CN113359760B - Method for eliminating vehicle collision in optimal path algorithm operation result - Google Patents
Method for eliminating vehicle collision in optimal path algorithm operation result Download PDFInfo
- Publication number
- CN113359760B CN113359760B CN202110749386.6A CN202110749386A CN113359760B CN 113359760 B CN113359760 B CN 113359760B CN 202110749386 A CN202110749386 A CN 202110749386A CN 113359760 B CN113359760 B CN 113359760B
- Authority
- CN
- China
- Prior art keywords
- group
- vertex
- real
- vehicle
- real vertex
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000012546 transfer Methods 0.000 claims abstract description 207
- 238000012937 correction Methods 0.000 claims abstract description 11
- 230000007704 transition Effects 0.000 claims description 14
- 238000011065 in-situ storage Methods 0.000 claims description 2
- 238000002715 modification method Methods 0.000 abstract description 3
- 230000002123 temporal effect Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- 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
- G05D1/0223—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving speed control of the vehicle
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- 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
- G05D1/0221—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving a learning process
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Traffic Control Systems (AREA)
Abstract
The application discloses a method for eliminating vehicle collision in an optimal path algorithm operation result, which comprises the steps of inquiring in a first group of optimal paths according to a point collision inquiry condition and an edge collision inquiry condition to obtain a first controllable vehicle, a second controllable vehicle and first to sixth real vertexes; generating a second vehicle transfer system by modifying the first vehicle transfer system by a first modification method; inputting a second vehicle transfer system into a path planning sub-algorithm to obtain a second group of transfer systems; and correcting the second group of transfer systems by a second correction method to generate a third group of transfer systems, and inputting the third group of transfer systems into an optimal path sub-algorithm to generate a second group of optimal paths.
Description
Technical Field
The application relates to the field of vehicle path planning, in particular to a method for eliminating vehicle collision in an optimal path algorithm operation result.
Background
In the prior art, the optimal path of the controllable vehicle can be planned through an optimal path algorithm. The optimal path algorithm can be used for path planning for a plurality of vehicles which start at different points and have the same speed and start at the same time, so that all controllable vehicles can enter a group optimal circular path, wherein the group optimal circular path is defined to minimize the maximum time interval for all controllable vehicles to complete task protocols; the map is defined with a plurality of real vertexes and adjacent real vertexes of the real vertexes; by "adjacent" is meant that two real vertices may be in direct communication, and it is particularly emphasized here that the map allows two adjacent real vertices to be in direct communication through two paths of different distances. The task protocol is defined to be a loadable vertex and a unloadable vertex for each controllable vehicle, and all controllable vehicles are considered to complete one-time loading and one-time unloading as all controllable vehicles complete one-time task protocol. In the optimal path algorithm, each vehicle does not need to stay after completing loading and unloading, and can be regarded as being completed instantly.
The optimal path algorithm comprises a path planning sub-algorithm and an optimal path sub-algorithm in sequence. Wherein, the Path Planning sub-algorithm (Multi-Robot Path Planning with Temporal Logic Constraints) refers to the International Journal of Robotics Research Journal 2013, volume 32, pages 8, 889 to 911, which is published by Ulusoy A et al and is named as optimization and Robustness in Multi-Robot Path Planning with Temporal Logic Constraints; the path planning algorithm is used to obtain a vehicle transfer system and a task specification to generate a group transfer system. The best Path sub-algorithm (Optimal Path Planning with Temporal logical Constraints) is described in detail in International Journal of Robotics Research Journal 2011, volume 30, pages 1695 to 1708, which is published by Smith S L et al under the name of Optimal Path Planning for Surveillance with Temporal logical Constraints; the best path sub-algorithm is used to acquire the group transfer system and generate a group best path. Wherein the vehicle transfer system comprises a transfer cost for each vehicle to transfer from any real vertex to an adjacent real vertex. The group transfer system comprises all group states and transfer costs for transferring from each group state to an adjacent group state; wherein the group state defines positions of all vehicles at the same time and at least one vehicle is located at a real vertex, and the positions of the vehicles which are not located at the real vertex in the group state are defined as virtual vertices; the neighboring group state refers to a group state that a certain group state can reach without going through other group states. The group optimal path comprises a group optimal pre-circulation entry path and a group optimal circulation path, and the group optimal pre-circulation entry path comprises a plurality of groups of states arranged according to time sequence; the group optimal circular path comprises a plurality of groups of states which are arranged in time sequence of the unidirectional circulation, wherein the state of the group arranged at the last in the group optimal circular path is the adjacent group state as the state of the group arranged at the top in the optimal circular path.
The optimal path algorithm can be regarded as an ideal optimal path planning method, which does not perform real-time control, and only performs path planning on each vehicle under an ideal state based on a known map and a task protocol. However, optimal path planning does not ensure that point and side collisions between vehicles occur. A point collision is a collision in which two or more vehicles are located at the same real vertex at the same time. The side collision means that two or more vehicles meet each other at a position between two adjacent real vertices. Such point and edge collisions are harmful or even unacceptable in some applications.
Disclosure of Invention
The present application is directed to overcoming the above-mentioned drawbacks and problems in the prior art, and providing a method for eliminating vehicle collisions in the operation result of the optimal path algorithm, wherein the group optimal path obtained by the method can avoid point collisions and side collisions of controllable vehicles, and simultaneously ensure that all controllable vehicles can effectively run through a map.
