CN112444263A - Global path planning method and device - Google Patents
Global path planning method and device Download PDFInfo
- Publication number
- CN112444263A CN112444263A CN202011003654.1A CN202011003654A CN112444263A CN 112444263 A CN112444263 A CN 112444263A CN 202011003654 A CN202011003654 A CN 202011003654A CN 112444263 A CN112444263 A CN 112444263A
- Authority
- CN
- China
- Prior art keywords
- point
- determining
- closest
- edge
- skeleton line
- 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
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
- G01C21/34—Route searching; Route guidance
- G01C21/3446—Details of route searching algorithms, e.g. Dijkstra, A*, arc-flags, using precalculated routes
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0214—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory in accordance with safety or protection criteria, e.g. avoiding hazardous areas
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
The invention provides a global path planning method, which comprises the following steps: acquiring a first skeleton line of a passable area in the unstructured road and determining a second skeleton line according to the first skeleton line; acquiring a right driving path of a passable area; determining a plurality of sidelines and a plurality of vertexes of the unstructured road according to the second skeleton line and the right driving path; the edge has a length attribute and a direction attribute; before global path planning, mapping the current position and the target point of the vehicle to a class structured map; determining a first nearest point set from the first position to the edge of the class structured map and a second nearest point set from the second position to the edge of the class structured map, and determining an evaluation function value from each first nearest point in the first nearest point set to each second nearest point in the second nearest point set; and determining a path from the first closest point with the minimum evaluation function value to the second closest point as an optimal path. Therefore, the efficiency of global path planning is improved, and the reasonability of the path search result is improved.
Description
Technical Field
The invention relates to the technical field of automatic driving, in particular to a global path planning method and a global path planning device.
Background
With the development of artificial intelligence technology and modern manufacturing industry, the automatic driving technology has gradually advanced into people's daily life and has profoundly changed people's mode of travel. The unmanned technology can be briefly divided into perception, prediction, positioning, decision, planning and control. Planning refers to path planning and is mainly divided into global path planning and local path planning. The goal of global path planning is to reasonably search an environmental area in a determined area, and generate an optimal collision-free path from the current position of a vehicle to any target position in the area after considering a safety threshold. Compared with a structured area, an unstructured area lacks information describing a topological relationship of a path, so that efficient global path planning in the unstructured area is more challenging.
The existing unmanned global path planning scheme is mainly divided into two types, namely a grid-based graph search algorithm and a random sampling-based search algorithm:
1. the grid-based graph search algorithm is represented by the a-x algorithm and its variants. The main idea is to grid the search area and search to the surrounding space by taking a grid as a sampling step length. Evaluating and screening each sampling point through a designed heuristic function, and continuously exploring after selecting the optimal sampling point until a target point is explored. After searching to the target point, the final search path is obtained according to the 'father node' set during searching.
2. The search algorithm based on random sampling is represented by the RRT algorithm. The main idea is that the starting point is used as a root node, a random expanded tree is generated by increasing leaf nodes through random sampling, and when the leaf nodes in the random tree contain a target point or enter a target area, a path from the starting point to the target point can be found in the random tree.
The rationality of the grid-based search algorithm to generate a path depends on the shape and accuracy of the grid. For an unstructured area with a large range and complex road conditions, different grid sizes and shapes need to be adopted according to the width and the shape of a passing area, so that the manufacturing cost of the map is increased, and the calculation amount is increased.
Although the map does not need to be rasterized in advance, the random sampling-based search algorithm is a pure random search algorithm and is insensitive to environment types, when a search space contains a large number of obstacles or narrow channel constraints, the convergence rate of the algorithm is low, the efficiency is greatly reduced, and a path is difficult to find in the environment with the narrow channel.
Disclosure of Invention
The embodiment of the invention aims to provide a global path planning method and a global path planning device, which are used for solving the problems that the rationality of a path generated by a grid-based search algorithm in the prior art depends on the shape and the precision of a grid, different grid sizes and shapes need to be adopted according to the width and the shape of a passing area, the manufacturing cost of a map is increased, the calculated amount is increased, and a random sampling-based search algorithm is not sensitive to the type of an environment.
In a first aspect, the present invention provides a global path planning method, where the global path planning method includes:
acquiring a first skeleton line of a passable area in an unstructured road; the unstructured road comprises a passable area and a non-passable area;
determining a second skeleton line according to the distance between each point in the first skeleton line and the boundary of the unstructured road;
acquiring a right driving path of the passable area;
determining a plurality of edges and a plurality of vertexes of the unstructured road according to the second skeleton line and the right driving path, wherein the edges and the vertexes form a class structured map; the edge lines have a length attribute and a direction attribute, and the intersection point of any two edge lines is the vertex of the unstructured road;
before global path planning, mapping the current position and the target point of the vehicle to the class structured map;
determining a first nearest point set from the first position to an edge of the class structured map and a second nearest point set from the second position to the edge of the class structured map according to a first position of a current position of a vehicle in the class structured map and a second position of a target point in the class structured map;
determining an evaluation function value from each first closest point in the first closest point set to each second closest point in the second closest point set according to the length from each first closest point in the first closest point set to the sampling point along the edge, the direction from each first closest point to the sampling point along the edge and the distance from the sampling point to each second closest point;
and determining a path from the first closest point with the minimum evaluation function value to the second closest point as an optimal path.
In one possible implementation manner, the acquiring the first skeleton line of the passable area in the unstructured road specifically includes:
rasterizing a passable area in the unstructured road, and dividing the passable area into a plurality of grids;
determining the grid value of each grid according to the distance from each grid to the impassable area;
and when any grid value represents that the grid value is far from different parts in the non-passable area, determining a first skeleton line according to the any grid value.
