CN116907490A - Multi-robot path planning method based on conflict search - Google Patents
Multi-robot path planning method based on conflict search Download PDFInfo
- Publication number
- CN116907490A CN116907490A CN202310630398.6A CN202310630398A CN116907490A CN 116907490 A CN116907490 A CN 116907490A CN 202310630398 A CN202310630398 A CN 202310630398A CN 116907490 A CN116907490 A CN 116907490A
- Authority
- CN
- China
- Prior art keywords
- coordinate point
- robot
- path
- robots
- task
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000004364 calculation method Methods 0.000 claims abstract description 13
- 239000003016 pheromone Substances 0.000 claims description 11
- 241000257303 Hymenoptera Species 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 3
- 230000007704 transition Effects 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/20—Instruments for performing navigational calculations
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/20—Instruments for performing navigational calculations
- G01C21/206—Instruments for performing navigational calculations specially adapted for indoor navigation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Theoretical Computer Science (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Manipulator (AREA)
- Numerical Control (AREA)
Abstract
The invention belongs to the technical field of robot path planning, and particularly relates to a multi-robot path planning method based on conflict searching. According to the method, the ant colony algorithm is utilized to calculate the optimal order completion sequence of each task respectively, so that the total path of all orders contained in completing each task is shortest; planning the path of each robot by using an A-algorithm based on the optimal order completion sequence of each task, so that the calculation speed is greatly increased; storing the planned paths of the robots into a search tree, searching for conflicts of paths among the robots, and resolving the conflicts. The method solves the problems that the task is completed for too long or even can not be completed under the traditional multi-robot multi-order path planning method, and effectively solves the problems of lower path planning efficiency and tortuous path planning caused by collision and deadlock among robots in multi-robot multi-order path planning.
Description
Technical Field
The invention belongs to the technical field of robot path planning, and particularly relates to a multi-robot path planning method based on conflict searching.
Background
At present, with development and progress of technology, robots are increasingly used. Because the multi-robot system can better realize information and resource sharing, has higher parallelism and robustness, and can complete more complex tasks. As the number of robots in the same system increases, the problem of multi-robot scheduling becomes increasingly important.
Multiple robots are performing tasks within the same system, which may create collisions on the path of travel. Conflicts are classified into intersection conflicts and opposite conflicts. The intersection collision means that two robots arrive at an intersection at the same time, and the opposite collision means that the two robots travel in opposite directions.
Path planning refers to the steps of giving a starting point and a target point in an environment with a certain obstacle, referring to certain criteria, such as shortest path factors, least time, least cost and the like, and searching an optimal collision-free path in the environment.
The path planning of the multi-robot system is based on ensuring that a single robot completes the path planning from a starting point to a target point, so as to avoid collision between each robot and surrounding obstacles and avoid collision between each robot and other robots. The path planning of the multi-robot system can complete coordination among the multiple robots, avoid collision, deadlock and other conditions, and enable the multiple robots to complete specified tasks together and smoothly. However, the conventional multi-robot path planning method has the disadvantages that the time required for completing the path planning task is too long, the efficiency is low, the planned path is relatively tortuous, and the path planning task can not be completed sometimes.
Disclosure of Invention
The invention aims to provide a multi-robot path planning method based on conflict search, which utilizes an ant colony algorithm to calculate the optimal order completion sequence of each task respectively, so that the total path for completing all orders contained in each task is shortest, plans the path of each robot based on the optimal order completion sequence of each task by utilizing an A-type algorithm, stores the planned paths of each robot into a search tree, carries out conflict search on the paths among the robots, carries out conflict resolution, and effectively improves the overall path planning efficiency of the multi-robot multi-order.
In order to achieve the above purpose, the invention adopts the following technical scheme:
a multi-robot path planning method based on conflict search comprises the following steps:
s1, building a grid map according to the actual situation of a task place, and marking obstacles on the map to separate a feasible path of the robot;
s2, determining robots for completing tasks based on tasks to be distributed and current state information of the robots, and updating the state information of the robots, wherein each task to be distributed comprises a plurality of orders;
s3, calculating the shortest total path of all orders contained in each task by using an ant colony algorithm based on the feasible paths of the robots in the grid map obtained in the step S1 so as to obtain the optimal sequence of all orders contained in each task;
s4, constructing an empty search tree, and adding the starting position of each robot as a root node into the search tree, wherein the search tree has only one root node;
s5, planning a path of each robot for completing the assigned task by using an A-type algorithm based on the order optimal completion sequence of each task obtained in the step S3;
s6, placing the planned paths of the robots on a search tree, performing conflict search on the paths among the robots, returning to the step S5 if the paths among the robots have conflict, and completing overall planning of the paths of the multiple robots if the paths among the robots do not have conflict.