In order to achieve the purpose, the following technical scheme is adopted:
a method for eliminating vehicle collision in the operation result of an optimal path algorithm is provided, the number of the vehicles is at least two, at least one of the vehicles is a controllable vehicle; the optimal path algorithm sequentially comprises a path planning sub-algorithm and an optimal path sub-algorithm; the path planning sub-algorithm is used for acquiring a vehicle transfer system and generating a group transfer system; the optimal path sub-algorithm is used for acquiring the group transfer system and generating a group optimal path as the operation result of the optimal path algorithm; the vehicle transfer system comprises a transfer cost of each vehicle for transferring from any real vertex to an adjacent real vertex; the group transfer system comprises all group states and transfer costs for transferring from each group state to an adjacent group state; wherein the group state defines positions of all vehicles at the same time and at least one vehicle is located at a real vertex, and the positions of the vehicles which are not located at the real vertex in the group state are defined as virtual vertices; the adjacent group state is a group state which can be reached by a certain group state without passing through other group states; the group optimal path comprises a group optimal pre-circulation entry path and a group optimal circulation path, and the group optimal pre-circulation entry path comprises a plurality of groups of states arranged according to time sequence; the group optimal circulating path comprises a plurality of groups of states which are arranged in time sequence and are in unidirectional circulation, wherein two groups of states which are adjacent to each other in time sequence, and the group state which is arranged at the last in the group optimal circulating path and the group state which is arranged at the forefront in the optimal circulating path are adjacent to each other in time sequence; the path planning sub-algorithm acquires a first vehicle transfer system and a task specification and generates a first group of transfer systems; the optimal path sub-algorithm acquires a first group of transfer systems and generates a first group of optimal paths as the operation result of the optimal path algorithm; the method is characterized by comprising the following steps: inquiring in the first group of optimal paths according to the point collision inquiry conditions to obtain all first controllable vehicles, a first real vertex, a second real vertex and a third real vertex; inquiring in the first group of optimal paths according to the side collision inquiry condition to obtain a second controllable vehicle, a fourth real vertex, a fifth vertex and a sixth real vertex; the point collision query condition is that at least two vehicles in the same group of states are positioned at the same real top point, and the group state meeting the point collision query condition in the first group of optimal paths is a first group state; in the first group of states, all the controllable vehicles which are positioned at the same real vertex with other vehicles are first controllable vehicles; the position of the first controllable vehicle in the first set of states is a first real vertex; the adjacent real vertex positioned in front of the first real vertex in the first vehicle optimal path corresponding to the first controllable vehicle is a second real vertex; an adjacent real vertex positioned behind the first real vertex in the first vehicle optimal path corresponding to the first controllable vehicle is a third real vertex; the first optimal path of the vehicle comprises a plurality of positions of any vehicle in the first group of optimal paths which are arranged in time sequence; the side collision query condition is that positions of at least two vehicles in two adjacent group states are interchanged with each other, the two adjacent group states meeting the side collision query condition in the first group of optimal paths are respectively a second group state and a third group state according to time sequence, the controllable vehicles meeting the side collision query condition in the first group of optimal paths are all second controllable vehicles, the position of the second controllable vehicle in the second group state is a first position, if the first position is a real vertex, the first position is defined as a fourth real vertex, if the first position is a virtual vertex, the real vertex which is located in front of the first position in the first vehicle optimal path corresponding to the second controllable vehicle and is closest to the first position is a fourth real vertex; the position of the second controllable vehicle in the third group of states is a second position, if the second position is a real vertex, the second position is defined as a fifth vertex, if the second position is a virtual vertex, the real vertex which is positioned behind the second position and is closest to the second position in the first vehicle optimal path corresponding to the second controllable vehicle is a fifth vertex; an adjacent real vertex positioned behind the fifth vertex in the first vehicle optimal path corresponding to the second controllable vehicle is a sixth real vertex; generating a second vehicle transfer system for the first vehicle transfer system by a first correction method; the second vehicle transfer system is defined to include a transfer cost for each vehicle to transfer from any real vertex to an adjacent real vertex or to itself; the first correction method includes: adding first transfer costs transferred to the first vehicle transfer system to all second real vertexes and all third real vertexes of all first controllable vehicles in the first vehicle transfer system, and adding second transfer costs transferred to the first vehicle transfer system to all sixth real vertexes of all second controllable vehicles in the first vehicle transfer system; the first transfer cost and the second transfer cost are both less than any transfer cost in the first vehicle transfer system; determining a corresponding seventh real vertex for all fourth real vertices of all second controllable vehicles in the first vehicle transfer system, wherein the seventh real vertex is a real vertex adjacent to the corresponding fourth real vertex, if the first vehicle transfer system has a transfer cost of transferring the fourth real vertex to the seventh real vertex and does not have a transfer cost of transferring the seventh real vertex to the fourth real vertex, a third transfer cost of transferring the seventh real vertex to the fourth real vertex is increased, and the third transfer cost is equal to the sum of the transfer cost of transferring the fourth real vertex to the seventh real vertex and an in-situ turning cost; if the first vehicle transfer system has the transfer cost of the seventh real vertex to the fourth real vertex and does not have the transfer cost of the fourth real vertex to the seventh real vertex, increasing the fourth transfer cost of the fourth real vertex to the seventh real vertex, wherein the fourth transfer cost is equal to the sum of the transfer cost of the seventh real vertex to the fourth real vertex and the in-place turning cost; inputting the second vehicle transfer system into a path planning sub-algorithm to obtain a second group of transfer systems; generating a third group of transfer systems for the second group of transfer systems by a second correction method; the second correction method includes: acquiring all group states meeting the point collision query condition in the second group of transfer systems according to the point collision query condition and defining the group states as a fourth group state, acquiring all two adjacent group states meeting the edge collision query condition in the second group of transfer systems according to the edge collision query condition and respectively defining the two adjacent group states as a fifth group state and a sixth group state according to time sequence; deleting all fourth group states, and modifying the transition costs of the transition to all fifth group states and all sixth group states to be more than or equal to three times of the maximum value of all transition costs in the second group transition system; inputting the third set of transfer systems into the optimal path sub-algorithm to obtain a second set of optimal paths.