In a possible implementation manner, the acquiring the right-going travel path of the passable area specifically includes:
acquiring a grid adjacent to the impassable area;
and determining a right driving path according to the grid values of the grids adjacent to the impassable area.
In one possible implementation, the grid size is determined according to the width of the vehicle during the rasterization process;
and determining the distance from the right driving path to the boundary of the unstructured road according to the width and the minimum turning radius of the vehicle.
In a possible implementation manner, the determining, according to a distance between each point in the first skeleton line and a boundary of the unstructured road, a second skeleton line specifically includes:
and when the distance between the point in the first skeleton line and the boundary of the passable area is smaller than a preset distance threshold, deleting the point in the first skeleton line to obtain a second skeleton line.
In a possible implementation manner, the step of having the direction attribute by the edge specifically includes:
acquiring any point on any sideline on the right driving path;
judging whether a first side point and a second side point of any point are in the unstructured road; the first side point and the second side point are arranged on two sides of any one point;
when the first side point is not in the unstructured road, determining the direction attribute of any edge as a first direction, and when the second side point is not in the unstructured road, determining the direction attribute of any edge as a second direction; the first direction and the second direction are opposite to each other.
In a possible implementation manner, the searching for the first closest point set on the edge of the first location to class structured map and the second closest point set on the edge of the second location to class structured map specifically includes:
determining a starting circle by taking the first position as a circle center and a preset first radius as a radius;
determining a first closest point set in the starting circle, and taking each first closest point in the first closest point set as a starting point of global path planning;
determining a terminal point circle by taking the second position as a circle center and a preset second radius as a radius;
determining a second set of closest points in the circle of end points, and using each second closest point in the second set of closest points as an end point of a solicitation path plan.
In a possible implementation manner, the determining, according to a length of each first closest point in the first closest point set to the sampling point along the edge, a direction of each first closest point to the sampling point along the edge, and a distance from the sampling point to each second closest point, an evaluation value of each first closest point in the first closest point set to each second closest point in the second closest point set specifically includes:
determining the length of each starting point to the sampling point along the edge line;
determining the direction of each starting point to the sampling point along the edge;
determining a first product of a preset first weight value and the length;
determining a second product of a preset second weight value and the direction;
determining a sum of the first product and the second product as a first cost value;
determining the distance from the sampling point to a target point as a second generation value;
and determining an evaluation function value according to the first generation value and the second generation value.
In a second aspect, the present invention provides a global path planning apparatus, including:
an acquisition unit for acquiring a first skeleton line of a passable area in an unstructured road; the unstructured road comprises a passable area and a non-passable area;
a determining unit, configured to determine a second skeleton line according to a distance between each point in the first skeleton line and a boundary of the unstructured road;
the acquisition unit is further used for acquiring a right driving path of the passable area;
the determining unit is further configured to determine, according to the second skeleton line and the right driving route, a plurality of edges and a plurality of vertices of an unstructured road, where the plurality of edges and the plurality of vertices form a class structured map; the edge lines have a length attribute and a direction attribute, and the intersection point of any two edge lines is the vertex of the unstructured road;
the mapping unit is used for mapping the current position and the target point of the vehicle to the class structured map before global path planning is carried out;
the determining unit is further configured to determine, according to a first position of a current position of the vehicle in the class structured map and a second position of the target point in the class structured map, a first closest point set from the first position to an edge of the class structured map and a second closest point set from the second position to the edge of the class structured map;
the determining unit is further configured to determine an evaluation function value of each first closest point in the first closest point set to each second closest point in the second closest point set according to a length of each first closest point in the first closest point set to the sampling point along the edge, a direction of each first closest point to the sampling point along the edge, and a distance of the sampling point to each second closest point;
the determining unit is further configured to determine a path from the first closest point to the second closest point where the evaluation function value is minimum as an optimal path.
In a third aspect, the invention provides an apparatus comprising a memory for storing a program and a processor for performing the method of any of the first aspects.
In a fourth aspect, the present invention provides a computer program product comprising instructions which, when run on a computer, cause the computer to perform the method according to any one of the first aspect.
In a fifth aspect, the invention provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the method of any of the first aspects.
By applying the global path planning method and the global path planning device provided by the embodiment of the invention, the topological processing can be carried out on the unstructured road to obtain the class structured map, so that the efficiency of global path planning is improved by carrying out the global path planning on the class structured map.
Drawings
Fig. 1 is a schematic flow chart of a global path planning method according to an embodiment of the present invention;
FIG. 2 is a schematic view of an unstructured road according to an embodiment of the present invention;
fig. 3 is a schematic diagram of generating a class structured map of an unstructured road according to an embodiment of the present invention;
FIG. 4 is a block diagram of FIG. 3;
fig. 5 is a first skeleton line of a sub-structured road according to an embodiment of the present invention;
fig. 6 is a schematic diagram of a second skeleton line and a right driving path according to a first embodiment of the present invention;
FIG. 7 is a schematic diagram illustrating the direction attribute of an edge according to an embodiment of the present invention;
FIG. 8 is a schematic diagram of an unstructured road with directional attribute according to an embodiment of the present invention;
FIG. 9 is a schematic diagram of a first closest point set and a second closest point set according to an embodiment of the present invention;
fig. 10 is a schematic structural diagram of a global path planning apparatus according to a second embodiment of the present invention.
Detailed Description
The present application will be described in further detail with reference to the following drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the relevant invention and not restrictive of the invention. It should be further noted that, for the convenience of description, only the portions related to the related invention are shown in the drawings.