Preferably, step S3 includes:
s3.1, numbering positions of all orders contained in a task, taking an initial position of a robot as a starting point position of an ant colony, and setting initial pheromone concentration for the robot;
s3.2, each ant randomly selects a first order to be accessed, each ant selects the next order to be accessed according to probability after the access is finished, and when all ants access all orders, the pheromone concentration on the path of each ant is updated;
s3.3, circulating the step S3.2 until the paths of all orders contained in the task are the shortest, and stopping circulating to obtain the optimal sequence of all orders contained in the task;
the order completion optimal sequence of each task is obtained through calculation through the method.
Preferably, the pheromone concentration on the path of each ant is updated based on the length of the path of each ant.
Preferably, step S5 includes:
s5.1, selecting a task, inputting a task order into the robot according to the optimal completion sequence obtained in the step S3, and sequentially taking the task order as an access target of the robot;
s5.2, planning an access path of each time according to the position of the access target input each time;
s5.3, when the last order position in the optimal order sequence of the task output in the step S3 is used as a final access target to be searched, the path planning of the robot is completed;
s5.4, taking the last order position in the order sequence as a terminal, taking the starting position of the robot as a starting point, reversely searching the parent coordinates from the terminal, and continuously advancing each coordinate along the parent coordinates until returning to the starting point to obtain a path for recording specific path information;
the path of each robot is planned by the method;
the starting position of each robot and the position of each access target in the algorithm have corresponding coordinate points on the grid map.
Preferably, step S5.2 includes:
s5.2.1, creating an OPEN table and a CLOSE table for a coordinate point corresponding to the current starting position of the robot, adding the coordinate of the starting position of the robot into the OPEN table, and setting the actual cost of the starting coordinate point to be 0;
s5.2.2, adding a coordinate point adjacent to the initial coordinate point into the OPEN table, deleting the initial coordinate point from the OPEN table and adding the initial coordinate point into the CLOSE table;
s5.2.3 selecting a coordinate point with the minimum total cost from the OPEN table as a coordinate point to be processed currently, deleting the coordinate point to be processed currently from the OPEN table and adding the coordinate point to the CLOSE table, judging whether the coordinate point to be processed currently is a coordinate point corresponding to an input access target, if yes, ending the access, otherwise executing a step S5.2.4;
s5.2.4 expanding adjacent coordinate points of the coordinate points to be processed currently, judging each adjacent coordinate point, skipping the coordinate point if the adjacent coordinate point is in the CLOSE table, recording the actual cost, estimated cost and total cost of the coordinate point if the adjacent coordinate point is in the OPEN table, adding the coordinate point into the OPEN table and calculating the actual cost, estimated cost and total cost of the coordinate point if the adjacent coordinate point is not in the OPEN table, taking the coordinate point to be processed currently as the father coordinate point of the coordinate point, and updating the OPEN table and the CLOSE table;
s5.2.5, circularly executing steps S5.2.3-S5.2.4 until the coordinate point to be processed currently is the coordinate point corresponding to the input access target, and completing the path planning of the robot from the current starting position to the input access target;
each time an access path is planned using an a-algorithm, an OPEN table and a CLOSE table are created for the current path planning process.
Preferably, the calculation formula of the total cost of each coordinate point is as follows:
f(x,y)＝g(x,y)+h(x,y)，
wherein x represents the abscissa of the coordinate point on the grid map; y represents the ordinate of the coordinate point on the grid map; g (x, y) represents the actual cost value of a coordinate point, i.e., the Manhattan distance from the start of the current access path plan to the coordinate point; h (x, y) represents the estimated cost value of a coordinate point, i.e., the Manhattan distance from the coordinate point to the coordinate point corresponding to the current access target; f (x, y) represents the total cost value of the coordinate points;
the calculation formula of the Manhattan distance is as follows:
d(i,j)＝|x i -x j |+|y i -y j |，
wherein i and j respectively represent two coordinate points, x i 、y i Respectively representing the abscissa and the ordinate of the i coordinate point, and x j 、y j The abscissa and ordinate of the j-coordinate point are represented, respectively.
