CN107560615B

CN107560615B - Parking path planning method for parking lot automatic driving system

Info

Publication number: CN107560615B
Application number: CN201710684053.3A
Authority: CN
Inventors: 戴一凡; 卢贤票; 齐麟; 徐巍; 代爽; 曾勇
Original assignee: Suzhou Automotive Research Institute of Tsinghua University
Current assignee: Tsinghua University; Suzhou Automotive Research Institute of Tsinghua University
Priority date: 2017-08-11
Filing date: 2017-08-11
Publication date: 2021-04-23
Anticipated expiration: 2037-08-11
Also published as: CN107560615A

Abstract

The invention discloses a parking path planning method for an automatic driving system of a parking lot, which comprises the following steps: performing mathematical modeling on the parking lot according to road network information in the parking lot; selecting corresponding nodes at a road intersection of a parking lot and each parking space, numbering the nodes, and establishing a corresponding matrix relation; reading the line information in the line information table, and taking the distance between the starting point and the end point as a weight; and constructing a Dijkstra map by using the node map of the road network, and calculating an optimal path according to a Dijkstra algorithm. The optimal path from the starting point of the parking lot to the parking space can be rapidly planned.

Description

Parking path planning method for parking lot automatic driving system

Technical Field

The invention belongs to the technical field of automatic driving decision of automobiles, and particularly relates to a parking path planning method for an automatic driving system of a parking lot.

Background

As automotive autopilot technology continues to develop, the use of autopilot in parking lots has also begun to gain widespread attention. In the conventional parking lot at present, the most troubling drivers are two problems: firstly, the difficulty of finding an empty parking space and parking a narrow parking space; and secondly, the parking lot is possibly blocked when a plurality of vehicles find empty parking spaces.

The automatic driving system for the parking lot not only can improve the convenience of the parking process, but also can effectively relieve the problem of parking lot blockage caused by finding parking spaces. The parking lot automatic driving system searches a parking space using an ultrasonic sensor and an intelligent wheel sensor, generates a parking path to a parking position, and controls driving. That is, the intelligent parking assist system allows a driver to automatically park a vehicle without steering wheel operation only by controlling forward and backward movements and brake operation.

Most of the existing path planning methods are applied to urban roads, and the field of automatic driving systems of parking lots does not relate to the problem of path planning. Therefore, the method for planning the path from the starting point of the parking lot to the parking space enables the field of automatic driving systems of the parking lot to be a problem to be solved urgently. The invention is achieved accordingly.

Disclosure of Invention

Aiming at the problems in the technical scheme, the invention provides a parking path planning method for an automatic driving system of a parking lot, which can rapidly plan an optimal path from a starting point of the parking lot to a parking space.

The technical scheme of the invention is as follows:

a parking path planning method for an automatic driving system of a parking lot comprises the following steps:

s01: performing mathematical modeling on the parking lot according to road network information in the parking lot;

s02: selecting corresponding nodes at a road intersection of a parking lot and each parking space, numbering the nodes, and establishing a corresponding matrix relation;

s03: reading the line information in the line information table, and taking the distance between the starting point and the end point as a weight;

s04: and constructing a Dijkstra map by using the node map of the road network, and calculating an optimal path according to a Dijkstra algorithm.

Preferably, the step S04 further includes smoothing the intersection to calculate a turning radius and obtain a turning trajectory.

Preferably, in step S01, the road network information includes at least a width direction of a parking lot road, a width and a length of a parking space, and a distribution status of the parking space.

Preferably, in step S02, an intersection of a center line of a forward road in front of each parking space and a center line of a parking space width is used as a parking space node; and if the curvature change of the road intersection does not exceed the set threshold, selecting the road intersection as a node, and if the curvature change is greater than the set threshold, selecting a curvature change point and a circular arc central point as the node.

Preferably, in step S04, the calculating an optimal path according to the dijkstra algorithm includes:

s11: a starting point set S, namely S = { v }, v is a starting point, the distance of v is 0, U comprises other node sets except v, namely U = { other nodes }, if v is adjacent to a node in U, then < U, v > has a weight, and if U is not an adjacent point of v, then < U, v > has a weight of infinity;

s12: selecting a node k with the minimum distance v from the U, and adding k into the S, wherein the selected distance is the length of the shortest path from v to k;

s13: modifying the distance of each node in the U by taking k as a newly considered intermediate point; if the distance from the starting point v to the node u through the node k is shorter than the original distance without the node k, modifying the distance value of the node u, wherein the modified distance value is the sum of the distance of the node k and the distance value from the node k to the node u;

s14: steps S12 and S13 are repeated until all nodes are contained in S.

Compared with the prior art, the invention has the advantages that:

the optimal path from the starting point of the parking lot to the parking space can be rapidly planned.

Drawings

The invention is further described with reference to the following figures and examples:

FIG. 1 is a flow chart of a parking path planning method for a parking lot autopilot system of the present invention;

FIG. 2 is a schematic of parking lot modeling;

FIG. 3 is a schematic diagram of selected nodes in a parking lot;

fig. 4 is a schematic diagram of a track at a road intersection after rounding.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail with reference to the accompanying drawings in conjunction with the following detailed description. It should be understood that the description is intended to be exemplary only, and is not intended to limit the scope of the present invention. Moreover, in the following description, descriptions of well-known structures and techniques are omitted so as to not unnecessarily obscure the concepts of the present invention.