It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The present application will be described in detail below with reference to the embodiments with reference to the attached drawings.
Fig. 1 is a schematic flowchart of a global path planning method according to an embodiment of the present invention. The execution subject of the application is a terminal, a server or a processor with a computing function. The present application will be described by taking an example of applying the method to an unmanned Vehicle, and when the method is applied to an unmanned Vehicle, an execution subject of the method is an Automated Vehicle Control Unit (AVCU), that is, a central processing Unit of the unmanned Vehicle corresponds to a "brain" of the unmanned Vehicle. The method comprises the following steps:
Specifically, during driving, the autonomous vehicle loads a map of a region where the autonomous vehicle is driving, and the map includes a structured road and an unstructured road. The unstructured roads are opposite to the structured roads, the structured roads refer to highways with good structuralization, such as expressways, urban arterial roads and the like, the roads have clear road sign lines, the background environment of the roads is single, and the geometric characteristics of the roads are obvious. For the structured road, the topological relation among the paths can be easily acquired, and then the global path planning is carried out according to the topological relation.
The unstructured roads generally refer to roads with low structuralization degrees, such as city non-main roads, rural streets and the like, and the roads have no lane lines and clear road boundaries, and are affected by shadows, water marks and the like, so that the road areas and the non-road areas are difficult to distinguish. The unstructured roadway comprises a passable area and a non-passable area.
Referring to fig. 2, fig. 2 is a schematic diagram of an unstructured road according to an embodiment of the present invention, where a lower point in fig. 2 may be used as a current position of a vehicle, and an upper point may be used as a target point of the current vehicle. The dark area is a passable area, and the white area is a non-passable area.
For an unstructured road like that in fig. 2, a first skeleton line thereof may be determined through a certain process. Here, the processing is rasterization processing, and the step of determining the first skeleton line by the rasterization processing is as follows:
firstly, rasterizing a passable area in an unstructured road, and dividing the passable area into a plurality of grids; secondly, determining the grid value of each grid according to the distance from each grid to the impassable area; and finally, when any grid value represents different parts in the distance non-passable area, determining the first skeleton line according to any grid value.
How to determine the first skeleton line is described below with reference to fig. 3.
In fig. 3, a black area is an impassable area, a gray area is an impassable area, the impassable area is regarded as an obstacle, the boundary of the unstructured road in fig. 3 is regarded as a single obstacle, the impassable area is subjected to rasterization processing, each grid value represents the distance to the nearest obstacle, the grid value of the grid occupied by the obstacle is 0, the grid value of the grid adjacent to the obstacle is 1, the grid values sequentially extend outwards until the two different obstacles extend to the same grid, for example, the grid value is 4, the distances to the obstacles on the left side and the right side are both 4, the grid with the grid value of 4 represents the grids extending from the two different obstacles to the same grid, and at this time, the grid with the extension end forms the first skeleton line of the area.
Fig. 4 is a block diagram of fig. 3, and referring to fig. 4, the passable area may be divided into 5 blocks according to white lines, for the zone 1, the obstacle closest to the point of the zone 1 is the obstacle on the left side, and the obstacles on other sides are not close to the obstacle on the left side. Regarding the zone 2, the obstacle closest to the zone 2 is the upper obstacle, and the obstacles on the other sides are not close to the upper obstacle. Similarly, for zone 5, the obstacle closest to zone 5 is the middle obstacle, and the obstacles on the other sides are not close to the middle obstacle.
Therefore, the grid-vertex connection line of the grid value 4 is used to distinguish different blocks, and the grid-vertex connection line of the grid value 4 is used as a skeleton line because the upper point is closer to the upper obstacle and the lower point is closer to the left obstacle.
Similarly, the reason why the grid of the grid value 4 takes the center point is similar, and the center point is the same distance from the left side, the upper side, and the middle obstacle, and therefore serves as the boundary point.
Thus, according to fig. 3 and 4, the first skeleton line of the unstructured road shown in fig. 2 may be determined, the first skeleton line of the unstructured road shown in fig. 2 is shown in fig. 5, and fig. 5 is the first skeleton line of the sub-structured road provided in the first embodiment of the present invention.
The grid size during the grid processing is determined according to the width of the vehicle, for example, the grid size and the width of the vehicle are in a fixed proportion, for example, 1:1, or the grid size is determined according to the length and the width of the vehicle, for example, the grid size is determined after the length square of the vehicle plus the width square, and then the grid size is determined according to the length and the width of the vehicle, so that the speed of the grid processing is improved.
And step 120, determining a second skeleton line according to the distance between each point in the first skeleton line and the boundary of the unstructured road.
Specifically, for a first skeleton line, the first skeleton line is composed of a plurality of sidelines, and when the distance from a point on the sideline in the first skeleton line to the boundary of the passable area is smaller than a preset distance threshold, the sideline in the first skeleton line is deleted to obtain a second skeleton line.
For example, if a distance from one point to the boundary of the unstructured road in a certain edge line on the first skeleton line is smaller than a preset distance threshold, the edge line is removed.
And step 130, acquiring a right driving path of the passable area.
Specifically, in one example, the distance from the route when driving to the right to the boundary of the unstructured road may be determined according to the width and the minimum turning radius of the unmanned vehicle, and thus the right driving route may be determined according to the distance and the boundary of the unstructured road. The distance here may be the sum of half the width of the vehicle and the minimum turning radius.
In another example, a grid adjacent to the impassable area may be obtained; and determining a right driving path according to the grid values of the grids adjacent to the impassable area. For example, referring to fig. 3, if the grid value adjacent to the impassable area is 1, the center points of the grids with the grid value of 1 are taken to connect the grid values according to the grid value of 1, so as to determine the right driving path.