Preferably, the information of each coordinate point on each robot path is stored in an array or a linked list in sequence.
In a preferred embodiment, in step S6, the planned paths of the robots are all placed in a node on the search tree.
As a preferred scheme, judging whether paths among the robots have collision according to path information of the robots in the nodes, if the paths among the robots have collision, generating a new node on the search tree, adding new constraints for the robots with collision in the new node, returning to the step S5, re-planning the paths of the robots based on the newly added constraints, and placing the re-planned paths of the robots into the new node added with the new constraints.
Preferably, the new constraint is that the respective robot is not allowed to access the respective location in different time steps or that the plurality of robots are not allowed to access the respective location simultaneously in the respective time steps.
The beneficial effects of the invention are as follows:
according to the method, the ant colony algorithm is utilized to calculate the optimal order completion sequence of each task respectively, so that the total path of all orders contained in completing each task is shortest; planning the path of each robot by using an A-algorithm based on the optimal order completion sequence of each task, so that the calculation speed is greatly increased; storing the planned paths of the robots into a search tree, searching for conflicts of paths among the robots, and resolving the conflicts. The method solves the problems that the task is completed for too long or even can not be completed under the traditional multi-robot multi-order path planning method, and effectively solves the problems of lower path planning efficiency and tortuous path planning caused by collision and deadlock among robots in multi-robot multi-order path planning.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flow chart of a method of multi-robot path planning based on conflicting searches.
Fig. 2 is a flowchart of calculating the optimal order of all orders included in completing each task using the ant colony algorithm.
Fig. 3 is a flow chart of planning the path of each robot to complete an assigned task using an a-algorithm.
Detailed Description
The following specific examples are presented to illustrate the present invention, and those skilled in the art will readily appreciate the additional advantages and capabilities of the present invention as disclosed herein. The invention may be practiced or carried out in other embodiments that depart from the specific details, and the details of the present description may be modified or varied from the spirit and scope of the present invention. It should be noted that the following embodiments and features in the embodiments may be combined with each other without conflict.
Referring to fig. 1, a multi-robot path planning method based on collision search includes the steps of:
a multi-robot path planning method based on conflict search comprises the following steps:
s1, building a grid map according to the actual situation of the task place, and marking obstacles on the map to separate out a feasible path of the robot. The grid method uses grids with the same size to divide the environment space, and the resolution of the grid map, namely the size of the space in the actual environment represented by each grid, is defined according to specific scenes.
S2, determining robots which finish the tasks based on the tasks to be distributed and the current state information of the robots, and updating the state information of the robots, wherein each task to be distributed comprises a plurality of orders. According to the current state of each robot, the capacity of each robot, the priority of the task and other factors, the task can be distributed to the most suitable robot.
And S3, calculating the shortest total path of all orders contained in each task by using an ant colony algorithm based on the feasible paths of the robots in the grid map obtained in the step S1 so as to obtain the optimal sequence of all orders contained in each task.
S4, constructing an empty search tree, and adding the starting position of each robot as a root node into the search tree, wherein the search tree has only one root node.
And S5, planning a path for each robot to complete the assigned task by using an A-based algorithm based on the order optimal completion sequence of each task obtained in the step S3. Each robot completes a task, each task comprises a plurality of orders, an ant colony algorithm is utilized to calculate the optimal order completion sequence of each task, the total path for completing all orders contained in each task is shortest, and based on the optimal order completion sequence of each task, the path of each robot is planned by utilizing an A-type algorithm, so that the calculation speed of the A-type algorithm can be effectively improved, and the problems that the path planning efficiency is low and even planning cannot be completed can be effectively avoided.
S6, placing the planned paths of the robots on a search tree, performing conflict search on the paths among the robots, returning to the step S5 if the paths among the robots have conflict, and completing overall planning of the paths of the multiple robots if the paths among the robots do not have conflict.
