distance required by a vehicle, and adopts different heuristic functions to carry out path planning on the two parts of paths respectively by combining the characteristics of the two parts of paths, so that the path searching process can obtain an optimal path with the minimum iteration times, the path planning efficiency is improved, and simple and efficient parking is realized.
The technical scheme provided by the invention is as follows:
a two-section autonomous parking path planning method based on a hybrid A-algorithm comprises the following steps:
dividing the parking path into a first section and a second section; the first section is a path from a vehicle entering a parking lot to a vehicle running to a minimum parking distance point, and the second section is a path from the minimum parking distance point to a parking termination point;
when the distance between the vehicle and the parking ending point is the minimum parking distance, judging that the vehicle reaches the minimum parking distance point;
carrying out path planning on the first segment by adopting a first heuristic function through a mixed A-type algorithm;
wherein the first heuristic function is:
cost(N i ，N i+1 )＝dis(N i ，N i+1 )*(1+a*backcost+b*turncost)；
carrying out path planning on the second section by adopting a second heuristic function through a mixed A-algorithm;
wherein the second heuristic is:
cost(N i ,N i+1 )＝dis(N i ,N i+1 )*{(1+a*backcost+b*turncost+cost[theta(N i ),theta(N i+1 )]}；
in the formula (N) i ,N i+1 ) Representing the slave current node N i Extending to the next node N i+1 Is not limited to the required cost of (1); dis (N) i ,N i+1 ) Representing the current node N i Extending to the next node N i+1 Is a distance from the current node N i Extending to the next node N i+1 The reversing cost and the steering cost; turncost represents the current node N i Extending to the next node N i+1 Steering cost theta (N) i ) Indicating that the vehicle is at the current node N i Is a course angle of (2); theta (N) i+1 ) Indicating that the vehicle is at the next node N i+1 Is a course angle of (2); cost [ theta (N) i ),theta(N i+1 )]Representing a heading angle theta (N) from a current node i ) Extending to the next node heading angle theta (N) i+1 ) The cost required; a. b represent weight coefficients, respectively.
Preferably, when parking vertically, the method for calculating the minimum parking distance includes:
S＝sinθ*R min +L；
wherein R is min A minimum turning radius indicating that the vehicle does not collide with the parking space; θ represents a minimum turning radius R of the vehicle at which the vehicle does not collide with the parking space min Front wheel corner when driving out of parking space; l represents the body length.
Preferably, when parallel parking is performed, the method for calculating the minimum parking distance is as follows:
S＝sin(θ 1 +θ 2 )*R min +L+L 2 ；
wherein R is min A minimum turning radius indicating that the vehicle does not collide with the parking space; θ 1 Indicating the angle of rotation, θ, of the vehicle away from the parking space 2 A corner indicating that the vehicle is driven out of the parking space and the posture of the vehicle body is recovered; l represents the length of the vehicle body, L 2 Indicating the distance between the front wheel center and the body center of the vehicle.
Preferably, before the parking path planning, the method further comprises: and performing rasterization and discretization on the environment modeling map.
Preferably, the two-stage autonomous parking path planning method based on the hybrid a-algorithm further includes: when the path planning is carried out, judging whether a vehicle collides with an environmental obstacle at the next node when node expansion is carried out each time, if so, discarding the expansion and carrying out node expansion planning again;
wherein, if the front pose is: [ x ] 0 ,y 0 ,theta 0 ]The pose of the next node is: [ x, y, theta ]]The method comprises the steps of carrying out a first treatment on the surface of the Then
x＝x 0 +D*cos(theta 0 )
y＝y 0 +D*sin(theta 0 )；
theta＝theta 0 +D/L*tan(delta)
Where G represents the resolution of the grid in the environment modeling map, D represents the resolution of the pose change for each node expansion, and delta represents the steering angle change value for each node expansion.
Preferably, in the first heuristic function, the value range of the backcost is 5-10, and the value range of the turncost is 5-10.
Preferably, in the second heuristic function, backcost=1 and turncost=1.