Specifically, a class structured map is actually a similar structured map, in which an unstructured road is treated as containing a plurality of edges and a plurality of vertices.
Referring to fig. 6, fig. 6 is a schematic diagram of a second skeleton line and a right driving route according to a first embodiment of the present invention, a solid line in fig. 6 is the second skeleton line, a dotted line is the right driving route, the second skeleton line and the right driving route in fig. 6 are called edges, an intersection point of each edge is called a vertex, a border is formed between two vertices, the border has not only a length attribute but also a direction attribute, the second skeleton line in the border is two-way passable, and the right driving route follows the principle of driving on the right.
The following description is made on how to determine the directional property of an edge:
specifically, any point on any sideline on the right driving path is obtained firstly; then judging whether a first side point and a second side point of any point are in the unstructured road; the first side point and the second side point are arranged on two sides of any point; finally, when the first side point is not in the unstructured road, determining the direction attribute of any side line as a first direction, and when the second side point is not in the unstructured road, determining the direction attribute of any side line as a second direction; the first direction and the second direction are opposite to each other.
Referring to FIG. 7, taking the side line AB as an example, if the distance from AB to the boundary of the unstructured road is l, given an initial direction of AB as indicated by the arrow in the figure, i.e. pointing from A to B, taking any point C on the side of AB, taking points C on the left and right sides of point C along the normal direction of point ClAnd Cr。ClAnd CrThe following formula is satisfied:
CCr＝CCris l + epsilon formula (1)
Wherein, CCrFrom point C to point CrLength of line segment of CClFrom point C to point ClThe length of the line segment, epsilon, is a sufficiently small value that can be considered infinitesimal small.
Can be detected by ClAnd CrAnd determining the direction attribute of the edge line AB if the road is in the unstructured road. If C is presentrNot in the unstructured road, the arrow in fig. 8 is the direction of the AB edge line, if C islAnd not in the unstructured road, the opposite direction of the arrow in fig. 8 is the direction of the AB edge line.
Referring to fig. 8, the class structured map of the unstructured road includes a length attribute, a direction attribute and a plurality of vertices of an edge. The direction attribute can be visually represented by an arrow, and the length attribute can be stored in a database, for example, an edge has an edge ID, and the corresponding edge ID corresponds to the length attribute of the edge. The edge ID and length attributes may be stored in a database in a tabular manner. If the ID of the edge AB is 01 and the corresponding length is 5m, 01 and 5 meters are stored in the database in a table.
Specifically, during the driving process of the automatic driving vehicle, the current position information of the vehicle may be obtained through a sensor mounted on the vehicle, such as an Inertial Measurement Unit (IMU), a Global Positioning System (GPS), and the like.
When the server receives the taxi calling service information of the user terminal, for example, the user terminal receives information of a departure place, a destination, departure time and the like input by the user, and after receiving the information of the departure place and the destination sent by the user terminal, the server screens out the target vehicle and sends the information of the departure place, the destination, the departure time, the number of departure persons and the like to the target vehicle.
For the target vehicle, when the vehicle is in the process of traveling, global path planning is required so as to travel from a departure place to a destination, and before the global path planning is performed, if the current position information and the destination, that is, the target point, are both in the obtained class structured map having a plurality of edges and a plurality of vertices, the current position of the vehicle and the target point can be mapped in the class structured map.
The current position of the vehicle can be longitude and latitude data, each point in the class structured map has position information, and the current longitude and latitude data of the vehicle can be mapped into the structured map, so that the current position and the target point of the vehicle in the structured map can be determined.
And step 160, determining a first closest point set from the first position to the boundary of the class structured map and a second closest point set from the second position to the boundary of the class structured map according to the first position of the current position of the vehicle in the class structured map and the second position of the target point in the class structured map.
Specifically, step 160 includes: firstly, determining an initial circle by taking a first position as a circle center and a preset first radius as a radius; secondly, determining a first closest point set in the starting circle, and taking each first closest point in the first closest point set as a starting point of global path planning; thirdly, determining a terminal point circle by taking the second position as a circle center and a preset second radius as a radius; finally, a second set of closest points in the circle of end points is determined, and each second closest point in the second set of closest points is used as an end point of the recruitment path plan.
The preset first radius and the preset second radius may be the same or different. The preset first radius may be determined according to the width of the vehicle and a fixed value, where the fixed value is related to the number of edge lines around the first position and the distance between adjacent edge lines around the first position, for example, referring to fig. 9, the first position is a small rectangular frame below, three edge lines are located at the first position, and after the circle is made by the preset first radius, all three edge lines are ensured to be located in the starting circle, and the obtained first closest point set is represented by three small pentagons around the small rectangular frame below.
Accordingly, the final circle is set in a similar manner to the initial circle, and will not be described herein again. The second position is a small rectangle frame above, three sidelines are arranged at the first position, and after a circle is made by a preset second radius, the obtained second nearest point set is represented by three small pentagons around the small rectangle frame above.
Specifically, step 170 includes: determining the length of each starting point to the sampling point along the edge line; determining the direction of each starting point to the sampling point along the edge; determining a first product of a preset first weight value and a length; determining a second product of a preset second weight value and the direction; determining a sum of the first product and the second product as a first cost value; determining the distance from the sampling point to the target point as a second generation value; and determining an evaluation function value according to the first generation value and the second generation value.
The sampling point is a point at each path search, and the path search can be realized by adopting a path search algorithm. For example, the a-algorithm in the heuristic path search algorithm, which is not limited in this application.