And placing the planned paths of the robots into one node on the search tree, and judging whether the paths of the robots have conflict or not according to the path information of the robots in the node. This step allows for interaction and coordination among multiple robots to ensure that the overall path is optimal and free of collisions.
Further, referring to fig. 2, step S3 includes:
s3.1, numbering positions of all orders contained in a task, taking an initial position of a robot as a starting point position of an ant colony, and setting initial pheromone concentration for the robot;
s3.2, each ant randomly selects a first order to be accessed, each ant selects a next order to be accessed according to probability after the access is finished, when all ants access all orders, the pheromone concentration on the path of each ant is updated, and the pheromone concentration on the path of each ant is updated based on the length of the path of each ant;
s3.3, circulating the step S3.2 until the paths of all orders contained in the task are the shortest, and stopping circulating to obtain the optimal sequence of all orders contained in the task;
the order completion optimal sequence of each task is obtained through calculation through the method.
Specifically:
when the shortest total path for completing all orders contained in each task is calculated by using an ant colony algorithm, the number of ants selected by each task is 10 times of the number of orders contained in the task.
The routing of ants is based on factors such as pheromone concentration and distance. Specifically, when each ant arrives at a certain order, the strength and distance of pheromones between the current order and all the non-accessed orders are calculated, and then the next order to be accessed is selected in a roulette manner according to the transfer probability.
The calculation formula of the probability transition is as follows:
wherein i and j respectively represent two orders, τ i,j Information element quantity, eta, representing order i through order j i,j Heuristic factors representing orders i through j, alpha, beta representing two constants, N representing the aggregate of all orders, p i,j Representing the transition probabilities of order i to order j.
And normalizing the transition probability:
where k represents all orders that have not been accessed, p' i,j Representing the transition probabilities of normalized order i to order j.
Further, referring to fig. 3, step S5 includes:
s5.1, selecting a task, inputting a task order into the robot according to the optimal completion sequence obtained in the step S3, and sequentially taking the task order as an access target of the robot;
s5.2, planning an access path of each time according to the position of the access target input each time;
s5.3, when the last order position in the optimal order sequence of the task output in the step S3 is used as a final access target to be searched, the path planning of the robot is completed;
s5.4, taking the last order position in the order sequence as a terminal, taking the starting position of the robot as a starting point, reversely searching the parent coordinates from the terminal, and continuously advancing each coordinate along the parent coordinates until returning to the starting point to obtain a path for recording specific path information; the shortest path from the starting point to the final target point can be obtained by using an a-algorithm, but no specific path information is recorded on the path, so that the path of each order of each robot for completing the assigned task needs to be constructed by traversing reversely from the final target point to the starting point.
The path of each robot is planned by the method;
the starting position of each robot and the position of each access target in the algorithm have corresponding coordinate points on the grid map.
Further, step S5.2 includes:
s5.2.1, creating an OPEN table and a CLOSE table for a coordinate point corresponding to the current starting position of the robot, adding the coordinate of the starting position of the robot into the OPEN table, and setting the actual cost of the starting coordinate point to be 0;
s5.2.2, adding a coordinate point adjacent to the initial coordinate point into the OPEN table, deleting the initial coordinate point from the OPEN table and adding the initial coordinate point into the CLOSE table;
s5.2.3 selecting a coordinate point with the minimum total cost from the OPEN table as a coordinate point to be processed currently, deleting the coordinate point to be processed currently from the OPEN table and adding the coordinate point to the CLOSE table, judging whether the coordinate point to be processed currently is a coordinate point corresponding to an input access target, if yes, ending the access, otherwise executing a step S5.2.4;
s5.2.4 expanding adjacent coordinate points of the coordinate points to be processed currently, judging each adjacent coordinate point, skipping the coordinate point if the adjacent coordinate point is in the CLOSE table, recording the actual cost, estimated cost and total cost of the coordinate point if the adjacent coordinate point is in the OPEN table, adding the coordinate point into the OPEN table and calculating the actual cost, estimated cost and total cost of the coordinate point if the adjacent coordinate point is not in the OPEN table, taking the coordinate point to be processed currently as the father coordinate point of the coordinate point, and updating the OPEN table and the CLOSE table;
s5.2.5, circularly executing steps S5.2.3-S5.2.4 until the coordinate point to be processed currently is the coordinate point corresponding to the input access target, and completing the path planning of the robot from the current starting position to the input access target;
when planning each access path by using an A-algorithm, creating an OPEN table and a CLOSE table for the current path planning process;
the information for each coordinate point on each robot path is stored in an array or linked list in order. And on the basis of the optimal order completion sequence of each task, planning the path of each robot by using an A-algorithm, so that the calculation speed is greatly increased.