The beneficial effects of the invention are as follows:
according to the two-section type autonomous parking path planning method based on the mixed A-algorithm, the parking path is divided into two parts according to the minimum parking distance required by the vehicle, and different heuristic functions are respectively adopted for path planning for the two parts of paths by combining the characteristics of the two parts of paths, so that the optimal path can be obtained in the path searching process with the minimum iteration times, the path planning efficiency is improved, and simple and efficient parking is realized.
Drawings
Fig. 1 is a flowchart of a two-stage autonomous parking path planning method based on a hybrid a algorithm according to the present invention.
Fig. 2a-2b are schematic views of a hybrid a-Shepp node extension of the algorithm Reeds-Shepp according to the present invention.
Fig. 3 is a schematic diagram of a calculation process of a minimum parking distance during vertical parking according to the present invention.
Fig. 4 is a schematic diagram of a calculation process of a parallel parking distance during vertical parking according to the present invention.
FIG. 5 is a schematic diagram illustrating a heuristic function according to the present invention.
FIG. 6 is a schematic diagram illustrating a second heuristic function according to the present invention.
Detailed Description
The present invention is described in further detail below with reference to the drawings to enable those skilled in the art to practice the invention by referring to the description.
As shown in fig. 1, the invention provides a two-stage autonomous parking path planning method based on a hybrid a-algorithm, which comprises the following steps:
1. discretization of environment modeling map
(1) The hybrid a algorithm is mainly applicable to global path planning in the case where the environment map is known. Since the node extension method of the hybrid a algorithm is based on a grid map, it is first necessary to discretize a known environment map. When grid discretization is performed, the resolution G of the grid, the resolution D of pose change of each node expansion and the turning radian delta with the largest requirement on vehicle dynamics are mainly considered.
As shown in fig. 2, each time a node is expanded, if the current pose is [ x, y, theta ], the pose calculation formula of the next node is as follows:
x＝x+D*cos(theta)；
y＝y+D*sin(theta)；
theta＝theta+D/L*tan(delta)；
where G represents the resolution of the grid in the environment modeling map and D represents the resolution of the pose change for each node expansion.
Before performing the hybrid a search, the shortest path searched based on the hybrid a algorithm is required to be used as a heuristic value of the hybrid a algorithm, and the maximum number of times n=g/D of expansion of two grid nodes is required.
2. Parking path segment design
1. Calculation of minimum parking distance
The minimum parking distance required is calculated according to the size of the vehicle, the information of the parking space, the minimum turning radius and the parking mode, and vertical parking and parallel parking are mainly considered.
The invention gives the definition of the minimum parking distance: minimum turning radius R of vehicle without collision with parking space min The distance from the parking space to the vehicle body parallel to the advancing direction of the vehicle. The minimum parking distance S is calculated mainly based on the parking space information, the minimum turning radius R, and the vehicle body length L. Firstly, according to parking space information, the minimum turning radius R can be solved to obtain the minimum turning radius R which does not collide with the parking space min . Wherein the minimum turning radius R is provided by the factory entity of the vehicle. At the same time, the minimum parking distance S must be determined by knowing the minimum turning radius R at which the vehicle does not collide with the parking space min The turning distance and the straight-line distance included in the minimum parking distance S are obtained by the front wheel turning angle θ and the vehicle body length L when the vehicle exits the parking space. In order to more clearly express the solution of the minimum parking distance, the invention provides a specific solution method of the minimum parking distance of vertical parking and horizontal parking.