Specifically, in one example, when a heuristic path search algorithm is adopted, evaluation may be performed through an evaluation function corresponding to the a-algorithm, where the evaluation function is shown in formula (2):
formula (2) g + h
The value of g is the first generation value of the search and represents the cost of searching from the starting point of the search to the current sampling point. The h value is a second cost value of the current sample point to the target point and can be represented by a distance from the sample point to the target point, where the distance includes, but is not limited to, Euclidean distance and Manhattan distance.
The first generation value can be calculated through formula (3).
g ═ α · l + β · dir formula (3)
Wherein l represents the length of the edge from the starting point to the current sampling point, the edge is divided by the top points, the edge between the two top points is used as the edge, dir is the direction from the starting point to the sampling point along the edge, wherein the forward direction is represented as 0, the backward direction is represented as 1, alpha is a first weight coefficient and represents the importance of the path length to the whole path, and beta is a second weight coefficient and represents the importance of the direction to the whole path.
Alpha and beta can be set according to requirements, for example, if the vehicle is not required to move forwards and backwards, the meaning of beta is not large if the target point can be reached at the shortest time, at the moment, the value of beta can be set to be smaller than the value of alpha, and the value of beta is infinitely small, and the specific value of beta needs to be determined through a plurality of times of analog calculation.
And step 180, determining a path from the first closest point with the minimum evaluation function value to the second closest point as an optimal path.
Subsequently, an evaluation function value can be determined according to the first generation value and the second generation value, and a corresponding optimal path with the minimum evaluation function value can be determined according to the evaluation function value.
For example, if the starting point has 1, 2, 3, the sampling point is a, and the end point is 4,5,6, then the starting point to the sampling point, the first generation value of the three paths of 1-a, 2-a, 3-a is calculated, the sampling point to the end point, the second generation value of the three paths of a-4, a-5, and a-6 is calculated, then the values are added arbitrarily, i.e. 1-a + a-4, 1-a + a-5, 1-a + a-6, 2-a + a-4, 2-a + a-5, 2-a + a-6, 3-a + a-4, 3-a + a-5, 3-a + a-6, and these 9 values are calculated to determine that the evaluation function value is the smallest as a local path, e.g. 1-a + a-4 is the smallest, the optimal path from the starting point 1 to the sampling point a and then to the end point 4 can be determined.
By applying the global path planning method provided by the embodiment of the invention, the topological processing can be carried out on the unstructured road to obtain the class structured map, so that the global path planning efficiency is improved through the class structured map, and meanwhile, the length attribute and the direction attribute of the sideline in the class structured map can be combined when the path evaluation is carried out, so that the reasonability of the path search result is improved.
Fig. 10 is a schematic structural diagram of a global path planning device according to a second embodiment of the present invention, and as shown in fig. 10, the global path planning device is applied to the global path planning method according to the first embodiment, and includes: an obtaining unit 210, a determining unit 220 and a mapping unit 230.
The obtaining unit 210 is configured to obtain a first skeleton line of a passable area in an unstructured road; the unstructured road comprises a passable area and a non-passable area;
the determining unit 220 is configured to determine a second skeleton line according to a distance between each point in the first skeleton line and a boundary of the unstructured road;
the obtaining unit 210 is further configured to obtain a right driving route of the passable area;
the determining unit 220 is further configured to determine, according to the second skeleton line and the right driving route, a plurality of edges and a plurality of vertices of the unstructured road, where the plurality of edges and the plurality of vertices form a similar structured map; the edge lines have a length attribute and a direction attribute, and the intersection point of any two edge lines is the vertex of the unstructured road;
the mapping unit 230 is configured to map the current position and the target point of the vehicle onto the class structured map before performing global path planning;
the determining unit 220 is further configured to determine, according to a first location of the current location of the vehicle in the class structured map and a second location of the target point in the class structured map, a first closest point set from the first location to an edge of the class structured map and a second closest point set from the second location to the edge of the class structured map;
the determining unit 220 is further configured to determine an evaluation function value from each first closest point in the first closest point set to each second closest point in the second closest point set according to a length from each first closest point in the first closest point set to the sampling point along the edge, a direction from each first closest point to the sampling point along the edge, and a distance from the sampling point to each second closest point;
the determining unit 220 is further configured to determine a path from the first closest point to the second closest point where the evaluation function value is minimum as an optimal path.
Further, the obtaining unit 210 is specifically configured to: rasterizing a passable area in the unstructured road, and dividing the passable area into a plurality of grids; determining the grid value of each grid according to the distance from each grid to the impassable area; when any grid value represents different parts in the distance-impassable area, the first skeleton line is determined according to any grid value.
Further, the obtaining unit 210 is specifically configured to: acquiring a grid adjacent to the impassable area; and determining a right driving path according to the grid values of the grids adjacent to the impassable area.
Determining the size of a grid according to the width of a vehicle during rasterization; and determining the distance from the right driving path to the boundary of the unstructured road according to the width and the minimum turning radius of the vehicle.
Further, the determining unit 220 is specifically configured to:
and when the distance between the point in the first skeleton line and the boundary of the passable area is smaller than a preset distance threshold, deleting the point in the first skeleton line to obtain a second skeleton line.
The edge having the directional attribute specifically includes:
acquiring any point on any sideline on the right driving path;
judging whether a first side point and a second side point of any point are in the unstructured road; the first side point and the second side point are arranged on two sides of any point;
when the first side point is not in the unstructured road, determining the direction attribute of any side line as a first direction, and when the second side point is not in the unstructured road, determining the direction attribute of any side line as a second direction; the first direction and the second direction are opposite to each other.