Further, the calculation formula of the total cost of each coordinate point is as follows:
f(x,y)＝g(x,y)+h(x,y)，
wherein x represents the abscissa of the coordinate point on the grid map; y represents the ordinate of the coordinate point on the grid map; g (x, y) represents the actual cost value of a coordinate point, i.e., the Manhattan distance from the start of the current access path plan to the coordinate point; h (x, y) represents the estimated cost value of a coordinate point, i.e., the Manhattan distance from the coordinate point to the coordinate point corresponding to the current access target; f (x, y) represents the total cost value of the coordinate points;
the calculation formula of the Manhattan distance is as follows:
d(i,j)＝|x i -x j |+|y i -y j |，
wherein i and j respectively represent two coordinate points, x i 、y i Respectively representing the abscissa and the ordinate of the i coordinate point, and x j 、y j The abscissa and ordinate of the j-coordinate point are represented, respectively.
Further, in step S6, the planned paths of the robots are all placed in one node on the search tree, and the node is a set of all paths and is a child node of the root node, and inherits the constraint of the root node. The set is provided with a plurality of matrixes, and the path planned by one robot is correspondingly stored in one matrix.
If the paths of the robots have conflicts, a new node is generated on the search tree, new constraints are added to the robots with the conflicts in the new node, step S5 is returned to, the paths of the robots are re-planned based on the newly added constraints, and the paths of the robots after re-planning are placed in the new node with the new constraints. The new node is a child node storing each robot path node after the previous planning.
The new constraint is that the respective robots are not allowed to access the respective locations in different time steps or that multiple robots are not allowed to access the respective locations simultaneously in the respective time steps.
Specifically:
conflicts are classified into point conflicts and edge conflicts.
If two robots arrive at the same coordinate point at the same time, then they are considered to have point collisions. If at two adjacent moments the start point coordinates of one robot and the end point coordinates of the other robot coincide, they are considered to have an edge collision.
The constraints of the new node will limit the robots that generate collisions when planning the new path. If there are two robots whose paths collide, one of which cannot appear at that time on the coordinate where the collision occurred, then the robot may stop passing the other robot first or select the other path in advance when planning the path.
The paths of the robots are planned again according to the added new constraint, so that the tasks of allocation of the robots can be completed while collision is avoided, the overall planning efficiency of the paths of the multiple robots and multiple orders can be effectively improved, and the problem that the paths of the robots are too tortuous is avoided.
The above examples are merely illustrative of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention, and various modifications and improvements made by those skilled in the art to the technical solution of the present invention should fall within the protection scope of the present invention without departing from the design spirit of the present invention.
Claims (10)
1. A multi-robot path planning method based on conflict search is characterized by comprising the following steps:
s1, building a grid map according to the actual situation of a task place, and marking obstacles on the map to separate a feasible path of the robot;
s2, determining robots for completing tasks based on tasks to be distributed and current state information of the robots, and updating the state information of the robots, wherein each task to be distributed comprises a plurality of orders;
s3, calculating the shortest total path of all orders contained in each task by using an ant colony algorithm based on the feasible paths of the robots in the grid map obtained in the step S1 so as to obtain the optimal sequence of all orders contained in each task;
s4, constructing an empty search tree, and adding the starting position of each robot as a root node into the search tree, wherein the search tree has only one root node;
s5, planning a path of each robot for completing the assigned task by using an A-type algorithm based on the order optimal completion sequence of each task obtained in the step S3;
s6, placing the planned paths of the robots on a search tree, performing conflict search on the paths among the robots, returning to the step S5 if the paths among the robots have conflict, and completing overall planning of the paths of the multiple robots if the paths among the robots do not have conflict.