(1) And (3) vertical parking: as shown in fig. 3, the minimum parking distance should be satisfied, and the distance required to exit from the parking space at the maximum steering wheel angle and resume the normal forward running posture in the case where the vehicle body does not collide with the parking space. And simultaneously satisfies the minimum turning radius and collision-free constraint. The center of a parking space and a headstock are taken as reference points, and a vehicle uses a minimum turning radius R which does not collide with the parking space min The magnitude of the front wheel turning angle when the vehicle exits the parking space is θ, and at this time, the minimum parking distance S can be obtained from the vehicle body size information. S is the length L from the center of the rear wheel to the head of the vehicle 1 And R is R min * And sin theta. However, in order to take account of errors in the path planning algorithm and to ensure implementation of a specific parking, a value of the minimum parking distance S that is offset from the theoretical value is to be given. Based on the above-mentioned considerations,and simplifying the calculation, selecting s=sinθ×r min +L is a calculation formula of the minimum parking distance. The theoretical calculation of the minimum parking distance is shown in fig. 3.
(2) Parallel parking: as shown in fig. 4, the minimum parking distance should be satisfied when parking in parallel: under the condition that the vehicle body does not collide with the parking space, the vehicle body is driven out from the parking space at the maximum steering wheel corner and the distance required by the normal forward driving posture is recovered; and simultaneously satisfying the minimum turning radius and the collision-free constraint. For parallel parking, the direction of the vehicle is the same after the parking space and the parking space is driven out, and then the wheel corner driven out at the moment comprises: angle of rotation θ from the original direction when the vehicle leaves the parking space 1 And restoring the rotation angle of the body posture to be theta 2 . The vehicle uses the front-rear symmetrical plane of the vehicle body as a reference plane, and the vehicle uses the minimum turning radius R which does not collide with the parking space min The magnitude of the accumulated front wheel rotation angle when the vehicle exits the parking space is theta 1 +θ 2 At this time, the minimum parking distance S can be obtained from the vehicle body size information. S is the length L from the center of the rear wheel to the head of the vehicle 1 Distance L between front wheel center and vehicle body center 2 And R is R min *sin(θ 1 +θ 2 ) A kind of electronic device. Also, in order to take account of errors in the path planning algorithm and to ensure implementation of a specific parking, a value of the minimum parking distance S that is large relative to the theoretical value is to be given. Based on the above consideration, and simplifying the calculation, s=sin (θ 1 +θ 2 )*R min +L+L 2 . The theoretical calculation of the minimum parking distance is shown in fig. 4.
2. Design of two-section heuristic function
The parking path is divided into two parts according to the minimum parking distance required by the vehicle, namely, the forward running process from entering a parking lot or a district to a minimum parking distance point (the position of the mass center of the vehicle when the distance between the parking path and the parking end point is the minimum parking distance) and the reversing and warehousing process of parking from the minimum parking distance point are realized.
Different heuristic functions are designed based on a hybrid a algorithm according to the driving characteristics of the vehicles with two paths. During forward driving, the vehicle always runs towards the target point, the obstacle avoidance is mainly considered, and fewer steering and reversing actions are required as much as possible. The heuristic should give a larger weight to the steering cost and the reversing cost. In the reversing driving process, the requirements on the vehicle pose are high, so that the heuristic function gives a larger weight to the minimum turning radius, and gives a smaller weight to the steering and reversing cost, thereby obtaining the optimal path with the minimum iteration times in the path searching process.
After the minimum parking distance is solved, the parking path should be segmented based on the minimum parking distance S, and different heuristic functions should be designed for the two-segment path planning according to the characteristics of the two-segment path driving. When the mixed A-algorithm is used for searching paths, the paths of the expansion nodes are mainly performed through Reeds-Shepp curves, so that reversing and steering behaviors of the node expansion process are mainly considered when the mixed A-algorithm is designed. Fig. 2a-2b show the manner in which the Reeds-Shepp curves expand nodes, mainly involving 3 node expansion directions in front and back.