Further, the seventh real vertex is a real vertex with the smallest virtual transfer cost among all adjacent real vertices corresponding to the fourth real vertex; the virtual transfer cost of any adjacent real vertex is determined according to the following method: if the second controllable vehicle in the first vehicle transfer system has neither the transfer cost for transferring from the fourth real vertex to the adjacent real vertex nor the transfer cost for transferring from the adjacent real vertex to the fourth real vertex, the virtual transfer cost of the adjacent real vertex is infinite; if the transfer cost from the fourth real vertex to the adjacent real vertex of the second controllable vehicle in the first vehicle transfer system also has the transfer cost from the adjacent real vertex to the fourth real vertex, the virtual transfer cost of the adjacent real vertex is the sum of the transfer cost from the fourth real vertex to the adjacent real vertex and the transfer cost from the adjacent real vertex to the fourth real vertex; if the second controllable vehicle in the first vehicle transfer system has the transfer cost from the fourth real vertex to the adjacent real vertex or the transfer cost from the adjacent real vertex to the fourth real vertex, the virtual transfer cost of the adjacent real vertex is correspondingly the sum of twice the transfer cost from the fourth real vertex to the adjacent real vertex and the in-place turning cost or the sum of twice the transfer cost from the adjacent real vertex to the fourth real vertex and the in-place turning cost.
Further, the vehicle also comprises uncontrollable vehicles, the uncontrollable vehicles run according to circulating paths, and the circulating paths of the uncontrollable vehicles do not intersect.
Compared with the prior art, the scheme has the following beneficial effects:
according to the second group of optimal paths obtained by the method, the fourth group state, the fifth group state and the sixth group state which are possible to generate point collision and side collision are deleted in the third group transfer system, so that the point collision and the side collision of the controllable vehicle can be avoided.
Generally, the elimination of the collision in the optimal path algorithm can be realized by directly deleting the point collision and the edge collision in the group transfer system, but the applicant finds that if the group transfer system is directly processed, the group transfer system can be divided into a plurality of strongly connected components, wherein part or all of the strongly connected components are not connected, so that the optimal path cannot be obtained. Therefore, the applicant creatively processes the real vertices meeting the query conditions of the point collision and the side collision in the first vehicle transfer system, namely adding wait to the real vertices before and after the point collision occurs, while for the side collision, only adding wait is still insufficient to avoid strong connection components, so that the applicant adds a return program to the real vertices where the side collision occurs, and adds wait to the real vertices after the side collision, and through mathematical reasoning and calculation, the technical scheme can avoid dividing the group transfer system into a plurality of strong connection components, and ensure that an optimal path can be obtained.
Defining a seventh real vertex as a real vertex with the minimum virtual transfer cost in all adjacent real vertices of the fourth real vertex; the second group of optimal paths can be further optimized, and collision avoidance cost is further reduced.
The method defined in the present application can be used simultaneously in situations involving uncontrollable vehicles (interfering vehicles).
Detailed Description
In the claims and specification, unless otherwise specified the terms "first", "second" or "third", etc., are used to distinguish between different items and are not used to describe a particular order.
In the claims and specification, unless otherwise defined, the terms "comprising", "having" and variations thereof mean "including but not limited to".
In the claims and specification, unless otherwise specified, the term "vehicle" means any object that can be moved between solid points on a "map" and which can be moved along a well-defined path controlled by a best-path algorithm.
In the claims and specification, unless otherwise specified, the term "transfer cost exists" means that the transfer cost of a vehicle from one real vertex to an adjacent real vertex is a value below a preset transfer cost threshold, which can be determined according to actual needs, and is generally a maximum value much higher than a normal transfer cost.
In the claims and the description, unless otherwise defined, the term "absence of a transfer cost" refers not only to the case where there is actually no transfer cost but also to the case where the corresponding transfer cost of the vehicle is higher than the above-mentioned transfer cost threshold.
In the claims and specification, unless otherwise specified, the term "delete group state" may refer to both deletion of that group of states in a group transfer system, i.e., no transfer of any other group of states to that group of states is allowed, and no transfer of that group of states to any other group of states is allowed.