Further, the determining unit 220 is specifically configured to:
determining a starting circle by taking the first position as a circle center and a preset first radius as a radius;
determining a first nearest point set in the starting circle, and taking each first nearest point in the first nearest point set as a starting point of global path planning;
determining a final point circle by taking the second position as a circle center and a preset second radius as a radius;
a second set of closest points in the circle of end points is determined and each second closest point in the second set of closest points is used as an end point for the solicitation path plan.
Further, the determining unit 220 is specifically configured to:
determining the length of each starting point to the sampling point along the edge line; determining the direction of each starting point to the sampling point along the edge; determining a first product of a preset first weight value and a length; determining a second product of a preset second weight value and the direction; determining a sum of the first product and the second product as a first cost value; determining the distance from the sampling point to the target point as a second generation value; and determining an evaluation function value according to the first generation value and the second generation value.
By applying the global path planning device provided by the embodiment of the invention, the topological processing can be carried out on the unstructured road to obtain the class structured map, so that the global path planning efficiency is improved through the class structured map, and meanwhile, when the path evaluation is carried out, the length attribute and the direction attribute of the sideline in the class structured map can be combined, so that the reasonability of the path search result is improved.
The third embodiment of the invention provides equipment, which comprises a memory and a processor, wherein the memory is used for storing programs, and the memory can be connected with the processor through a bus. The memory may be a non-volatile memory such as a hard disk drive and a flash memory, in which a software program and a device driver are stored. The software program is capable of performing various functions of the above-described methods provided by embodiments of the present invention; the device drivers may be network and interface drivers. The processor is used for executing a software program, and the software program can realize the method provided by the first embodiment of the invention when being executed.
A fourth embodiment of the present invention provides a computer program product including instructions, which, when the computer program product runs on a computer, causes the computer to execute the method provided in the first embodiment of the present invention.
The fifth embodiment of the present invention provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the method provided in the first embodiment of the present invention is implemented.
Those of skill would further appreciate that the various illustrative components and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied in hardware, a software module executed by a processor, or a combination of the two. A software module may reside in Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The above embodiments are provided to further explain the objects, technical solutions and advantages of the present invention in detail, it should be understood that the above embodiments are merely exemplary embodiments of the present invention and are not intended to limit the scope of the present invention, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (10)
1. A global path planning method is characterized by comprising the following steps:
acquiring a first skeleton line of a passable area in an unstructured road; the unstructured road comprises a passable area and a non-passable area;
determining a second skeleton line according to the distance between each point in the first skeleton line and the boundary of the unstructured road;
acquiring a right driving path of the passable area;
determining a plurality of edges and a plurality of vertexes of the unstructured road according to the second skeleton line and the right driving path, wherein the edges and the vertexes form a class structured map; the edge lines have a length attribute and a direction attribute, and the intersection point of any two edge lines is the vertex of the unstructured road;
before global path planning, mapping the current position and the target point of the vehicle to the class structured map;
determining a first nearest point set from the first position to an edge of the class structured map and a second nearest point set from the second position to the edge of the class structured map according to a first position of a current position of a vehicle in the class structured map and a second position of a target point in the class structured map;
determining an evaluation function value from each first closest point in the first closest point set to each second closest point in the second closest point set according to the length from each first closest point in the first closest point set to the sampling point along the edge, the direction from each first closest point to the sampling point along the edge and the distance from the sampling point to each second closest point;
and determining a path from the first closest point with the minimum evaluation function value to the second closest point as an optimal path.
2. The method according to claim 1, wherein the obtaining of the first skeleton line of the passable area in the unstructured road comprises:
rasterizing a passable area in the unstructured road, and dividing the passable area into a plurality of grids;
determining the grid value of each grid according to the distance from each grid to the impassable area;
and when any grid value represents that the grid value is far from different parts in the non-passable area, determining a first skeleton line according to the any grid value.
3. The method according to claim 2, wherein the obtaining of the right-hand travel path of the passable area comprises:
acquiring a grid adjacent to the impassable area;
and determining a right driving path according to the grid values of the grids adjacent to the impassable area.
4. The method of claim 2, wherein the rasterization process determines a grid size based on the width of the vehicle;
and determining the distance from the right driving path to the boundary of the unstructured road according to the width and the minimum turning radius of the vehicle.
5. The method according to claim 1, wherein the determining a second skeleton line according to the distance between each point in the first skeleton line and the boundary of the unstructured road specifically comprises:
and when the distance between the point in the first skeleton line and the boundary of the passable area is smaller than a preset distance threshold, deleting the point in the first skeleton line to obtain a second skeleton line.
6. The method of claim 1, wherein the edge having the directional property specifically comprises:
acquiring any point on any sideline on the right driving path;
judging whether a first side point and a second side point of any point are in the unstructured road; the first side point and the second side point are arranged on two sides of any one point;
when the first side point is not in the unstructured road, determining the direction attribute of any edge as a first direction, and when the second side point is not in the unstructured road, determining the direction attribute of any edge as a second direction; the first direction and the second direction are opposite to each other.
7. The method of claim 1, wherein searching for the first set of closest points on the edge of the first location to class structured map and the second set of closest points on the edge of the second location to class structured map specifically comprises:
determining a starting circle by taking the first position as a circle center and a preset first radius as a radius;
determining a first closest point set in the starting circle, and taking each first closest point in the first closest point set as a starting point of global path planning;
determining a terminal point circle by taking the second position as a circle center and a preset second radius as a radius;
determining a second set of closest points in the circle of end points, and using each second closest point in the second set of closest points as an end point of a solicitation path plan.
8. The method according to claim 7, wherein the determining the evaluation value of each first closest point in the first set of closest points to each second closest point in the second set of closest points according to the length of each first closest point in the first set of closest points to the sampling point along the edge, the direction of each first closest point to the sampling point along the edge, and the distance of the sampling point to each second closest point specifically comprises:
determining the length of each starting point to the sampling point along the edge line;
determining the direction of each starting point to the sampling point along the edge;
determining a first product of a preset first weight value and the length;
determining a second product of a preset second weight value and the direction;
determining a sum of the first product and the second product as a first cost value;
determining the distance from the sampling point to a target point as a second generation value;
and determining an evaluation function value according to the first generation value and the second generation value.
9. A global path planning apparatus, comprising:
an acquisition unit for acquiring a first skeleton line of a passable area in an unstructured road; the unstructured road comprises a passable area and a non-passable area;
a determining unit, configured to determine a second skeleton line according to a distance between each point in the first skeleton line and a boundary of the unstructured road;
the acquisition unit is further used for acquiring a right driving path of the passable area;
the determining unit is further configured to determine, according to the second skeleton line and the right driving route, a plurality of edges and a plurality of vertices of an unstructured road, where the plurality of edges and the plurality of vertices form a class structured map; the edge lines have a length attribute and a direction attribute, and the intersection point of any two edge lines is the vertex of the unstructured road;
the mapping unit is used for mapping the current position and the target point of the vehicle to the class structured map before global path planning is carried out;
the determining unit is further configured to determine, according to a first position of a current position of the vehicle in the class structured map and a second position of the target point in the class structured map, a first closest point set from the first position to an edge of the class structured map and a second closest point set from the second position to the edge of the class structured map;
the determining unit is further configured to determine an evaluation function value of each first closest point in the first closest point set to each second closest point in the second closest point set according to a length of each first closest point in the first closest point set to the sampling point along the edge, a direction of each first closest point to the sampling point along the edge, and a distance of the sampling point to each second closest point;
the determining unit is further configured to determine a path from the first closest point to the second closest point where the evaluation function value is minimum as an optimal path.
10. A computer-readable storage medium, characterized in that a computer program is stored on the computer-readable storage medium, which computer program, when being executed by a processor, is adapted to carry out the method of any one of claims 1-8.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011003654.1A CN112444263B (en) | 2020-09-22 | 2020-09-22 | Global path planning method and device |
| PCT/CN2021/118566 WO2022063005A1 (en) | 2020-09-22 | 2021-09-15 | Global path planning method and apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011003654.1A CN112444263B (en) | 2020-09-22 | 2020-09-22 | Global path planning method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112444263A true CN112444263A (en) | 2021-03-05 |
| CN112444263B CN112444263B (en) | 2023-05-23 |
Family
ID=74736831
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011003654.1A Active CN112444263B (en) | 2020-09-22 | 2020-09-22 | Global path planning method and device |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN112444263B (en) |
| WO (1) | WO2022063005A1 (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113119995A (en) * | 2021-03-11 | 2021-07-16 | 京东鲲鹏(江苏)科技有限公司 | Path searching method and device, equipment and storage medium |
| CN113219973A (en) * | 2021-05-08 | 2021-08-06 | 浙江工业大学 | Efficient local path control method for mobile robot |
| CN113670323A (en) * | 2021-08-17 | 2021-11-19 | 京东鲲鹏(江苏)科技有限公司 | Method, device, equipment and medium for determining target area |
| CN113928337A (en) * | 2021-09-30 | 2022-01-14 | 华为技术有限公司 | Method for guiding vehicle to run, related system and storage medium |
| WO2022063005A1 (en) * | 2020-09-22 | 2022-03-31 | 北京智行者科技有限公司 | Global path planning method and apparatus |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116543310B (en) * | 2023-06-30 | 2023-10-31 | 眉山环天智慧科技有限公司 | Road line extraction method based on Voronoi diagram and kernel density |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160210863A1 (en) * | 2015-01-19 | 2016-07-21 | The Aerospace Corporation | Autonomous nap-of-the-earth (anoe) flight path planning for manned and unmanned rotorcraft |
| CN106092102A (en) * | 2016-07-20 | 2016-11-09 | 广州极飞电子科技有限公司 | A kind of unmanned plane paths planning method and device |
| CN109557928A (en) * | 2019-01-17 | 2019-04-02 | 湖北亿咖通科技有限公司 | Automatic driving vehicle paths planning method based on map vector and grating map |
| CN110823241A (en) * | 2019-11-19 | 2020-02-21 | 齐鲁工业大学 | Robot path planning method and system based on passable area skeleton extraction |
| CN110850871A (en) * | 2019-10-21 | 2020-02-28 | 深圳市银星智能科技股份有限公司 | Machine path planning method and mobile robot |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005321370A (en) * | 2004-04-05 | 2005-11-17 | Sony Corp | Navigation system, data processing method and computer program |
| CN107664503A (en) * | 2016-07-29 | 2018-02-06 | 上海汽车集团股份有限公司 | Vehicle path planning method and device |
| CN106767860B (en) * | 2016-11-21 | 2019-12-03 | 江苏大学 | A method of shortening intelligent automobile path planning search time based on heuristic search algorithm |
| CN107992050B (en) * | 2017-12-20 | 2021-05-11 | 广州汽车集团股份有限公司 | Method and device for planning local path motion of unmanned vehicle |
| CN111521189B (en) * | 2020-04-10 | 2022-02-15 | 北京智行者科技有限公司 | Cleaning path planning method and device |
| CN112444263B (en) * | 2020-09-22 | 2023-05-23 | 北京智行者科技股份有限公司 | Global path planning method and device |
-
2020
- 2020-09-22 CN CN202011003654.1A patent/CN112444263B/en active Active
-
2021
- 2021-09-15 WO PCT/CN2021/118566 patent/WO2022063005A1/en active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160210863A1 (en) * | 2015-01-19 | 2016-07-21 | The Aerospace Corporation | Autonomous nap-of-the-earth (anoe) flight path planning for manned and unmanned rotorcraft |
| CN106092102A (en) * | 2016-07-20 | 2016-11-09 | 广州极飞电子科技有限公司 | A kind of unmanned plane paths planning method and device |
| CN109557928A (en) * | 2019-01-17 | 2019-04-02 | 湖北亿咖通科技有限公司 | Automatic driving vehicle paths planning method based on map vector and grating map |
| CN110850871A (en) * | 2019-10-21 | 2020-02-28 | 深圳市银星智能科技股份有限公司 | Machine path planning method and mobile robot |
| CN110823241A (en) * | 2019-11-19 | 2020-02-21 | 齐鲁工业大学 | Robot path planning method and system based on passable area skeleton extraction |
Non-Patent Citations (3)
| Title |
|---|
| JULIAN等: "Semantic Grid-Based Road Model Estimation for Autonomous Driving", 《2019 IEEE INTELLIGENT VEHICLES SYMPOSIUM (IV)》 * |
| 刘远刚等: "地图目标群间骨架线提取的算法研究", 《武汉大学学报信息科学版》 * |
| 林恒: "分类栅格图像序列渐变技术研究及应用", 《中国博士学位论文全文数据库 基础科学辑 第09期》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022063005A1 (en) * | 2020-09-22 | 2022-03-31 | 北京智行者科技有限公司 | Global path planning method and apparatus |
| CN113119995A (en) * | 2021-03-11 | 2021-07-16 | 京东鲲鹏(江苏)科技有限公司 | Path searching method and device, equipment and storage medium |
| CN113219973A (en) * | 2021-05-08 | 2021-08-06 | 浙江工业大学 | Efficient local path control method for mobile robot |
| CN113219973B (en) * | 2021-05-08 | 2022-06-24 | 浙江工业大学 | Local path control method of mobile robot |
| CN113670323A (en) * | 2021-08-17 | 2021-11-19 | 京东鲲鹏(江苏)科技有限公司 | Method, device, equipment and medium for determining target area |
| CN113928337A (en) * | 2021-09-30 | 2022-01-14 | 华为技术有限公司 | Method for guiding vehicle to run, related system and storage medium |
| CN113928337B (en) * | 2021-09-30 | 2023-10-20 | 华为技术有限公司 | Method for guiding vehicle to run and related system and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2022063005A1 (en) | 2022-03-31 |
| CN112444263B (en) | 2023-05-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112444263A (en) | Global path planning method and device | |
| Hashemi et al. | A critical review of real-time map-matching algorithms: Current issues and future directions | |
| US11423677B2 (en) | Automatic detection and positioning of pole-like objects in 3D | |
| US10429195B2 (en) | Method, apparatus, and computer program product for generation of a route using time and space | |
| CN111857160A (en) | Unmanned vehicle path planning method and device | |
| CN109489675A (en) | The path planning based on cost for automatic driving vehicle | |
| US11168989B2 (en) | Supervised point map matcher | |
| US11161501B2 (en) | System and method for optimizing a path for obstacle evasion for a vehicle | |
| US20190376795A1 (en) | Map matching method and apparatus | |
| US10818035B1 (en) | Representing vanishing points using relative distances | |
| JP7330142B2 (en) | Method, Apparatus, Device and Medium for Determining Vehicle U-Turn Path | |
| CN114440916B (en) | Navigation method, device, equipment and storage medium | |
| US11120566B2 (en) | Determining vanishing points based on feature maps | |
| US11227167B2 (en) | Determining vanishing points based on lane lines | |
| EP4177570A1 (en) | Standard-definition to high-definition navigation route determination | |
| CN113924459A (en) | Dynamic sensor range detection for vehicle navigation | |
| WO2023039666A1 (en) | Systems, methods, and media for occlusion-aware motion planning | |
| CN114295119A (en) | Map construction method and device | |
| RU2574040C2 (en) | Method and device for determining probability of exiting area identified on digital map as open terrain | |
| CN112185149B (en) | Path planning method and system based on urban road network data | |
| US20210270629A1 (en) | Method and apparatus for selecting a path to a destination | |
| CN113048981A (en) | DEM-oriented method for road-free regional path planning algorithm | |
| Selloum et al. | Lane level positioning using particle filtering | |
| CN112734811A (en) | Obstacle tracking method, obstacle tracking device and chip | |
| KR101622176B1 (en) | Method generating terrain data for global path planning |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20210806 Address after: B4-006, maker Plaza, 338 East Street, Huilongguan town, Changping District, Beijing 100096 Applicant after: Beijing Idriverplus Technology Co.,Ltd. Address before: 401122 No.1, 1st floor, building 3, No.21 Yunzhu Road, Yubei District, Chongqing Applicant before: Chongqing Zhixing Information Technology Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| CB02 | Change of applicant information |
Address after: B4-006, maker Plaza, 338 East Street, Huilongguan town, Changping District, Beijing 100096 Applicant after: Beijing Idriverplus Technology Co.,Ltd. Address before: B4-006, maker Plaza, 338 East Street, Huilongguan town, Changping District, Beijing 100096 Applicant before: Beijing Idriverplus Technology Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |