CN111307152A - Reverse generation planning method for autonomous parking path - Google Patents
Reverse generation planning method for autonomous parking path Download PDFInfo
- Publication number
- CN111307152A CN111307152A CN202010098379.XA CN202010098379A CN111307152A CN 111307152 A CN111307152 A CN 111307152A CN 202010098379 A CN202010098379 A CN 202010098379A CN 111307152 A CN111307152 A CN 111307152A
- Authority
- CN
- China
- Prior art keywords
- path
- vehicle
- parking space
- parking
- paths
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/20—Instruments for performing navigational calculations
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/14—Traffic control systems for road vehicles indicating individual free spaces in parking areas
- G08G1/141—Traffic control systems for road vehicles indicating individual free spaces in parking areas with means giving the indication of available parking spaces
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
Abstract
The invention relates to an autonomous parking path reverse generation planning method, which comprises the following steps: step 1, obtaining target parking space information, including parking space type and position information; step 2, determining a parking initial state and a target state according to the target parking space information; step 3, generating a corresponding target path tree by using a geometric method according to the parking space information and the target state; step 4, generating a plurality of paths connected with the destination path tree end point by using a sampling method according to the initial state of the vehicle; and 5, selecting an optimal path from the multiple paths according to the cost function. The parking method provided by the invention can generate paths for parking into the parking spaces from the initial positions aiming at different parking spaces and different initial poses of vehicles by utilizing the prior parking space information.
Description
Technical Field
The invention relates to an automatic parking technology, in particular to a method for planning a parking path.
Background
With the development of the automatic driving technology, the research on the automatic parking technology is also paid attention to extensively, and the problem of path planning of automatic parking is also paid attention to extensively in academic and industrial fields. In order to realize the autonomous parking function, the task of planning the autonomous parking path is to find an optimal collision-free track when the position and the type information of a parking space are given, and to control the vehicle to move to a target position under the constraint of vehicle movement. According to parking habits, people are usually used to park in a backing garage mode, when backing into a garage, a vehicle needs to travel a distance forwards and then back into a parking space, however, the task is difficult for autonomous parking, in the path planning process, path planning of automatic driving and path planning of parking are usually carried out separately, the parking process also involves two paths, and the problem of combining the path of a forward section and the path of a backing section means that the process is more complex than the task of planning the path of ordinary autonomous driving.
The traditional parking path planning method generally adopts an indirect method, a vehicle with a front section track runs along a straight line to search a parking space, a vehicle with a back track generally adopts a minimum turning radius method to roughly track the track, and the track has large error, so that the vehicle collides with surrounding vehicles, and the failure of parking is caused. For different parking space types, the indirect method needs to adopt different trajectory planning methods. Because the vehicle needs to travel along a straight line to search for the parking space, when the vehicle has a small error, a large error is generated in the parking process, and the failure of parking is usually caused.
In patent CN102975715A, a method for planning parallel parking paths of a vehicle in any posture is described, in which a fitting lattice needs to be established, a spline curve is fitted by traversing the lattice connecting the start point and the end point of the vehicle, and finally a trajectory is selected by combining with the kinematic constraint of the vehicle. This method is only for one parking space type, and the path generation process needs to fit a lattice, increasing the difficulty of path generation.
A path planning method for automatic parallel parking is provided in the patent with application number 201810071201.9, which performs simulation to obtain a parallel parking path plan according to the geometric parameters of the vehicle type to be parked. This method is only suitable for parking path planning with a fixed starting position.
The methods are planned from the backing section track, the influence of different poses on the parking track is effectively solved, but for the automatically driven vehicle, the vehicle needs to autonomously drive to the starting position of the backing section during parking, how the vehicle drives to the starting point of the backing section is not explained in the methods, and the combination of autonomous parking path planning and automatic driving path planning is not mentioned.
Disclosure of Invention
The method is suitable for parking path planning of different parking space types and different initial parking poses, and can be combined with path planning of automatic driving to overcome the defects of the existing parking path planning method.
The invention provides an autonomous parking path reverse generation planning method, which comprises the following steps:
step 2, determining a parking initial state and a target state according to the target parking space information;
step 3, generating a corresponding target path tree by using a geometric method according to the parking space information and the target state;
step 4, generating a plurality of paths connected with the destination path tree end point by using a sampling method according to the initial state of the vehicle;
and 5, selecting an optimal path from the multiple paths according to the cost function.
Further, the step 1 specifically includes: the autonomous parking vehicle acquires parking space information through a communication mode based on V2X or through storage, wherein the parking space information comprises horizontal and vertical coordinate information of a parking space relative to the vehicle, heading information when the vehicle is parked in the parking space, and a final pose E (X) of the vehicle when the vehicle is parked in the parking spacee,Ye,θe) Wherein X ise,Ye,θeRespectively is the abscissa, ordinate and course of the final position of the vehicle.
Further, the step 2 specifically includes: when the vehicle runs near the parking space, the vehicle acquires the current pose S (X) through a sensors,Ys,θs)，Xs、Ys、θsNamely the abscissa, the ordinate and the course of the current position of the vehicle.
Further, the step 3 specifically includes: the vehicle takes the current pose as a starting point and moves to an end point state G (X)g,Yg,θg) And planning a path, and generating a target tree by taking the terminal state as a starting point on the assumption that the vehicle is already parked in the parking space and the vehicle is driven out of the parking space based on a reverse driving method.
Further, the step 4 specifically includes: and generating a plurality of parking prediction paths by connecting the generated paths to the target tree from the current position, namely the current pose, of the vehicle and taking the target tree end point as an end point of path generation.
Further, when the path is generated, the end point of the target tree is used as a sampling point, a cubic Hermite spline curve generation method with the minimum curvature change rate is used for generating a cluster of smooth curves connecting the current vehicle position to the end point of the target tree, and a plurality of smooth curves from the current pose S (X) are obtaineds,Ys,θs) To the final pose E (X)e,Ye,θe) Each path representing a state that the vehicle may perform,
further, the step 5 specifically includes: in the generated parking prediction path, for part of the path, the curvature of the path is too large, the vehicle kinematic model constraint is not satisfied, obstacles exist on part of the path, the estimation basis is the distance from the obstacles, the smoothness degree and the path length, the weighted calculation is carried out on the estimation basis, and the cost function C is combined withfAnd selecting an optimal path, namely the final parking path.
Further, the method for generating the target path tree in step 3 includes: a parallel parking space target tree generation method, a vertical parking space target tree generation method and an inclined parking space target tree generation method.
Further, the parallel parking space target tree generating method, the vertical parking space target tree generating method and the inclined parking space target tree generating method specifically include the following steps:
assuming that a vehicle is parked in a parking space, driving the vehicle out of the parking space by various predefined paths, wherein the point E is a starting point of driving the vehicle out of the parking space, the point E1 is a first steering point, and G1-Gi are reversing points; and calculating by using E1 points as starting points and different turning radii R to generate a plurality of paths E1Gi, and obtaining a plurality of reversing paths from G1 to Gi by adopting a reverse planning method, namely different paths for reversing the vehicle into the parking space.
E1Gi[k].x＝E1.x+(R*sin(θ))
E1Gi[k].y＝E1.y+(R-R*cos(θ))
In the formula, R is the turning radius of the vehicle corresponding to different steering wheel turning angles, theta represents the included angle between a road point and a road starting point on an arc path in the generation process of the road point, k represents a plurality of paths, namely the kth path corresponding to the target tree, x and y respectively represent the horizontal and vertical coordinates of the road point relative to the position of the vehicle, the plurality of paths are generated by utilizing the formula, and the plurality of paths form a group of predicted paths for the vehicle to select to run, namely the target tree, and are a group of arcs with different radii.
Further, the optimal path includes two parts: a front-end path from a portion of the sampling method and a back-end path from a portion of the target tree.
Has the advantages that:
the parking method provided by the invention can generate paths for parking into the parking spaces from the initial positions aiming at different parking spaces and different initial poses of vehicles by utilizing the prior parking space information. The method comprises the steps of adopting a reverse planning mode, assuming that a vehicle is already parked in a parking space, when the vehicle drives out of the parking space, driving out the parking space by a plurality of methods according to different driving-out strategies, wherein the different driving-out strategies represent that different paths can drive the vehicle out of the parking space, the paths driven out of the parking space can be regarded as a group of path trees, the process that the vehicle drives out of the parking space is reversed, the process that the vehicle drives into the parking space can be regarded as the process that the vehicle drives into the parking space, adopting a sampling method, generating paths connected with the path trees from the current vehicle position, and screening through a cost function, so that the final parking path can be.
Drawings
FIG. 1 parking space type;
FIG. 2 illustrates a parallel parking space target tree of the present invention;
FIG. 3 illustrates a vertical parking space target tree of the present invention;
FIG. 4 illustrates a skewed parking space target tree of the present invention;
FIG. 5 is a state space sampling of the present invention;
FIG. 6 is a schematic diagram of an autonomous parking path planning process for parallel parking spaces according to the present invention;
FIG. 7 is a schematic diagram of an autonomous parking path planning for parallel parking spaces according to the present invention;
FIG. 8 is a schematic diagram of a parallel parking path planning process for different starting poses of the present invention;
FIG. 9 is a schematic diagram of a vertical parking path planning process for different initial poses of the present invention;
fig. 10 is a schematic diagram of an oblique parking path planning process with different initial poses according to the present invention.
The specific implementation mode is as follows:
the technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, rather than all embodiments, and all other embodiments obtained by a person skilled in the art based on the embodiments of the present invention belong to the protection scope of the present invention without creative efforts.
The parking method provided by the invention can generate paths for parking into the parking spaces from the initial positions aiming at different parking spaces and different initial poses of vehicles by utilizing the prior parking space information. The method comprises the steps of adopting a reverse planning mode, assuming that a vehicle is already parked in a parking space, when the vehicle drives out of the parking space, driving out the parking space by a plurality of methods according to different driving-out strategies, wherein the different driving-out strategies represent that different paths can drive the vehicle out of the parking space, the paths driven out of the parking space can be regarded as a group of path trees, the process that the vehicle drives out of the parking space is reversed, the process that the vehicle drives into the parking space can be regarded as the process that the vehicle drives into the parking space, adopting a sampling method, generating paths connected with the path trees from the current vehicle position, and screening through a cost function, so that the final parking path can be.
An autonomous parking path reverse generation planning method is used for planning autonomous parking paths of vehicles with different starting poses and different parking space types, and specifically comprises the following steps:
step 2, determining a parking initial state and a target state according to the target parking space information;
step 3, generating a corresponding target path tree by using a geometric method according to the parking space information and the target state;
step 4, generating a plurality of paths connected with the destination path tree end point by using a sampling method according to the initial state of the vehicle;
step 5, selecting an optimal path according to the cost function;
the method has no requirement on the initial pose of the vehicle during parking, the path generation is based on a geometric method, the path of the vehicle for driving out of the parking space is obtained through the geometric method, the geometric method is simple in calculation, the vehicle drives out of the parking space according to different turning radii, a path tree is generated, different path trees are generated for different parking spaces, and the problems that the parking strategy of the geometric method is limited by the shape of the parking space, the requirement on the parking initial position is strict through the geometric method, and the parking operation is failed due to the fact that the improper initial position can be caused are solved.
The parking path planning further comprises a parking path generating step, as follows:
referring to fig. 1, parking spaces are generally divided into three types: vertical parking spaces, parallel parking spaces and inclined parking spaces. The invention firstly assumes that the automobile is already parked in the parking space, and generates three different target trees according to different parking spaces, wherein the three different target trees are applied to the three different parking spaces. Each target tree isThere are tens of paths. The pose of the vehicle after being parked in the parking space is the final pose E (X)e,Ye,θe) (i.e., vehicle position abscissa, heading) from the final pose E (X)e,Ye,θe) Initially, a set of target trees of tens of paths is generated with different turning radii, and the end point state of the target tree is G (X)g,Yg,θg) (i.e., horizontal and vertical coordinates, direction of destination tree end point location) then the present invention uses a sampling method to derive the starting pose S (X) of the vehicles,Ys,θs) (i.e., the horizontal and vertical coordinates of the current position of the vehicle, the heading) and the end point G (X) of the target tree to be generatedg,Yg,θg) And finally, selecting an optimal track by combining conditions such as kinematic constraint, obstacle constraint, track length and the like as a sampled target point, and then generating a final path. This path consists of two parts: a front-end path from a portion of the sampling method and a back-end path from a portion of the target tree.
Step 2, when the vehicle runs to the position near the parking space, the vehicle acquires the current pose S (X) through a sensors,Ys,θs) And taking the current pose as a starting point and moving to an end point state G (X)g,Yg,θg) And (6) planning the path.
The method for generating the target path tree in step 3 correspondingly comprises the following steps:
(1) parallel parking space target tree generation method
As shown in fig. 2, the parallel parking space target tree: a vehicle exits a parallel parking space which will maintain the maximum turning angle and will travel out of the parking space, point E being the starting point for the vehicle to exit the parking space, point E1 being the first turning point, and points G1-Gi being the reversing points in the figure. Steering wheel angles at points E to E1 allow the vehicle to be driven at the minimum radius, which is obviously the most relaxed parking environment and more space efficient. Meanwhile, when the vehicle is driven out to the point E1, the vehicle will have more various methods to drive out of the parking space, when the vehicle is driven at a fixed steering wheel angle, a path with a fixed turning radius is generated, and when the steering wheel angles of the vehicles are different, a plurality of paths are generated from G1 to Gi.
E1Gi[k].x＝E1.x+(R*sin(b))
E1Gi[k].y＝E1.y+(R-R*cos(b))
The method comprises the following steps that R is vehicle turning radius corresponding to different steering wheel turning angles, b represents an included angle between a road point and a road starting point on an arc path in a path point generating process, k represents a plurality of paths, namely a kth path corresponding to a target tree, x and y respectively represent horizontal and vertical coordinates of the road point relative to a vehicle position, the paths can be generated by the formula, the paths form a group of predicted paths which can be selected by a vehicle and are used as the target tree, and the target tree is formed by a group of arcs with different radiuses as shown in figure 2.
(2) Vertical parking space target tree generation method
As shown in fig. 3, the vertical parking space target tree: a vehicle exits a vertical parking space and will travel outside the parking space maintaining a straight line, wherein point E is the starting point of the vehicle exiting the parking space, point E1 is the first turning point, and points G1-Gi are the destination of the target tree. And (3) keeping the straight-line driving state of the vehicle from E to E1, driving the vehicle forwards for a certain distance, and when the vehicle drives out to E1, driving the vehicle out of the vertical parking space by more methods, generating other paths with fixed turning angles by taking the minimum turning radius as a constraint from G1 to Gi, wherein the generation method of the target tree for the vertical parking space and the generation method of the target tree for the parallel parking space are parking prediction paths consisting of a group of arcs with different radii.
(3) Inclined parking space target tree generation method
As shown in fig. 4, the diagonal parking space target tree: a vehicle exits a parking space in a diagonal direction, and the vehicle will first keep a straight line and travel out of the parking space, wherein point E is the starting point of the vehicle exiting the parking space, point E1 is the first turning point, and points G1-Gi are the destination tree points. And (3) keeping the straight-line driving state of the vehicle from E to E1, driving the vehicle forwards for a certain distance, and when the vehicle drives out to E1, driving the vehicle out of the vertical parking space by more methods, generating other paths with fixed turning angles from G1 to Gi by taking the minimum turning radius as a constraint, wherein the inclined parking space target tree generation method and the parallel parking space target tree generation method are parking prediction paths consisting of a group of arcs with different radii.
Method for generating front path
The reference path is a path curve for guiding the unmanned vehicle to pass through the task road section to reach the terminal, when the reference path exists, the sampling method samples in a state space along the reference path, a cubic Hermite spline curve generation method with the minimum curvature change rate is used for generating a cluster of smooth curves, the curves with the minimum steering wheel variation are represented, and a feasible path set is generated, so that the motion controller can track easily and simultaneously accord with the road shape constraint.
A set of terminal states of the state space is first sampled according to road geometry using multi-resolution sampling, as shown in fig. 5. And uniformly taking points along the reference path at the pre-aiming distance rho by the state of the sampling terminal. Preview distance P, i.e. P in the figure0The distance to Pg should be greater than the minimum collision distance dminAnd is smaller than the maximum sensing distance dmax。
Evaluation basis of selected track: distance to an obstacle; the degree of smoothing; and errors from historical track data. Finally, an optimal trajectory is selected based on these evaluations.
In the present invention, when the vehicle is within the parking area during parking, the current pose S (X) of the vehicle is followeds,Ys,θs) And no reference path is used at the end point of the target tree, so that when the path is generated, the end point of the target tree is used as a sampling point, and a cubic Hermite spline curve generation method with the minimum curvature change rate is used for generating a cluster of smooth curves connecting the current vehicle position to the target tree, so that a plurality of smooth curves from the current pose S (X) can be obtaineds,Ys,θs) To the final pose E (X)e,Ye,θe) And finally, selecting the optimal path by performing weighted calculation on the evaluation basis according to the distance from the obstacle, the smoothness and the path length as the evaluation basis.
Cost function calculation
Distance of vehicle from obstacle position:
Cobdis＝f((xs,ys),(xob,yob))
(xs,ys) As the coordinates of the position of the vehicle body at the present time, (x)ob,yob) Is the position coordinates of the obstacle;
path length:
Cleng.h＝f((xs,ys),(xg,yg))+f((xg,yg),(xe,ye))
the evaluation of the path length is divided into two parts, mainly comprising the current position (x) of the vehicles,ys) Position to the end of the target tree (x)g,yg) And target tree endpoint location (x)g,yg) Distance to final position (x)e,ye)；
Smoothness of the trajectory:
Csmooth＝f(k)
ki represents the curvature of each point in the planned path:
wherein the content of the first and second substances,
4θi＝|θi+9-θi|
x, y represent the horizontal and vertical coordinates of the waypoints, Δ 6iIs the distance between two adjacent waypoints;
path Smoothness (smoothenness, f (k)):
n represents the total number of waypoints;
and obtaining an optimal path by performing weighted calculation on the formula:
Cf＝αCobdis+βClength+γCsmooth
α, gamma corresponds to Cobdis，Clength，CsmoothThe cost of the track can be calculated by the distributed weight values, and the minimum value of the cost is selected as an optimal path, namely a final parking path.
The specific path planning implementation method comprises the following steps:
on the premise of meeting the requirements of a kinematic model and constraints, how to plan a reasonable parking path in a parking area is the key of the unmanned vehicle automatic parking system. The parking path finally generated by adopting the method consists of two parts: the front-end path is from a portion of the sampling method and the back-end path is from a portion of the target tree.
As shown in FIG. 6, assume that the current starting pose of the vehicle is S (X)s,Ys,θs) The parking space type is known as a parallel parking space, and the position is known as E (X)e,Ye,θe) In the Gi parking prediction paths, some paths do not meet the vehicle kinematic model constraint because of overlarge curvature of the paths, and some paths have obstacles, so that the cost function C is combinedfAn optimal path may be selected.
Combining cost function CfThe final path is taken as shown in fig. 7, and the path is the final parking path. The vehicle tracks the path to park in the parking space.
An implementation mode of autonomous parking path planning with different initial poses is as follows:
as shown in fig. 8, when parking horizontally, a plurality of paths are generated from a parking space (only three representative paths are selected in the figure, and when a vehicle runs to a path tree end point, three paths which are perpendicular to, parallel to and form a certain included angle with the parking space are formed), a path tree is formed, the vehicle forms a plurality of paths by taking the path tree end point as a sampling point from any pose, finally an optimal path is selected by combining conditions such as kinematic constraint, path smoothness, obstacle constraint, path length and the like, and if the path represented by a dashed-dotted line in the figure is an optimal path, the vehicle tracks the dashed-dotted line path, then tracks a target tree path connected with the dashed-dotted line path, and parks in the parking space.
As shown in fig. 9, when parking vertically, multiple paths are generated from the parking space to form a target path tree, the vehicle forms multiple paths from any pose with the end point of the path tree as a sampling point, and finally, an optimal path is selected by combining conditions such as kinematic constraint, trajectory smoothness, obstacle constraint, path length and the like.
As shown in fig. 10, when parking in an inclined manner, multiple paths are generated from a parking space to form a target path tree, a vehicle forms multiple paths from any pose and taking a path tree end point as a sampling point, and finally an optimal path is selected by combining conditions such as kinematic constraint, trajectory smoothness, obstacle constraint, path length and the like.
The invention is applicable to various parking scenes, and the parking paths obtained by applying the invention to the parallel parking spaces, the vertical parking spaces and the inclined parking spaces with different parking poses are respectively shown in the figures 8, 9 and 10. It should be noted that the present invention does not limit the obstacles around the parking space to other vehicles, but may also limit the number of obstacles around the parking space to 1, and other numbers of obstacles are also applicable to the application scenario of the present invention.
Although illustrative embodiments of the present invention have been described above to facilitate the understanding of the present invention by those skilled in the art, it should be understood that the present invention is not limited to the scope of the embodiments, but various changes may be apparent to those skilled in the art, and it is intended that all inventive concepts utilizing the inventive concepts set forth herein be protected without departing from the spirit and scope of the present invention as defined and limited by the appended claims.
Claims (10)
1. An autonomous parking path reverse generation planning method is characterized by comprising the following steps:
step 1, obtaining target parking space information, including parking space type and position information;
step 2, determining a parking initial state and a target state according to the target parking space information;
step 3, generating a corresponding target path tree by using a geometric method according to the parking space information and the target state;
step 4, generating a plurality of paths connected with the destination path tree end point by using a sampling method according to the initial state of the vehicle;
and 5, selecting an optimal path from the multiple paths according to the cost function.
2. The autonomous parking path reverse generation planning method according to claim 1, characterized in that:
the step 1 specifically comprises: the autonomous parking vehicle acquires parking space information through a communication mode based on V2X or through storage, wherein the parking space information comprises horizontal and vertical coordinate information of a parking space relative to the vehicle, heading information when the vehicle is parked in the parking space, and a final pose E (X) of the vehicle when the vehicle is parked in the parking spacee,Ye,θe) Wherein X ise,Ye,θeRespectively final for the vehiclePosition abscissa, ordinate, course.
3. The autonomous parking path reverse generation planning method according to claim 1, characterized in that:
the step 2 specifically comprises: when the vehicle runs near the parking space, the vehicle acquires the current pose S (X) through a sensors,Ys,θs)，Xs、Ys、θsNamely the abscissa, the ordinate and the course of the current position of the vehicle.
4. The autonomous parking path reverse generation planning method according to claim 1, characterized in that:
the step 3 specifically comprises: the vehicle takes the current pose as a starting point and moves to an end point state G (X)g,Yg,θg) And planning a path, and generating a target tree by taking the terminal state as a starting point on the assumption that the vehicle is already parked in the parking space and the vehicle is driven out of the parking space based on a reverse driving method.
5. The autonomous parking path reverse generation planning method according to claim 1, characterized in that:
the step 4 specifically comprises: and generating a plurality of parking prediction paths by connecting the generated paths to the target tree from the current position, namely the current pose, of the vehicle and taking the target tree end point as an end point of path generation.
6. The autonomous parking path reverse generation planning method according to claim 5, characterized in that:
when a path is generated, the end point of a target tree is taken as a sampling point, a cubic Hermite spline curve generation method with the minimum curvature change rate is used for generating a cluster of smooth curves connecting the current vehicle position to the end point of the target tree, and a plurality of smooth curves from the current pose S (X)s,Ys,θs) To the final pose E (X)e,Ye,θe) Each path representing one of the possible executions of the vehicleStatus.
7. The autonomous parking path reverse generation planning method according to claim 1, characterized in that:
the step 5 specifically comprises: for part of paths, because the curvature of the paths is too large, the vehicle kinematic model constraint is not satisfied, obstacles exist on part of tracks, the path length is used as evaluation basis according to the distance from the obstacles, the smoothness degree and the evaluation basis, the evaluation basis is subjected to weighted calculation and combined with the cost function CfAnd selecting the optimal path, namely the final parking path.
8. The autonomous parking path reverse generation planning method according to claim 1, characterized in that:
the method for generating the target path tree in step 3 correspondingly comprises the following steps: a parallel parking space target tree generation method, a vertical parking space target tree generation method and an inclined parking space target tree generation method.
9. The autonomous parking path reverse generation planning method according to claim 8, characterized in that:
the parallel parking space target tree generation method, the vertical parking space target tree generation method and the inclined parking space target tree generation method specifically comprise the following steps:
assuming that a vehicle is parked in a parking space, driving the vehicle out of the parking space by various predefined paths, wherein the point E is a starting point of driving the vehicle out of the parking space, the point E1 is a first steering point, and G1-Gi are reversing points; calculating by using E1 points as starting points and different turning radii R to generate a plurality of paths E1Gi, and obtaining a plurality of reversing paths from G1 to Gi by adopting a reverse planning method, namely different paths for reversing vehicles to enter a parking space;
E1Gi[k].x＝E1.x+(R*sin(θ))
E1Gi[k].y＝E1.y+(R-R*cos(θ))
in the formula, R is the turning radius of the vehicle corresponding to different steering wheel turning angles, theta represents the included angle between a road point and a road starting point on an arc path in the generation process of the road point, k represents a plurality of paths, namely the kth path corresponding to the target tree, x and y respectively represent the horizontal and vertical coordinates of the road point relative to the position of the vehicle, the plurality of paths are generated by utilizing the formula, and the plurality of paths form a group of predicted paths for the vehicle to select to run, namely the target tree, and are a group of arcs with different radii.
10. The autonomous parking path reverse generation planning method according to claim 1, characterized in that:
the optimal path comprises two parts: a front-end path from a portion of the sampling method and a back-end path from a portion of the target tree.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010098379.XA CN111307152B (en) | 2020-02-18 | 2020-02-18 | Reverse generation planning method for autonomous parking path |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010098379.XA CN111307152B (en) | 2020-02-18 | 2020-02-18 | Reverse generation planning method for autonomous parking path |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111307152A true CN111307152A (en) | 2020-06-19 |
| CN111307152B CN111307152B (en) | 2022-02-11 |
Family
ID=71158285
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010098379.XA Active CN111307152B (en) | 2020-02-18 | 2020-02-18 | Reverse generation planning method for autonomous parking path |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111307152B (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111923902A (en) * | 2020-08-10 | 2020-11-13 | 华人运通(上海)自动驾驶科技有限公司 | Parking control method and device, electronic equipment and storage medium |
| CN111959498A (en) * | 2020-07-14 | 2020-11-20 | 重庆智行者信息科技有限公司 | Vertical parking method and device for automatically driving vehicle and vehicle |
| CN112477850A (en) * | 2020-11-27 | 2021-03-12 | 北京罗克维尔斯科技有限公司 | Parking path planning method and device, vehicle-mounted equipment and storage medium |
| CN112509375A (en) * | 2020-10-20 | 2021-03-16 | 东风汽车集团有限公司 | Parking dynamic display method and system |
| CN112606830A (en) * | 2020-12-29 | 2021-04-06 | 吉林大学 | Two-section type autonomous parking path planning method based on mixed A-star algorithm |
| CN112677959A (en) * | 2020-12-23 | 2021-04-20 | 广州小鹏自动驾驶科技有限公司 | Parking method and device |
| CN113483775A (en) * | 2021-06-30 | 2021-10-08 | 上海商汤临港智能科技有限公司 | Path prediction method and device, electronic equipment and computer readable storage medium |
| WO2022007227A1 (en) * | 2020-07-10 | 2022-01-13 | 广东小鹏汽车科技有限公司 | Automatic parking method and vehicle |
| CN114255594A (en) * | 2021-12-28 | 2022-03-29 | 吉林大学 | Autonomous passenger-riding parking motion planning and motion control method |
| CN114312757A (en) * | 2021-12-22 | 2022-04-12 | 华人运通(上海)自动驾驶科技有限公司 | Parking planning method based on four-wheel steering and vehicle |
| WO2022142592A1 (en) * | 2020-12-31 | 2022-07-07 | 华为技术有限公司 | Front-first parking method, device and system |
| WO2022222401A1 (en) * | 2021-04-21 | 2022-10-27 | 阿波罗智联(北京)科技有限公司 | Valet parking method and apparatus, and device and autonomous driving vehicle |
| CN116605211A (en) * | 2023-07-19 | 2023-08-18 | 广汽埃安新能源汽车股份有限公司 | Parking path planning method and device, electronic equipment and storage medium |
| CN117734676A (en) * | 2024-02-19 | 2024-03-22 | 知行汽车科技(苏州)股份有限公司 | Automatic parking method, device, equipment and storage medium |
| CN117734676B (en) * | 2024-02-19 | 2024-05-03 | 知行汽车科技(苏州)股份有限公司 | Automatic parking method, device, equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105035075A (en) * | 2015-06-24 | 2015-11-11 | 合肥中科自动控制系统有限公司 | Path planning method for autonomous parallel parking |
| CN105857306A (en) * | 2016-04-14 | 2016-08-17 | 中国科学院合肥物质科学研究院 | Vehicle autonomous parking path programming method used for multiple parking scenes |
| CN106515722A (en) * | 2016-11-08 | 2017-03-22 | 西华大学 | Vertical parking track planning method |
| US20170220045A1 (en) * | 2012-09-28 | 2017-08-03 | Waymo Llc | Methods and Systems for Transportation to Destinations by a Self-Driving Vehicle |
| US20190001967A1 (en) * | 2017-06-30 | 2019-01-03 | MAGNETI MARELLI S.p.A. | Path planning method for computing optimal parking maneuvers for road vehicles and corresponding system |
| CN109927716A (en) * | 2019-03-11 | 2019-06-25 | 武汉环宇智行科技有限公司 | Autonomous method of vertically parking based on high-precision map |
-
2020
- 2020-02-18 CN CN202010098379.XA patent/CN111307152B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170220045A1 (en) * | 2012-09-28 | 2017-08-03 | Waymo Llc | Methods and Systems for Transportation to Destinations by a Self-Driving Vehicle |
| CN105035075A (en) * | 2015-06-24 | 2015-11-11 | 合肥中科自动控制系统有限公司 | Path planning method for autonomous parallel parking |
| CN105857306A (en) * | 2016-04-14 | 2016-08-17 | 中国科学院合肥物质科学研究院 | Vehicle autonomous parking path programming method used for multiple parking scenes |
| CN106515722A (en) * | 2016-11-08 | 2017-03-22 | 西华大学 | Vertical parking track planning method |
| US20190001967A1 (en) * | 2017-06-30 | 2019-01-03 | MAGNETI MARELLI S.p.A. | Path planning method for computing optimal parking maneuvers for road vehicles and corresponding system |
| CN109927716A (en) * | 2019-03-11 | 2019-06-25 | 武汉环宇智行科技有限公司 | Autonomous method of vertically parking based on high-precision map |
Non-Patent Citations (3)
| Title |
|---|
| MINGBO DU.ETC: "An Improved RRT-based Motion Planner for Autonomous Vehicle in", 《2014 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS & AUTOMATION (ICRA)》 * |
| 杨妮娜，等: "平行泊车的路径规划方法及其仿真研究", 《电子测量技术》 * |
| 王道斌，等: "两种自主泊车路径规划方法的对比研究", 《电子测量技术》 * |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022007227A1 (en) * | 2020-07-10 | 2022-01-13 | 广东小鹏汽车科技有限公司 | Automatic parking method and vehicle |
| CN111959498A (en) * | 2020-07-14 | 2020-11-20 | 重庆智行者信息科技有限公司 | Vertical parking method and device for automatically driving vehicle and vehicle |
| CN111923902A (en) * | 2020-08-10 | 2020-11-13 | 华人运通(上海)自动驾驶科技有限公司 | Parking control method and device, electronic equipment and storage medium |
| CN111923902B (en) * | 2020-08-10 | 2022-03-01 | 华人运通(上海)自动驾驶科技有限公司 | Parking control method and device, electronic equipment and storage medium |
| CN112509375B (en) * | 2020-10-20 | 2022-03-08 | 东风汽车集团有限公司 | Parking dynamic display method and system |
| CN112509375A (en) * | 2020-10-20 | 2021-03-16 | 东风汽车集团有限公司 | Parking dynamic display method and system |
| CN112477850A (en) * | 2020-11-27 | 2021-03-12 | 北京罗克维尔斯科技有限公司 | Parking path planning method and device, vehicle-mounted equipment and storage medium |
| CN112677959A (en) * | 2020-12-23 | 2021-04-20 | 广州小鹏自动驾驶科技有限公司 | Parking method and device |
| CN112606830A (en) * | 2020-12-29 | 2021-04-06 | 吉林大学 | Two-section type autonomous parking path planning method based on mixed A-star algorithm |
| CN112606830B (en) * | 2020-12-29 | 2023-12-29 | 吉林大学 | Two-section type autonomous parking path planning method based on mixed A-algorithm |
| WO2022142592A1 (en) * | 2020-12-31 | 2022-07-07 | 华为技术有限公司 | Front-first parking method, device and system |
| WO2022222401A1 (en) * | 2021-04-21 | 2022-10-27 | 阿波罗智联(北京)科技有限公司 | Valet parking method and apparatus, and device and autonomous driving vehicle |
| CN113483775A (en) * | 2021-06-30 | 2021-10-08 | 上海商汤临港智能科技有限公司 | Path prediction method and device, electronic equipment and computer readable storage medium |
| CN114312757A (en) * | 2021-12-22 | 2022-04-12 | 华人运通(上海)自动驾驶科技有限公司 | Parking planning method based on four-wheel steering and vehicle |
| CN114255594A (en) * | 2021-12-28 | 2022-03-29 | 吉林大学 | Autonomous passenger-riding parking motion planning and motion control method |
| CN114255594B (en) * | 2021-12-28 | 2024-03-15 | 吉林大学 | Autonomous passenger parking motion planning and motion control method |
| CN116605211B (en) * | 2023-07-19 | 2023-09-26 | 广汽埃安新能源汽车股份有限公司 | Parking path planning method and device, electronic equipment and storage medium |
| CN116605211A (en) * | 2023-07-19 | 2023-08-18 | 广汽埃安新能源汽车股份有限公司 | Parking path planning method and device, electronic equipment and storage medium |
| CN117734676A (en) * | 2024-02-19 | 2024-03-22 | 知行汽车科技(苏州)股份有限公司 | Automatic parking method, device, equipment and storage medium |
| CN117734676B (en) * | 2024-02-19 | 2024-05-03 | 知行汽车科技(苏州)股份有限公司 | Automatic parking method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111307152B (en) | 2022-02-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111307152B (en) | Reverse generation planning method for autonomous parking path | |
| Li et al. | Time-optimal maneuver planning in automatic parallel parking using a simultaneous dynamic optimization approach | |
| CN113306549B (en) | Automatic parking trajectory planning algorithm | |
| JP3432784B2 (en) | How to record track data and sensor data of a moving vehicle | |
| Laugier et al. | Sensor-based control architecture for a car-like vehicle | |
| CN107901917B (en) | A kind of automatic driving vehicle Trajectory Tracking Control method based on sliding coupling estimation of trackslipping | |
| Sungwoo et al. | Easy path planning and robust control for automatic parallel parking | |
| Hsieh et al. | A parking algorithm for an autonomous vehicle | |
| CN106515722A (en) | Vertical parking track planning method | |
| CN109927716A (en) | Autonomous method of vertically parking based on high-precision map | |
| CN111152784B (en) | Intelligent passenger-riding parking local path planning method | |
| CN113895463B (en) | Path planning method suitable for turning around of automatic driving vehicle | |
| CN112606830A (en) | Two-section type autonomous parking path planning method based on mixed A-star algorithm | |
| CN111547045B (en) | Automatic parking method and device for vertical parking spaces | |
| CN102658819A (en) | Automobile automatic parking method based on humanoid intelligent control | |
| CN111959498A (en) | Vertical parking method and device for automatically driving vehicle and vehicle | |
| CN110509923A (en) | Automatic Pilot paths planning method, system, computer readable storage medium and vehicle | |
| CN116185014A (en) | Intelligent vehicle global optimal track planning method and system based on dynamic planning | |
| CN113589685B (en) | Vehicle moving robot control system and method based on deep neural network | |
| CN113885525A (en) | Path planning method and system for automatically driving vehicle to get rid of trouble, vehicle and storage medium | |
| CN116465427B (en) | Intelligent vehicle lane changing obstacle avoidance path planning method based on space-time risk quantification | |
| Huy et al. | A practical and optimal path planning for autonomous parking using fast marching algorithm and support vector machine | |
| CN113741463B (en) | Fixed-point and directional parking control method and system for unmanned ground maneuvering platform | |
| US20230330853A1 (en) | System and Method for Controlling a Motion of a Robot | |
| CN112706770B (en) | Vehicle entry control system and method considering steer-by-wire delay |
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 |