In the claims and specification, unless otherwise defined, the term "cost of turning around in place" may refer to the length of time a vehicle is used to turn around in place.
An embodiment is provided below for explaining the method for eliminating the vehicle collision in the operation result of the optimal path algorithm provided by the present application in detail.
In this embodiment, the number of vehicles is at least two, and the vehicles include controllable vehicles and uncontrollable vehicles, wherein the controllable vehicles refer to vehicles whose operation paths can be changed by inputting corresponding instructions to the operation systems of the controllable vehicles, and the uncontrollable vehicles refer to vehicles which operate along the circulation paths and whose respective circulation paths do not intersect. All vehicles run on a defined map and the starting points of all vehicles differ from one another, but all vehicles are at the same speed and the same departure time. A plurality of real vertexes and adjacent real vertexes of the real vertexes are defined in the map.
In this embodiment, the definitions of the optimal path algorithm, the path planning sub-algorithm, the optimal path sub-algorithm, the map, the real vertices, the adjacent real vertices, the vehicle transfer system, the task specification, the group state system, the group state, the virtual vertices, the adjacent group state, the group optimal path before entering the loop, and the group optimal loop path are the same as the related definitions in the background art.
In this embodiment, firstly, an optimal path algorithm needs to be run and a corresponding operation result is obtained, which includes inputting a first vehicle transfer system and a task specification into a path planning sub-algorithm and obtaining a first group of transfer systems, inputting the first group of transfer systems into an optimal path sub-algorithm and obtaining a first group of optimal paths, and then implementing the method for eliminating vehicle collision in the operation result of the optimal path algorithm provided by the present application to provide and obtain a second group of optimal paths, so as to implement collision avoidance.
Specifically, the first step includes obtaining all the first controllable vehicles, the first real vertex, the second real vertex and the third real vertex in the first group of optimal paths according to the point collision query condition query, and obtaining all the second controllable vehicles, the fourth real vertex, the fifth real vertex and the sixth real vertex in the first group of optimal paths according to the edge collision query condition query.
The point collision query condition is that at least two vehicles in the same group of states are located at the same real vertex. The group states meeting the point collision query condition in the first group of optimal paths are a first group state, all the controllable vehicles which are positioned at the same real vertex with other vehicles in the first group state are first controllable vehicles, and the position of the first controllable vehicle in the first group state is a first real vertex. And the adjacent real vertex positioned in front of the first real vertex in the first vehicle optimal path corresponding to the first controllable vehicle is a second real vertex, and the adjacent real vertex positioned behind the first real vertex in the first vehicle optimal path corresponding to the first controllable vehicle is a third real vertex. The first optimal path of the vehicle comprises a plurality of positions of any vehicle in the first group of optimal paths in time sequence.
For example, there are a controllable vehicle a and a controllable vehicle B, and the position of the controllable vehicle a in the group state 1 and the position of the controllable vehicle B in the group state 1 are both located at a same real vertex 1, then the real vertex 1 is a first real vertex, the controllable vehicle a and the controllable vehicle B are both first controllable vehicles, and the group state 1 is a first group state; taking the controllable vehicle a as an example, the controllable vehicle a operates according to the controllable vehicle a optimal path corresponding to the controllable vehicle a in the first group of optimal paths, where the controllable vehicle a optimal path includes a plurality of positions to be passed through, which are arranged in time sequence during the operation of the controllable vehicle a, the controllable vehicle a optimal path includes the first real vertex, and a real vertex before the first real vertex in the controllable vehicle a optimal path is the second real vertex corresponding to the controllable vehicle a, and a real vertex after the first real vertex in the controllable vehicle a optimal path is the third real vertex corresponding to the controllable vehicle a.
The side collision query condition is that the positions of at least two vehicles in two adjacent group states are interchanged with each other; two adjacent group states meeting the side collision query condition in the first group of optimal paths are respectively a second group state and a third group state according to time sequence, the controllable vehicles meeting the side collision query condition in the first group of optimal paths are all second controllable vehicles, the position of the second controllable vehicle in the second group state is a first position, the first position is defined as a fourth real vertex if the first position is a real vertex, the real vertex which is positioned in front of the first position and is closest to the first position in the first vehicle optimal path corresponding to the second controllable vehicle in the first position is a fourth real vertex in the first position if the first position is a virtual vertex, the position of the second controllable vehicle in the third group state is a second position, the second position is defined as a fifth vertex in the second position if the second position is a real vertex, and the real vertex which is positioned in back of the second position and is closest to the second position in the first vehicle optimal path corresponding to the second controllable vehicle in the second position if the second position is a virtual vertex is a fifth vertex, and the adjacent real vertex positioned behind the fifth vertex in the optimal path of the first vehicle corresponding to the second controllable vehicle is a sixth real vertex.
For example, there are a controllable vehicle C and a controllable vehicle D, in a group state one located before in time sequence, the controllable vehicle C is located at position 1, and the controllable vehicle D is located at position 2, and in a group state two located after in time sequence and adjacent to the group state one, the controllable vehicle C is located at position 2, and the controllable vehicle D is located at position 1, and at this time, the controllable vehicle C and the controllable vehicle D meet a side collision query condition, both are second controllable vehicles, the group state one is a second group state, the group state two is a third group state, position 1 is a first position, and position 2 is a second position. Taking the controllable vehicle C as an example, it operates according to the controllable vehicle C optimal path corresponding to the controllable vehicle C optimal path in the first set of optimal paths, in the controllable vehicle C optimal path, if the position 1 is a real vertex, the real vertex at the position 1 is defined as a fourth real vertex, if the position 1 is an imaginary vertex, the real vertex before the position 1 and closest to the position 1 is defined as a fourth real vertex, if the position 2 is a real vertex, the real vertex at the position 2 is defined as a fifth vertex, if the position 2 is an imaginary vertex, the real vertex after the position 2 and closest to the position 2 is defined as a fifth vertex, and the adjacent real vertex after the fifth vertex is defined as a sixth real vertex.
Step two includes modifying the first vehicle transfer system according to the first modification method and generating a second vehicle transfer system defined to include a transfer penalty for each vehicle to transfer from any one real vertex to an adjacent real vertex or to itself. Wherein increasing the transfer cost to itself serves to index the vehicle to wait at the corresponding real vertex for a time at least equivalent to its transfer cost.
Specifically, the first correction method includes: adding a first transfer cost for transferring to the first controllable vehicle to all second real vertexes and all third real vertexes of all first controllable vehicles in the first vehicle transfer system; meanwhile, second transfer costs transferred to the first vehicle transfer system are added to all sixth real vertexes of all second controllable vehicles in the first vehicle transfer system, and a real vertex adjacent to the fourth real vertex is determined for all fourth real vertexes of all second controllable vehicles in the first vehicle transfer system and is defined as a seventh real vertex; if the first vehicle transfer system has the transfer cost of the fourth real vertex transferred to the seventh real vertex and does not have the transfer cost of the seventh real vertex transferred to the fourth real vertex, increasing a third transfer cost of the seventh real vertex transferred to the fourth real vertex, wherein the third transfer cost is equal to the sum of the transfer cost of the fourth real vertex transferred to the seventh real vertex and the in-place turning cost, and if the first vehicle transfer system has the transfer cost of the seventh real vertex transferred to the fourth real vertex and does not have the transfer cost of the fourth real vertex transferred to the seventh real vertex, increasing a fourth transfer cost of the fourth real vertex transferred to the seventh real vertex, wherein the fourth transfer cost is equal to the sum of the transfer cost of the seventh real vertex transferred to the fourth real vertex and the in-place turning cost.
The first transfer cost and the second transfer cost are both smaller than any transfer cost in the first vehicle transfer system, and as a preferable scheme, the first transfer cost and the second transfer cost may be set to be the greatest common divisor of all transfer costs in the first vehicle transfer system.
As a preferable aspect, the seventh real vertex is defined as a real vertex having the smallest virtual transition cost among all adjacent real vertices of the fourth real vertex. The virtual transfer cost of any adjacent real vertex is determined according to the following method: if the second controllable vehicle in the first vehicle transfer system has neither the transfer cost for transferring from the fourth real vertex to the adjacent real vertex nor the transfer cost for transferring from the adjacent real vertex to the fourth real vertex, the virtual transfer cost for the adjacent real vertex is infinite; if the second controllable vehicle in the first vehicle transfer system has the transfer cost from the fourth real vertex to the adjacent real vertex and also has the transfer cost from the adjacent real vertex to the fourth real vertex, the virtual transfer cost of the adjacent real vertex is the sum of the transfer cost from the fourth real vertex to the adjacent real vertex and the transfer cost from the adjacent real vertex to the fourth real vertex; if the second controllable vehicle in the first vehicle transfer system has the transfer cost from the fourth real vertex to the adjacent real vertex or the transfer cost from the adjacent real vertex to the fourth real vertex, the virtual transfer cost of the adjacent real vertex is the sum of two times of the transfer cost from the fourth real vertex to the adjacent real vertex and the in-place turning cost, or the sum of two times of the transfer cost from the adjacent real vertex to the fourth real vertex and the in-place turning cost.
And step three, inputting the second vehicle transfer system into a path planning sub-algorithm to obtain a second group of transfer systems.
Step four includes modifying the second set of transfer systems according to the second modification method to generate a third set of transfer systems.
Specifically, the second correction method includes: acquiring all group states meeting the point collision query condition in the second group of transfer systems according to the point collision query condition and defining the group states as a fourth group state, acquiring all two adjacent group states meeting the edge collision query condition in the second group of transfer systems according to the edge collision query condition and respectively defining the two adjacent group states as a fifth group state and a sixth group state according to time sequence; deleting all fourth group states, and modifying the transition costs of the transition to all fifth group states and all sixth group states to be more than or equal to three times of the maximum value of all transition costs in the second group transition system; generally, the transition costs of transitioning to all fifth set states and all sixth set states should be modified to a maximum.
And step five, inputting the third group of transfer systems into the optimal path sub-algorithm to obtain a second group of optimal paths.