Example (b):

the preferred embodiments of the present invention will be further described with reference to the accompanying drawings.

As shown in fig. 1, a parking path planning method for an automatic driving system of a parking lot includes the steps of:

and performing mathematical modeling on the parking lot according to the road network information in the parking lot. The width direction of the parking lot road, the width length of the parking spaces, and the distribution state of the parking spaces are obtained through the road network information of the parking lot. As shown in fig. 2, a parking lot is mathematically modeled in the example;

there are 31 parking spaces per row in the example, for a total of 93 parking spaces. Each parking space is 6 meters in length and 3 meters in width. Wherein the green rectangle represents the lawn and has a width of 1 meter and a length of 6 meters. The width of the road between two rows of parking spaces is 10 meters.

Selecting corresponding nodes at the intersection of the parking lot road and each parking space, numbering the nodes, and establishing a corresponding matrix relationship as shown in FIG. 3; and taking the intersection point of the center line of the forward road in front of each parking space and the center line of the width of the parking space as a parking space node, if the curvature change at the road intersection does not exceed a set threshold, selecting the road intersection point as the node, and if the curvature change is greater than the set threshold, selecting the curvature change point and the arc center point as the node.

Reading the line information in the line information table to select a starting point and an end point, and taking the distance as a weight; the actual distance between the connectable adjacent nodes is used as a weight, and the distance between the non-connectable nodes is set to a larger value. The destination position is the position information of an empty parking space obtained through the communication between the vehicle and the parking lot, and the position information of the empty parking space is read to find out the corresponding node number. The automatic driving automobile can run by setting the speed, the time is not required to be considered as the weight, only the distance is taken as the weight, and the calculation time is saved.

In the example, the leftmost point in the middle row is used as a starting point v, and an empty parking space is randomly selected as an end point. The forward distance of the adjacent nodes is used as a weight value, and the weight value between the node which cannot be directly reached and the reverse node is set to be a larger value, so that the calculation time is reduced, and the method has great significance for automatically driving the automobile.

Constructing Dijkstra maps from node maps of the road network;

the optimal path is calculated according to the Dijkstra algorithm, and the method comprises the following steps:

a. initially in the example, S contains only the set of starting points, i.e., S ═ v, where v is 0 in distance. U contains the set of other nodes except v, i.e. U = { the rest nodes }, if v is adjacent to the node in U, then < U, v > normally has weight, if U is not the adjacent point of v, then < U, v > weight is ∞.

b. And selecting a node k with the minimum distance v from the U, and adding k into S (the selected distance is the shortest path length from v to k).

c. Modifying the distance of each node in the U by taking k as a newly considered intermediate point; if the distance from the starting point v to the node u (passing through the node k) is shorter than the original distance (not passing through the node k, i.e., v reaches u directly), the distance value of the node u is modified, and the modified distance value is the distance of the node k plus the distance value from k to u.

d. Repeating steps b and c until all nodes are contained in S.

As shown in fig. 4, smoothing is performed on the intersection curve; the arc at the road intersection needs to be tangent to the straight line by processing the arc by using the arc, and the turning radius is half of the road width, so that the turning track can be obtained. So that the automatically driven vehicle can pass through the road intersection without being restricted by dynamics problems. In the example, the turning radius is 3.5 meters and the steering angle is 90 degrees. The method can ensure that the automatically driven vehicle can stably pass through the intersection, and the situation that the vehicle cannot pass through the intersection due to the picture of the line angle is avoided.

It is to be understood that the above-described embodiments of the present invention are merely illustrative of or explaining the principles of the invention and are not to be construed as limiting the invention. Therefore, any modification, equivalent replacement, improvement and the like made without departing from the spirit and scope of the present invention should be included in the protection scope of the present invention. Further, it is intended that the appended claims cover all such variations and modifications as fall within the scope and boundaries of the appended claims or the equivalents of such scope and boundaries.

Claims

1. A parking path planning method for an automatic driving system of a parking lot is characterized by comprising the following steps:

s02: selecting corresponding nodes at a road intersection of a parking lot and each parking space, numbering the nodes, and establishing a corresponding matrix relation; taking the intersection point of the center line of the forward road in front of each parking space and the center line of the width of the parking space as a parking space node; if the curvature change of the road intersection does not exceed a set threshold, selecting the road intersection as a node, and if the curvature change is greater than the set threshold, selecting a curvature change point and a circular arc central point as nodes;

s03: reading the line information in the line information table, taking the distance between the starting point and the end point as a weight, taking the forward distance of the adjacent node as a weight, and setting the weight between the node which cannot be directly reached and the reverse node as a larger value;

2. The parking path planning method for the parking lot automatic driving system according to claim 1, wherein the step S04 further comprises smoothing the intersection, calculating a turning radius, and obtaining a turning trajectory.

3. The parking path planning method for the automatic driving system of the parking lot according to claim 1, wherein in step S01, the road network information at least includes a width direction of a parking lot road, a width and a length of a parking space, and a distribution status of the parking space.

4. The parking path planning method for the parking lot automated driving system according to claim 1, wherein in the step S04, calculating the optimal path according to dijkstra' S algorithm includes:

s14: steps S12 and S13 are repeated until all nodes are contained in S.