2. The method for planning a path for multiple robots based on collision searching according to claim 1, wherein step S3 comprises:
s3.1, numbering positions of all orders contained in a task, taking an initial position of a robot as a starting point position of an ant colony, and setting initial pheromone concentration for the robot;
s3.2, each ant randomly selects a first order to be accessed, each ant selects the next order to be accessed according to probability after the access is finished, and when all ants access all orders, the pheromone concentration on the path of each ant is updated;
s3.3, circulating the step S3.2 until the paths of all orders contained in the task are the shortest, and stopping circulating to obtain the optimal sequence of all orders contained in the task;
the order completion optimal sequence of each task is obtained through calculation through the method.
3. The method of claim 2, wherein the pheromone concentration on each ant path is updated based on the length of each ant path.
4. The method for planning a path for multiple robots based on collision searching according to claim 2, wherein step S5 comprises:
s5.1, selecting a task, inputting a task order into the robot according to the optimal completion sequence obtained in the step S3, and sequentially taking the task order as an access target of the robot;
s5.2, planning an access path of each time according to the position of the access target input each time;
s5.3, when the last order position in the optimal order sequence of the task output in the step S3 is used as a final access target to be searched, the path planning of the robot is completed;
s5.4, taking the last order position in the order sequence as a terminal, taking the starting position of the robot as a starting point, reversely searching the parent coordinates from the terminal, and continuously advancing each coordinate along the parent coordinates until returning to the starting point to obtain a path for recording specific path information;
the path of each robot is planned by the method;
the starting position of each robot and the position of each access target in the algorithm have corresponding coordinate points on the grid map.
5. The method for multiple robot path planning based on collision search according to claim 4, wherein step S5.2 comprises:
s5.2.1, creating an OPEN table and a CLOSE table for a coordinate point corresponding to the current starting position of the robot, adding the coordinate of the starting position of the robot into the OPEN table, and setting the actual cost of the starting coordinate point to be 0;
s5.2.2, adding a coordinate point adjacent to the initial coordinate point into the OPEN table, deleting the initial coordinate point from the OPEN table and adding the initial coordinate point into the CLOSE table;
s5.2.3 selecting a coordinate point with the minimum total cost from the OPEN table as a coordinate point to be processed currently, deleting the coordinate point to be processed currently from the OPEN table and adding the coordinate point to the CLOSE table, judging whether the coordinate point to be processed currently is a coordinate point corresponding to an input access target, if yes, ending the access, otherwise executing a step S5.2.4;
s5.2.4 expanding adjacent coordinate points of the coordinate points to be processed currently, judging each adjacent coordinate point, skipping the coordinate point if the adjacent coordinate point is in the CLOSE table, recording the actual cost, estimated cost and total cost of the coordinate point if the adjacent coordinate point is in the OPEN table, adding the coordinate point into the OPEN table and calculating the actual cost, estimated cost and total cost of the coordinate point if the adjacent coordinate point is not in the OPEN table, taking the coordinate point to be processed currently as the father coordinate point of the coordinate point, and updating the OPEN table and the CLOSE table;
s5.2.5, circularly executing steps S5.2.3-S5.2.4 until the coordinate point to be processed currently is the coordinate point corresponding to the input access target, and completing the path planning of the robot from the current starting position to the input access target;
each time an access path is planned using an a-algorithm, an OPEN table and a CLOSE table are created for the current path planning process.
6. The method for planning a path for multiple robots based on collision search according to claim 5, wherein the total cost of each coordinate point is calculated as follows:
f(x,y)＝g(x,y)+h(x,y)，
wherein x represents the abscissa of the coordinate point on the grid map; y represents the ordinate of the coordinate point on the grid map; g (x, y) represents the actual cost value of a coordinate point, i.e., the Manhattan distance from the start of the current access path plan to the coordinate point; h (x, y) represents the estimated cost value of a coordinate point, i.e., the Manhattan distance from the coordinate point to the coordinate point corresponding to the current access target; f (x, y) represents the total cost value of the coordinate points;
the calculation formula of the Manhattan distance is as follows:
d(i,j)＝|x i -x j |+|y i -y j |，
wherein i and j respectively represent two coordinate points, x i 、y i Respectively representing the abscissa and the ordinate of the i coordinate point, and x j 、y j The abscissa and ordinate of the j-coordinate point are represented, respectively.
7. The method of claim 6, wherein the information of each coordinate point on each robot path is sequentially stored in an array or linked list.
8. The method for planning paths for multiple robots based on collision search according to claim 1, wherein in step S6, the planned paths of each robot are placed in a node on a search tree.
9. The method for planning paths of multiple robots based on collision search according to claim 8, wherein whether paths between robots have collision is determined according to path information of each robot in the nodes, if the paths between the robots have collision, a new node is generated on the search tree, new constraints are added to the robot having collision in the new node, the path of each robot is re-planned based on the newly added constraints, and the path of each robot after re-planning is put into the new node to which the new constraints are added.
10. A multi-robot path planning method based on conflicting searches according to claim 9, characterized in that the new constraint is that no respective robots are allowed to access respective locations in different time steps or that multiple robots are not allowed to access respective locations simultaneously in respective time steps.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310630398.6A CN116907490A (en) | 2023-05-31 | 2023-05-31 | Multi-robot path planning method based on conflict search |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310630398.6A CN116907490A (en) | 2023-05-31 | 2023-05-31 | Multi-robot path planning method based on conflict search |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116907490A true CN116907490A (en) | 2023-10-20 |
Family
ID=88357144
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310630398.6A Pending CN116907490A (en) | 2023-05-31 | 2023-05-31 | Multi-robot path planning method based on conflict search |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116907490A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118129763A (en) * | 2024-05-07 | 2024-06-04 | 浙江大学 | Multi-robot path planning method based on conflict search under uncertain time |
-
2023
- 2023-05-31 CN CN202310630398.6A patent/CN116907490A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118129763A (en) * | 2024-05-07 | 2024-06-04 | 浙江大学 | Multi-robot path planning method based on conflict search under uncertain time |
| CN118129763B (en) * | 2024-05-07 | 2024-07-23 | 浙江大学 | Multi-robot path planning method based on conflict search under uncertain time |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109115226B (en) | Route planning method for avoiding multi-robot conflict based on jumping point search | |
| CN109839110B (en) | Multi-target point path planning method based on rapid random search tree | |
| CN113031603B (en) | Task priority based collaborative path planning method for multiple logistics robots | |
| CN114510056B (en) | Method for planning steady moving global path of indoor mobile robot | |
| CN103413209B (en) | Many client many warehouses logistics distribution routing resources | |
| US10365110B2 (en) | Method and system for determining a path of an object for moving from a starting state to an end state set avoiding one or more obstacles | |
| Fraigniaud et al. | Collective tree exploration | |
| CN111708364B (en) | AGV path planning method based on A-algorithm improvement | |
| CN115237135A (en) | Mobile robot path planning method and system based on conflict | |
| CN111611274A (en) | Database query optimization method and system | |
| CN109974711A (en) | A kind of AGV multiple target point autonomous navigation method towards wisdom factory | |
| CN114815802A (en) | Unmanned overhead traveling crane path planning method and system based on improved ant colony algorithm | |
| CN115454070B (en) | K-Means ant colony algorithm multi-robot path planning method | |
| CN116907490A (en) | Multi-robot path planning method based on conflict search | |
| CN112327890A (en) | Underwater multi-robot path planning based on WHCA algorithm | |
| CN112836974A (en) | DQN and MCTS based box-to-box inter-zone multi-field bridge dynamic scheduling method | |
| CN112819211A (en) | Multi-region scheduling route planning method based on ant colony iterative algorithm | |
| CN114625137A (en) | AGV-based intelligent parking path planning method and system | |
| Tang et al. | Solving multi-agent target assignment and path finding with a single constraint tree | |
| Zhou et al. | A multi-AGV fast path planning method based on improved CBS algorithm in workshops | |
| CN117170381A (en) | Path planning method for multi-four-way shuttle | |
| Riman et al. | A Priority-based Modified A∗ Path Planning Algorithm for Multi-Mobile Robot Navigation | |
| CN112862212B (en) | Multi-AGV (automatic guided vehicle) scheduling method, device and equipment based on improved sparrow search algorithm | |
| CN115981322A (en) | Multi-agent path planning method, device, equipment and medium in dynamic environment | |
| Haiming et al. | Algorithm of path planning based on time window for multiple mobile robots in warehousing system |
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 |