In the method for eliminating the vehicle collision in the operation result of the optimal path algorithm in the embodiment, the first group of optimal paths are obtained by performing the first calculation on the operation path of the vehicle based on the existing optimal path algorithm, and the fourth group state, the fifth group state and the sixth group state which are possible to generate the point collision and the side collision are deleted in the third group transfer system in the second group of optimal paths obtained according to the method in the embodiment, so that the point collision and the side collision of the controllable vehicle can be avoided. Meanwhile, as the applicant creatively processes the real vertexes meeting the query conditions of the point collision and the side collision in the first vehicle transfer system, namely, the real vertexes before and after the point collision are added with the waiting, and for the side collision, the addition of the waiting is still insufficient to avoid the generation of the strong connection components, the applicant adds a return program on the real vertexes with the side collision, and the addition of the waiting on the real vertexes after the side collision is carried out, and through mathematical reasoning and calculation, the technical scheme can avoid the division of the group transfer system into a plurality of strong connection components, and can ensure that the optimal path can be obtained.
In this embodiment, the seventh real vertex is defined as the real vertex with the minimum virtual transfer cost among all the adjacent real vertices of the fourth real vertex; the second group of optimal paths can be further optimized, and collision avoidance cost is further reduced.
The uncontrollable vehicle in the present embodiment is not essential. When the uncontrollable vehicles are not included, the number of controllable vehicles must be two or more. In this case, the related object of the invention can be achieved also by applying the present embodiment.
The description of the above specification and examples is intended to be illustrative of the scope of the present application and is not intended to be limiting.
Claims (3)
1. A method for eliminating vehicle collision in the operation result of an optimal path algorithm,
the number of the vehicles is at least two, and at least one of the vehicles is a controllable vehicle;
the optimal path algorithm sequentially comprises a path planning sub-algorithm and an optimal path sub-algorithm; the path planning sub-algorithm is used for acquiring a vehicle transfer system and generating a group transfer system; the optimal path sub-algorithm is used for acquiring the group transfer system and generating a group optimal path as the operation result of the optimal path algorithm;
the vehicle transfer system comprises a transfer cost of each vehicle for transferring from any real vertex to an adjacent real vertex;
the group transfer system comprises all group states and transfer costs for transferring from each group state to an adjacent group state; wherein the group state defines positions of all vehicles at the same time and at least one vehicle is located at a real vertex, and the positions of the vehicles which are not located at the real vertex in the group state are defined as virtual vertices; the adjacent group state is a group state which can be reached by a certain group state without passing through other group states;
the group optimal path comprises a group optimal pre-circulation entry path and a group optimal circulation path, and the group optimal pre-circulation entry path comprises a plurality of groups of states arranged according to time sequence; the group optimal circulating path comprises a plurality of groups of states which are arranged in time sequence and are in unidirectional circulation, wherein two groups of states which are adjacent to each other in time sequence, and the group state which is arranged at the last in the group optimal circulating path and the group state which is arranged at the forefront in the optimal circulating path are adjacent to each other in time sequence;
the path planning sub-algorithm acquires a first vehicle transfer system and a task specification and generates a first group of transfer systems; the optimal path sub-algorithm acquires a first group of transfer systems and generates a first group of optimal paths as the operation result of the optimal path algorithm; the method is characterized by comprising the following steps:
inquiring in the first group of optimal paths according to the point collision inquiry conditions to obtain all first controllable vehicles, a first real vertex, a second real vertex and a third real vertex; inquiring in the first group of optimal paths according to the side collision inquiry condition to obtain a second controllable vehicle, a fourth real vertex, a fifth vertex and a sixth real vertex; the point collision query condition is that at least two vehicles in the same group of states are positioned at the same real top point, and the group state meeting the point collision query condition in the first group of optimal paths is a first group state; in the first group of states, all the controllable vehicles which are positioned at the same real vertex with other vehicles are first controllable vehicles; the position of the first controllable vehicle in the first set of states is a first real vertex; the adjacent real vertex positioned in front of the first real vertex in the first vehicle optimal path corresponding to the first controllable vehicle is a second real vertex; an adjacent real vertex positioned behind the first real vertex in the first vehicle optimal path corresponding to the first controllable vehicle is a third real vertex; the first optimal path of the vehicle comprises a plurality of positions of any vehicle in the first group of optimal paths which are arranged in time sequence; the side collision query condition is that positions of at least two vehicles in two adjacent group states are interchanged with each other, the two adjacent group states meeting the side collision query condition in the first group of optimal paths are respectively a second group state and a third group state according to time sequence, the controllable vehicles meeting the side collision query condition in the first group of optimal paths are all second controllable vehicles, the position of the second controllable vehicle in the second group state is a first position, if the first position is a real vertex, the first position is defined as a fourth real vertex, if the first position is a virtual vertex, the real vertex which is located in front of the first position in the first vehicle optimal path corresponding to the second controllable vehicle and is closest to the first position is a fourth real vertex; the position of the second controllable vehicle in the third group of states is a second position, if the second position is a real vertex, the second position is defined as a fifth vertex, if the second position is a virtual vertex, the real vertex which is positioned behind the second position and is closest to the second position in the first vehicle optimal path corresponding to the second controllable vehicle is a fifth vertex; an adjacent real vertex positioned behind the fifth vertex in the first vehicle optimal path corresponding to the second controllable vehicle is a sixth real vertex;
generating a second vehicle transfer system for the first vehicle transfer system by a first correction method; the second vehicle transfer system is defined to include a transfer cost for each vehicle to transfer from any real vertex to an adjacent real vertex or to itself; the first correction method includes: adding first transfer costs transferred to the first vehicle transfer system to all second real vertexes and all third real vertexes of all first controllable vehicles in the first vehicle transfer system, and adding second transfer costs transferred to the first vehicle transfer system to all sixth real vertexes of all second controllable vehicles in the first vehicle transfer system; the first transfer cost and the second transfer cost are both less than any transfer cost in the first vehicle transfer system; determining a corresponding seventh real vertex for all fourth real vertices of all second controllable vehicles in the first vehicle transfer system, wherein the seventh real vertex is a real vertex adjacent to the corresponding fourth real vertex, if the first vehicle transfer system has a transfer cost of transferring the fourth real vertex to the seventh real vertex and does not have a transfer cost of transferring the seventh real vertex to the fourth real vertex, a third transfer cost of transferring the seventh real vertex to the fourth real vertex is increased, and the third transfer cost is equal to the sum of the transfer cost of transferring the fourth real vertex to the seventh real vertex and an in-situ turning cost; if the first vehicle transfer system has the transfer cost of the seventh real vertex to the fourth real vertex and does not have the transfer cost of the fourth real vertex to the seventh real vertex, increasing the fourth transfer cost of the fourth real vertex to the seventh real vertex, wherein the fourth transfer cost is equal to the sum of the transfer cost of the seventh real vertex to the fourth real vertex and the in-place turning cost;
inputting the second vehicle transfer system into a path planning sub-algorithm to obtain a second group of transfer systems;
generating a third group of transfer systems for the second group of transfer systems by a second correction method; the second correction method includes: acquiring all group states meeting the point collision query condition in the second group of transfer systems according to the point collision query condition and defining the group states as a fourth group state, acquiring all two adjacent group states meeting the edge collision query condition in the second group of transfer systems according to the edge collision query condition and respectively defining the two adjacent group states as a fifth group state and a sixth group state according to time sequence; deleting all fourth group states, and modifying the transition costs of the transition to all fifth group states and all sixth group states to be more than or equal to three times of the maximum value of all transition costs in the second group transition system;
inputting the third set of transfer systems into the optimal path sub-algorithm to obtain a second set of optimal paths.
2. The method for eliminating vehicle collision in the operation result of the optimal path algorithm as claimed in claim 1, wherein the seventh real vertex is a real vertex with the smallest virtual transfer cost among all the adjacent real vertices corresponding to the fourth real vertex;
the virtual transfer cost of any adjacent real vertex is determined according to the following method:
if the second controllable vehicle in the first vehicle transfer system has neither the transfer cost for transferring from the fourth real vertex to the adjacent real vertex nor the transfer cost for transferring from the adjacent real vertex to the fourth real vertex, the virtual transfer cost of the adjacent real vertex is infinite;
if the transfer cost from the fourth real vertex to the adjacent real vertex of the second controllable vehicle in the first vehicle transfer system also has the transfer cost from the adjacent real vertex to the fourth real vertex, the virtual transfer cost of the adjacent real vertex is the sum of the transfer cost from the fourth real vertex to the adjacent real vertex and the transfer cost from the adjacent real vertex to the fourth real vertex;
if the second controllable vehicle in the first vehicle transfer system has the transfer cost from the fourth real vertex to the adjacent real vertex or the transfer cost from the adjacent real vertex to the fourth real vertex, the virtual transfer cost of the adjacent real vertex is correspondingly the sum of twice the transfer cost from the fourth real vertex to the adjacent real vertex and the in-place turning cost or the sum of twice the transfer cost from the adjacent real vertex to the fourth real vertex and the in-place turning cost.
3. A method for eliminating vehicle collisions in the results of the best path algorithm as claimed in claim 1, wherein: the vehicle also comprises uncontrollable vehicles, wherein the uncontrollable vehicles run along circulating paths, and the circulating paths of the uncontrollable vehicles do not intersect.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110749386.6A CN113359760B (en) | 2021-07-01 | 2021-07-01 | Method for eliminating vehicle collision in optimal path algorithm operation result |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110749386.6A CN113359760B (en) | 2021-07-01 | 2021-07-01 | Method for eliminating vehicle collision in optimal path algorithm operation result |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113359760A CN113359760A (en) | 2021-09-07 |
| CN113359760B true CN113359760B (en) | 2022-04-22 |
Family
ID=77537892
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110749386.6A Active CN113359760B (en) | 2021-07-01 | 2021-07-01 | Method for eliminating vehicle collision in optimal path algorithm operation result |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113359760B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018099782A1 (en) * | 2016-11-30 | 2018-06-07 | Robert Bosch Gmbh | Method for collision testing with computing-time efficiency in path planning for a vehicle |
| CN110488826A (en) * | 2019-08-20 | 2019-11-22 | 集美大学 | A kind of AGV collision prevention method, terminal device and storage medium suitable for path conflict |
| CN111332279A (en) * | 2018-12-18 | 2020-06-26 | 北京京东尚科信息技术有限公司 | Parking path generation method and device |
| WO2020161928A1 (en) * | 2019-02-06 | 2020-08-13 | 三菱電機株式会社 | Vehicle control device and vehicle control method |
| WO2021048966A1 (en) * | 2019-09-12 | 2021-03-18 | 三菱電機株式会社 | Drive assistance device, vehicle control device, drive assistance system, and drive assistance method |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10884430B2 (en) * | 2015-09-11 | 2021-01-05 | The Trustees Of The University Of Pennsylvania | Systems and methods for generating safe trajectories for multi-vehicle teams |
| FR3041590B1 (en) * | 2015-09-30 | 2018-08-17 | Renault S.A.S | SYSTEM FOR CONTROLLING THE DIRECTION OF A MOTOR VEHICLE IN THE EVENT OF AN IMMINENT COLLISION WITH AN OBSTACLE |
| CN106919181A (en) * | 2016-10-20 | 2017-07-04 | 湖南大学 | A kind of unmanned plane barrier-avoiding method |
| EP3455591A4 (en) * | 2017-06-19 | 2019-07-10 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for displaying movement of vehicle on map |
| US11230273B2 (en) * | 2019-05-02 | 2022-01-25 | Liebherr Mining Equipment Newport News Co. | Method for autonomously controlling a vehicle |
-
2021
- 2021-07-01 CN CN202110749386.6A patent/CN113359760B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018099782A1 (en) * | 2016-11-30 | 2018-06-07 | Robert Bosch Gmbh | Method for collision testing with computing-time efficiency in path planning for a vehicle |
| CN111332279A (en) * | 2018-12-18 | 2020-06-26 | 北京京东尚科信息技术有限公司 | Parking path generation method and device |
| WO2020161928A1 (en) * | 2019-02-06 | 2020-08-13 | 三菱電機株式会社 | Vehicle control device and vehicle control method |
| CN110488826A (en) * | 2019-08-20 | 2019-11-22 | 集美大学 | A kind of AGV collision prevention method, terminal device and storage medium suitable for path conflict |
| WO2021048966A1 (en) * | 2019-09-12 | 2021-03-18 | 三菱電機株式会社 | Drive assistance device, vehicle control device, drive assistance system, and drive assistance method |
Non-Patent Citations (6)
| Title |
|---|
| Concurrent Optimal Trajectory Planning for Indoor Quadrotor Formation Switching;Xu, Y等;《 JOURNAL OF INTELLIGENT & ROBOTIC SYSTEMS 》;20190531;全文 * |
| Emissions and light absorption of carbonaceous aerosols from on-road vehicles in an urban tunnel in south China;Runqi Zhang等;《Science of the Total Environment》;20210602;全文 * |
| Improved Surendra background updating algorithm for moving vehicle detection;Lan Weiyao 等;《2018 Chinese Control And Decision Conference (CCDC》;20181231;全文 * |
| Improved Surendra Background Updating Algorithm for Moving Vehicle Detection;Zhao Luyao等;《Department of Automation》;20181231;全文 * |
| 基于超声波的全自动平行泊车路径规划;肖祖铭等;《内燃机与配件》;20200730(第14期);全文 * |
| 无人驾驶智能车路径引导的研究;宋洁;《CNKI-优秀硕士论文》;20130715(第07期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113359760A (en) | 2021-09-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109724612B (en) | AGV path planning method and device based on topological map | |
| CN108762270B (en) | Improved path planning algorithm for variable probability bidirectional fast search random tree | |
| CN107203190B (en) | inertial navigation AGV scheduling method and system based on complex path | |
| Li et al. | Optimal control-based online motion planning for cooperative lane changes of connected and automated vehicles | |
| CN113085850B (en) | Vehicle obstacle avoidance method and device, electronic equipment and storage medium | |
| CN108262747B (en) | Motion trajectory planning method and device, robot and storage medium | |
| CN112595337B (en) | Obstacle avoidance path planning method and device, electronic device, vehicle and storage medium | |
| US20200042014A1 (en) | Coordination of paths of a plurality of movable machines | |
| CN113682312B (en) | Autonomous channel switching method and system integrating deep reinforcement learning | |
| Van Parys et al. | Online distributed motion planning for multi-vehicle systems | |
| CN111813099B (en) | Driving control method and device for unmanned vehicle, computer equipment and vehicle | |
| Feng et al. | Model-based decision making with imagination for autonomous parking | |
| CN113359760B (en) | Method for eliminating vehicle collision in optimal path algorithm operation result | |
| CN115755951A (en) | Unmanned aerial vehicle obstacle avoidance method for quickly recovering flight path | |
| CN114347982B (en) | Path planning method and system for automatic parking | |
| Sato et al. | Target-enclosing strategies for multi-agent using adaptive control strategy | |
| CN114840001A (en) | Multi-vehicle collaborative track planning method in closed environment | |
| Van Koevering et al. | Provable probabilistic safety and feasibility-assured control for autonomous vehicles using exponential control barrier functions | |
| CN114700944A (en) | Heterogeneous task-oriented double-robot collaborative path planning method | |
| CN113009922B (en) | Scheduling management method for robot walking path | |
| CN112706760A (en) | Unmanned parking path planning method for special road scene | |
| CN115903854B (en) | Automatic driving real-time track planning method for dynamic structured road | |
| Zhang et al. | An improved dynamic step size RRT algorithm in complex environments | |
| Wang et al. | Speed planning for autonomous driving in dynamic urban driving scenarios | |
| US20210362704A1 (en) | Valet parking system and valet parking method |
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 |