(1) From entering the parking lot to the minimum parking distance point: the section of the path is mainly a forward running process at a low speed, and the vehicle is considered to always run along the target point because the distance from the vehicle to the target point is long at the moment, so that the pose constraint of the vehicle is not excessively cared. Meanwhile, when the vehicle runs in the forward direction at a low speed, the time for reaching a target point or the energy of a driver is greatly increased by carrying out steering and reversing actions, so that the planned path does not have more steering and reversing actions, and larger values are given to steering cost turncost and reversing cost backcost as much as possible in the constraint of a heuristic function, so that the iteration times of searching and the steering and reversing actions are reduced, and the relatively better path is ensured. If the cost function is used to represent the slave current node N i Extending to the next node N i+1 The cost function is as follows:
cost(N i ，N i+1 )＝dis(N i ，N i+1 )*(1+a*backcost+b*turncost)；
in the above formula, the backcost and the turn cost are respectively at the current node N i Extend to the next sectionPoint N i+1 The reversing cost and the steering cost are both 5-10 in the value ranges of the backcost and the turncost; thereby serving to reduce the reversing and steering actions. a and b respectively represent the weights of various parameters, and can be respectively taken as the ratio of the length of a reversing path to the length of a steering path to the length of the whole path; for between two adjacent nodes, 1 can be taken directly. This is mainly due to the fact that the turning path and the reversing path occupy the length of the node expansion path, and for two adjacent nodes, only one path is generally mainly used, so that 1 is directly taken.
Fig. 6 is a simplified illustration of how the path segment is planned by reducing steering and reversing actions, thereby providing a basis for the design of a heuristic for the path segment. As shown in fig. 6, the vehicle enters the parking lot in the illustrated pose, and wants to reach the target point G, for example, path planning based on heuristic function one, the preferred extended path should be path 2. Because no turns are required at the beginning of the expansion of the path nodes and the entire path only needs one turn, while path 1 needs two turns.
(2) From the minimum parking distance point to the realization of parking: the section of path mainly realizes parking in a correct vehicle pose by steering and reversing, and cannot collide with a parking space during parking. The steering cost turncost and the reversing cost backcost should be given smaller values, even without adding extra steering cost and reversing cost. And the constraint of the dynamic course angle and the minimum turning radius should be increased so as to ensure that the vehicle can park in a better path. Using cost [ theta (Ni), theta (Ni+1)]To represent the heading angle theta (N) from the current node i ) Extending to the next node heading angle theta (N) i+1 ) The required cost is achieved, so that the pose is approximated, and the error of the final reached end point and the target end point pose is ensured to be within a threshold range. Then if the cost function is used to represent the slave current node N i Extending to the next node N i+1 The cost of the second heuristic function is estimated as follows:
cost(N i ,N i+1 )＝dis(N i ,N i+1 )*{(1+a*backcost+b*turncost+cost[theta(N i ),theta(N i+1 )]}；
and determining whether the expansion of the node can be achieved without collision under the minimum turning radius, and if not, discarding the expansion of the node. In the above formula, the backcost and the turn cost are respectively at the current node N i Extending to the next node N i+1 Cost of reverse and cost of steering, cost [ theta (Ni), theta (Ni+1)]Representing the required cost of the course angle change for both poses. According to the invention, the backseat cost and the turncost are respectively assigned to 1, so that the reversing and steering actions are not additionally limited; let cost [ theta (N) i ),theta(N i+1 )]The value is 5 times to 10 times of the radian change value of the two poses. And after the path nodes are expanded to the position and posture errors of the end point within a certain range, the whole path searching process is realized through a mixed A-type algorithm. a and b represent weights of various parameters respectively, and can be taken as a ratio of the length of a reversing path to the length of a steering path to the length of the whole path. For between two adjacent nodes, 1 can be taken directly.
Fig. 6 shows a specific planning result of the path, which is used for showing the main movement form of the vehicle when parking is realized. As can be seen from fig. 6, the section mainly comprises a turning behavior and a reversing behavior. The meaning and the effectiveness of the segmentation are further described, so that basis is provided for heuristic function design of path planning of the second segment.
3. Path generation
The node expansion mode of the mixed A-algorithm is mainly based on the mixed A-algorithm and an RS curve, wherein two left graphs of the graph represent six node expansion directions of the RS curve, namely forward straight, left turn, right turn, backward straight, left turn and right turn, and each steering angle change is delta. When node expansion is carried out from the current grid based on a heuristic function I, searching is carried out by using a mixed A-type algorithm, and the node expansion times n are firstly judged 1 Whether the current pose is smaller than n=g/D, and RS connection judgment is performed once every five nodes are expanded, namely whether smooth connection from the current pose to the next grid can be achieved or not is judged by traversing 48 RS curves. After RS connection is made, discretization is performed on the RS curve,namely, the RS curve is expanded through discrete poses, and when nodes are expanded each time, if the current pose is [ x, y, theta ]]The pose calculation formula of the next node is as follows:
x＝x+D*cos(theta)；
y＝y+D*sin(theta)；
theta＝theta+D/L*tan(delta)；
the node expansion is performed while the collision detection of the path is necessary, namely, whether the vehicle model collides with an environmental obstacle or not is detected, if collision occurs, the RS curve is abandoned, the searching is performed again, and finally, the path from entering the parking lot to the minimum parking distance point is obtained. In particular, the determination of whether the vehicle reaches the parking end point is made by determining whether the position errors of the vehicle pose [ x, y, theta ] and the end point are within a threshold range, that is, determining that the vehicle reaches the parking end point when the vehicle pose is [ x±Δx, y±Δy, theta±Δtheta ]. According to experience, deltax is less than or equal to 0.05 meter, deltay is less than or equal to 0.5 meter, and deltatheta is less than or equal to 0.05 radian. Similarly, the path planning from the minimum parking point to the specific process of parking is mainly performed based on a heuristic function II, and the node expansion mode is similar to that of the first-stage path planning as well as collision detection.
The invention mainly designs the parking path in a segmentation way based on a mixed A-algorithm. Firstly, the global path planning problem in the parking process is solved, and compared with the existing research, the method has more practical significance in considering path planning only near parking places, and meanwhile, the parking waiting time can be saved. Secondly, in the path planning process, the parking path is segmented based on the characteristics of the mixed A algorithm, so that the expansion number of nodes can be reduced as much as possible in both paths, and the path calculation time can be saved.
Although embodiments of the present invention have been disclosed above, it is not limited to the details and embodiments shown and described, it is well suited to various fields of use for which the invention would be readily apparent to those skilled in the art, and accordingly, the invention is not limited to the specific details and illustrations shown and described herein, without departing from the general concepts defined in the claims and their equivalents.
Claims (5)
1. The two-section type autonomous parking path planning method based on the mixed A algorithm is characterized by comprising the following steps of:
dividing the parking path into a first section and a second section; the first section is a path from a vehicle entering a parking lot to a vehicle running to a minimum parking distance point, and the second section is a path from the minimum parking distance point to a parking termination point;
when the distance between the vehicle and the parking ending point is the minimum parking distance, judging that the vehicle reaches the minimum parking distance point;
carrying out path planning on the first segment by adopting a first heuristic function through a mixed A-type algorithm;
wherein the first heuristic function is:
cost(N i ，N i+1 )＝dis(N i ，N i+1 )*(1+a*backcost+b*turncost)；
carrying out path planning on the second section by adopting a second heuristic function through a mixed A-algorithm;
wherein the second heuristic is:
cost(N i ,N i+1 )＝dis(N i ,N i+1 )*{1+a*backcost+b*turncost+cost[theta(N i ),theta(N i+1 )]}；
in the formula (N) i ,N i+1 ) Representing the slave current node N i Extending to the next node N i+1 Is not limited to the required cost of (1); dis (N) i ,N i+1 ) Representing the current node N i Extending to the next node N i+1 Is a distance from the current node N i Extending to the next node N i+1 The reversing cost and the steering cost; turncost represents the current node N i Extending to the next node N i+1 Steering cost theta (N) i ) Indicating that the vehicle is at the current node N i Is a course angle of (2); theta (N) i+1 ) Indicating that the vehicle is at the next node N i+1 Is a course angle of (2); cost [ theta (N) i ),theta(N i+1 )]Representing heading angle theta (N) from current node i ) Extending to the next node heading angle theta (N) i+1 ) The cost required; a. b represents weight coefficients, respectively;
the method for calculating the minimum parking distance during vertical parking comprises the following steps:
S＝sinθ*R min +L；
wherein R is min A minimum turning radius indicating that the vehicle does not collide with the parking space; θ represents a minimum turning radius R of the vehicle at which the vehicle does not collide with the parking space min Front wheel corner when driving out of parking space; l represents the length of the vehicle body;
the method for calculating the minimum parking distance during parallel parking comprises the following steps:
S＝sin(θ 1 +θ 2 )*R min +L+L 2 ；
wherein R is min A minimum turning radius indicating that the vehicle does not collide with the parking space; θ 1 Indicating the angle of rotation, θ, of the vehicle away from the parking space 2 A corner indicating that the vehicle is driven out of the parking space and the posture of the vehicle body is recovered; l represents the length of the vehicle body, L 2 Indicating the distance between the front wheel center and the body center of the vehicle.
2. The hybrid a-algorithm based two-segment autonomous parking path planning method of claim 1, further comprising, prior to parking path planning: and performing rasterization and discretization on the environment modeling map.
3. The hybrid a-algorithm based two-segment autonomous parking path planning method of claim 2, further comprising: when the path planning is carried out, judging whether a vehicle collides with an environmental obstacle at the next node when node expansion is carried out each time, if so, discarding the expansion and carrying out node expansion planning again;
wherein, if the front pose is: [ x ] 0 ,y 0 ,theta 0 ]The pose of the next node is: [ x, y, theta ]]The method comprises the steps of carrying out a first treatment on the surface of the Then
Where G represents the resolution of the grid in the environment modeling map, D represents the resolution of the pose change for each node expansion, and delta represents the steering angle change value for each node expansion.
4. The two-stage autonomous parking path planning method based on the hybrid a algorithm according to claim 3, wherein in the first heuristic function, the value range of the backcost is 5-10, and the value range of the turn cost is 5-10.
5. The hybrid a-algorithm based two-segment autonomous parking path planning method of claim 4, wherein in the second heuristic function, backcost = 1 and turn cost = 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011594837.5A CN112606830B (en) | 2020-12-29 | 2020-12-29 | Two-section type autonomous parking path planning method based on mixed A-algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011594837.5A CN112606830B (en) | 2020-12-29 | 2020-12-29 | Two-section type autonomous parking path planning method based on mixed A-algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112606830A CN112606830A (en) | 2021-04-06 |
| CN112606830B true CN112606830B (en) | 2023-12-29 |
Family
ID=75249365
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011594837.5A Active CN112606830B (en) | 2020-12-29 | 2020-12-29 | Two-section type autonomous parking path planning method based on mixed A-algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112606830B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113219990B (en) * | 2021-06-02 | 2022-04-26 | 西安电子科技大学 | Robot path planning method based on adaptive neighborhood and steering cost |
| CN113701771B (en) * | 2021-07-29 | 2023-08-01 | 东风悦享科技有限公司 | Parking path planning method and device, electronic equipment and storage medium |
| CN113581167B (en) * | 2021-09-01 | 2023-03-31 | 华域汽车***股份有限公司 | Full-automatic parking control method and controller based on straight-line circular arc path |
| CN113830079B (en) * | 2021-10-19 | 2023-09-01 | 同济大学 | Method and system for online planning continuous curvature parking path of arbitrary initial pose |
| CN113978547B (en) * | 2021-10-21 | 2022-08-02 | 江铃汽车股份有限公司 | Automatic driving steering control method and system |
| CN114255594B (en) * | 2021-12-28 | 2024-03-15 | 吉林大学 | Autonomous passenger parking motion planning and motion control method |
| CN115223389B (en) * | 2022-07-15 | 2023-11-21 | 西南交通大学 | Parking guidance path planning method based on dynamic road section cost |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110806218A (en) * | 2019-11-29 | 2020-02-18 | 北京京东乾石科技有限公司 | Parking path planning method, device and system |
| CN111307152A (en) * | 2020-02-18 | 2020-06-19 | 中国科学院合肥物质科学研究院 | Reverse generation planning method for autonomous parking path |
| AU2020101761A4 (en) * | 2020-08-11 | 2020-09-17 | Nanjing University Of Science & Technology | Method for planning path of parking agv based on improved dijkstra algorithm |
| CN111768647A (en) * | 2020-05-29 | 2020-10-13 | 同济大学 | Autonomous parking method and device based on mobile edge calculation |
| CN111897341A (en) * | 2020-08-05 | 2020-11-06 | 三一专用汽车有限责任公司 | Parking path planning method, parking path planning device and computer-readable storage medium |
-
2020
- 2020-12-29 CN CN202011594837.5A patent/CN112606830B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110806218A (en) * | 2019-11-29 | 2020-02-18 | 北京京东乾石科技有限公司 | Parking path planning method, device and system |
| CN111307152A (en) * | 2020-02-18 | 2020-06-19 | 中国科学院合肥物质科学研究院 | Reverse generation planning method for autonomous parking path |
| CN111768647A (en) * | 2020-05-29 | 2020-10-13 | 同济大学 | Autonomous parking method and device based on mobile edge calculation |
| CN111897341A (en) * | 2020-08-05 | 2020-11-06 | 三一专用汽车有限责任公司 | Parking path planning method, parking path planning device and computer-readable storage medium |
| AU2020101761A4 (en) * | 2020-08-11 | 2020-09-17 | Nanjing University Of Science & Technology | Method for planning path of parking agv based on improved dijkstra algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112606830A (en) | 2021-04-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112606830B (en) | Two-section type autonomous parking path planning method based on mixed A-algorithm | |
| CN109927716B (en) | Autonomous vertical parking method based on high-precision map | |
| CN102495631B (en) | Intelligent control method of driverless vehicle tracking desired trajectory | |
| TW202019743A (en) | Driving track planning system and method for self-driving car capable of enhancing accuracy and safety of track planning | |
| Pérez et al. | Trajectory generator for autonomous vehicles in urban environments | |
| CN111679678B (en) | Track planning method and system for transverse and longitudinal separation and computer equipment | |
| CN111338335A (en) | Vehicle local track planning method under structured road scene | |
| Zheng et al. | RRT based path planning for autonomous parking of vehicle | |
| CN104977933A (en) | Regional path tracking control method for autonomous land vehicle | |
| CN113689735B (en) | Vehicle lane change collision early warning method and device | |
| CN113978452B (en) | Automatic parallel parking path planning method | |
| CN112269384B (en) | Vehicle dynamic trajectory planning method combining obstacle behavior intention | |
| US11480962B1 (en) | Dynamic lane expansion | |
| CN102587237A (en) | Right-turning lane design method by considering turning characteristic of large vehicle | |
| CN114162113A (en) | Parallel parking path planning method combined with multi-section curve | |
| CN116182884A (en) | Intelligent vehicle local path planning method based on transverse and longitudinal decoupling of frenet coordinate system | |
| Wu et al. | Dynamic path planning for autonomous driving on branch streets with crossing pedestrian avoidance guidance | |
| CN109606355A (en) | Auxiliary is parked the method for trajectory planning | |
| Wang et al. | Research on self-parking path planning algorithms | |
| CN113815645B (en) | Automatic driving behavior decision system and motion planning method suitable for annular intersection | |
| CN115903828A (en) | Automatic parking path planning method based on four-wheel steering of vehicle | |
| CN116360420A (en) | CLF-CBF optimized unmanned vehicle motion planning control method | |
| CN114537383A (en) | Internet vehicle curve collision early warning method and system based on driver steady state preview | |
| CN108725579A (en) | Direction-controlling method and system, model, terminal during parking and vehicle | |
| CN111857112B (en) | Automobile local path planning method and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |