CN112327829A - Distributed multi-robot cooperative motion control method, system, medium and application - Google Patents
Distributed multi-robot cooperative motion control method, system, medium and application Download PDFInfo
- Publication number
- CN112327829A CN112327829A CN202011103507.1A CN202011103507A CN112327829A CN 112327829 A CN112327829 A CN 112327829A CN 202011103507 A CN202011103507 A CN 202011103507A CN 112327829 A CN112327829 A CN 112327829A
- Authority
- CN
- China
- Prior art keywords
- robot
- path
- rob
- local
- planning
- 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
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0287—Control of position or course in two dimensions specially adapted to land vehicles involving a plurality of land vehicles, e.g. fleet or convoy travelling
- G05D1/0291—Fleet control
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0287—Control of position or course in two dimensions specially adapted to land vehicles involving a plurality of land vehicles, e.g. fleet or convoy travelling
- G05D1/0289—Control of position or course in two dimensions specially adapted to land vehicles involving a plurality of land vehicles, e.g. fleet or convoy travelling with means for avoiding collisions between vehicles
Abstract
The invention belongs to the technical field of cooperative control, and discloses a distributed multi-robot cooperative motion control method, a system, a medium and application, wherein each robot plans a local optimal path by using local information on the premise of not considering other robots; and collision avoidance between the robot and other robots detected by the vehicle-mounted sensors is realized by utilizing a communication and coordination strategy. The distributed multi-robot cooperative motion control system comprises: the first planning module is used for planning a local optimal path on the premise of not considering other robots by using local information; and the second planning module is used for realizing collision avoidance between the robot and other robots detected by the vehicle-mounted sensors by utilizing a communication and coordination strategy. The invention is suitable for cooperative motion control of a limited plurality of robots. The robot cooperative control tasks are as follows: both robots avoid static obstacles and other robots in real time while ultimately reaching the target position with minimal expense.
Description
Technical Field
The invention belongs to the technical field of cooperative control, and particularly relates to a distributed multi-robot cooperative motion control method, system, medium and application.
Background
At present: with the development of science and technology, the application of mobile robots is becoming more common, and mobile robot systems that can operate in various environments, such as Unmanned aircrafts (Unmanned aerial vehicles), Unmanned vehicles (Unmanned Ground vehicles), Unmanned undersea vehicles (Unmanned undersea vehicles), and the like, have been developed. Meanwhile, with the progress of industrialization, many fields such as automatic manufacturing, flexible production, search and rescue, environmental monitoring, safety and health face a large number of tasks with complex operations and large scale, and a single robot cannot well complete the tasks. Compared with a single robot system, the multi-robot system has a series of remarkable advantages due to the mutual cooperation of a plurality of robots, for example, the multi-robot system can reduce the complexity of task solution, promote the high efficiency of task completion, increase the reliability of the system, simplify the design of the system and the like. Due to these excellent characteristics, the multi-robot system is receiving more and more attention, and has attracted a great deal of study by scholars.
Most researches in the prior art assume that global environment information is known, however, in an actual system, a general robot can only know the environment information within the range of a sensor, and static obstacles or dynamic obstacles are likely to be newly added in the environment at any time. Therefore, the proposed offline path planning algorithm with knowledge of global environment information is not suitable for practical systems.
Through the above analysis, the problems and defects of the prior art are as follows: most of the current researches are offline path planning methods proposed on the basis of assuming known global information, and in an actual system, a general robot can only know environmental information within a sensor range, and a static obstacle or a dynamic obstacle is likely to be newly added in the environment at any time, so that the method is not suitable for the actual system.
The difficulty in solving the above problems and defects is: the planning needs to collect environment data, the dynamic updating of the environment model can be corrected at any time, the modeling and the searching of the environment are integrated, the robot system is required to have high-speed information processing capacity and computing capacity, high robustness on environment errors and noise is required, and real-time feedback and correction can be performed on the planning result.
The significance of solving the problems and the defects is as follows: the provided multi-mobile-robot coordinated quadratic programming method based on the rolling window fully utilizes the local environment measured by each robot in real time, and performs online programming in a rolling mode, so that the online calculation amount is greatly reduced, and the system has strong robustness.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a distributed multi-robot cooperative motion control method, a distributed multi-robot cooperative motion control system, a distributed multi-robot cooperative motion control medium and application.
The invention is realized in such a way that a distributed multi-robot cooperative motion control method comprises the following steps:
each robot plans a local optimal path by using local information on the premise of not considering other robots;
and collision avoidance between the robot and other robots detected by the vehicle-mounted sensors is realized by utilizing a communication and coordination strategy.
Further, the robot Rob of the distributed multi-robot cooperative motion control methodiThe rolling planning algorithm comprises the following specific steps:
step one, initializing a starting point SiEnd point GiThe working environment WS, the view radius R and the robot step length delta a, wherein the delta a is less than or equal to R;
step two, refreshing robot RobiIS information set ofi(tk) That is, the robot RobiThe view field or the rolling window is used for acquiring the environmental information in the sensing range of the current robot;
thirdly, selecting local sub-targets on the premise of not considering other robots and dynamic obstacles: if G isi∈ISi(tk) Let Psubi(tk)＝GiOtherwise, to
And P issubi(tk) The sub-target mapping rule is met;
refers to the boundary of the field of view, FD(tk) Refers to a feasible point set in the robot working environment WS;
step four, according to A*Algorithm in ISi(tk) In-computation local optimal path FPi(PRi(tk)，Psubi(tk))；
Step five, whether the sensor detects other robots RobjIf yes, Step6 is executed in sequence; if not, turning to Step 8;
step six, estimating Rob of robotiAnd RobjIf so, executing the step seven in sequence; if not, turning to the step eight;
step seven, collision avoidance is carried out through an optimization coordination strategy to obtain a newly planned local path FP'i(PRi(tk)，Psubi(tk) Let FPi(PRi(tk)，Psubi(tk))＝FP′i(PRi(tk)，Psubi(tk) Updating the planned local path;
step eight, the robot follows the planned local path FPi(PRi(tk)，Psubi(tk) One step with a step size of ε, 0<ε<R；
If PRi(tk)＝GiWhen the target is reached, quitting; otherwise, k is k +1, and step two is carried out.
Further, the time for the robot to travel one step length epsilon from the current position according to the planned local path is delta t, firstly, the robot RiThe vehicle-mounted sensor detects the robot RjRobot RjR is also detectedi。RiAnd RjCommunicate with each other, exchange respective local paths, and determine RiAnd RjIf the local paths have conflict, the problem of predicting the relative positions of the two tracks is converted into the judgment of whether the 2 track point sets have intersection; the specific process is as follows:
(1) if the two tracks do not intersectSet, then robot R within Δ t timeiAnd robot RjThe robot continues to move forward according to the original path without collision;
(2) if the path tracks of the two paths have overlapped road sections, the two paths may run in the same direction on the overlapped road sections, so that a frontal collision occurs, and a local path coordination correction strategy is started;
(3) if the path tracks of the robot R and the robot R have 1 or more intersection points and the robot R move laterally and generate side collision, the robot R is randomly selectediOr RjA pause strategy is initiated. The pause time is a rolling period, namely the time T required by the robot to move by a step length epsilon, and the other robot continues to move according to the original plan.
Further, the collision avoidance strategy for various prediction situations of the distributed multi-robot cooperative motion control method comprises:
(1) if the situation that the front collision is possible is predicted, starting a local path correction coordination strategy;
(2) if the situation that side collision is about to occur is predicted, the robot only needs to wait for the time T required by moving one step length epsilon in situ and then moves according to the original planned path;
(3) and if the robot is predicted not to collide with the other robot, directly traveling according to the original planned path.
Further, the basic idea of modifying the local path coordination strategy is as follows: firstly, planning a local optimal path by using local information on the premise of not considering other robots and dynamic obstacles, and recording the planning performed at the moment as the 1 st planning; then, the robot Rob is predictediAnd RobjWhen positive conflict possibly occurs, planning a new local optimal path by each robot under the condition of considering the opposite side, determining which robot travels along the new path by comparing the influence of the new local path on the global objective function, and recording the planning performed at the moment as the 2 nd planning; robiIs longer than the original path by delta i, RobjIs longer than the original path by Δ j. If Δ i < Δ j, RobiSelecting a new path, RobjProceeding along the originally planned path; if Δ i > Δ j, RobjSelecting a new path, RobiTravel along the originally planned path; if Δ i ═ Δ j, Rob is chosen randomlyiOr RobjTravel along the newly planned path.
In the collision avoidance process, if the robot RiAnd robot RjIf the optimal track from the current position to the sub-target position cannot be planned, setting the track increment to be + ∞, and then carrying out R operation on the robotiAnd robot RjComparing the trace increments of (a); if the two robots can not plan the optimal track from the current position to the sub-target position, one robot is randomly selected, the robot is retreated by a step length epsilon from the current position, information interaction is carried out with other robots to ensure that the retreating operation can not collide with other robots, and the other robot continues to advance along the original track.
Further comprising: rob learning by information interaction between robotsjCurrently planned local path, predicting robot RobiAnd RobjPath conflict of tkThe coordination strategy of the time is as follows:
the method comprises the following steps: at robot RiIS ofi(tk) In (1), the robot RjPropagates along its path as a static obstacle, giving IS'i(tk)；
Step two: according to A*Algorithm IS'i(tk) Calculating a local optimal path FP'ki(PRi(tk)，P′subi(tk) There may be two cases for this step: FP 'can be planned out'ki(PRi(tk)，P′subi(tk) Let Δ i be FP'ki(PRi(tk)，P′subi(tk))-FPki(PRi(tk)，Psubi(tk) FP) hereki(PRi(tk)，Psubi(tk) And FP'ki(PRi(tk)，P′subi(tk) Respectively, the local optimal paths planned by the layer 1 planning and the layer 2 planning respectively correspond to the P according to the sub-target mapping rulesubi(tk)、P′subi(tk)；
No regulations marking FP'ki(PRi(tk)，P′subi(tk))；
Step three: with robot RobjCommunication, knowing FP'kj(PRj(tk)，P′subj(tk) Whether it is programmable;
step four: according to RobiAnd RobjAnd (3) determining a coordination result according to the planning condition of the new local path:
if RobiNew local paths can be planned, and RobjNot, then RobiStep size epsilon of travel along new local pathiInstant FPki(PRi(tk)，Psubi(tk))＝FP′ki(PRi(tk)，P′subi(tk))；
If RobjNew local paths can be planned, and RobiNot, then RobiThe step length epsilon is finished along the original local pathi。
If RobiAnd RobjA new local path can be planned, comparing Δ i and Δ j:
if Δ i < Δ j, RobiStep size epsilon of travel along new local pathiInstant FPki(PRi(tk)Psubi(tk))＝FP′ki(PRi(tk)，P′subi(tk))；
If delta i is equal to delta j, then a random number RandomNum is randomly selected from [0, 1 ], if 0.5 ≦ RandomNum < 1, then RobiStep size ε i is traveled along the original local path, otherwise, RobiStep size epsilon of travel along new local pathiInstant FPki(PRi(tk)，Psubi(tk))＝FP′ki(PRi(tk)，P′subi(tk))；
If RobiAnd RobjIf no new local path can be planned, d (P) is comparedRi(tk)，Gi) And d (P)Rj(tk)，Gj)：
If d (P)Ri(tk)，Gi)＜d(PRj(tk)，Gj) Then RobiBack by one step epsilon along the path traversedi；
If d (P)Ri(tk)Gi)＝d(PRj(tk)，Gj) Randomly selecting a number RandomNum in [0, 1), if the RandomNum is more than or equal to 0 and less than 0.5, then RobiBack by one step epsilon along the path traversedi。
Further, after the local target is determined, the local path planning problem in the rolling window is summarized as P under the condition of known environment informationR(t) to Psub(t) planning problem, using A*Solving the planning problem by an algorithm;
the path planning algorithm flow based on the rolling window is as follows:
step one, aligning with a starting point PsEnd point PgInitializing a working space Wspace, a vision radius R of the robot and a step length epsilon;
step two, if the target point is reached, the planning is terminated;
step three, updating the map information in the current rolling window;
fourthly, generating a local optimization sub-target P according to the sub-target mapping rulesub(t)；
Step five, according to the local sub-target Psub(t) planning a local path in the current window with the known environment information;
step six, moving forward one step according to the local path, wherein the step length is epsilon, and epsilon is more than 0 and less than or equal to R;
step seven, returning to the step two;
the rolling path planning based on the A-algorithm comprises the following steps:
step one, initializing a starting point PinitAnd finallyPoint P _ { good }, field of view radius R of working environment WS, Rob;
step two, if PgoalE.g. IS (t), let local sub-target Psub(t)＝PgoalOtherwise, determining the sub-targets by the above principle;
step three, planning a slave P by using A-star algorithmR(t) to Psub(t) the optimal path;
step four, if Psub(t)＝PgoalLet Rob move to Psub(t), finishing planning and quitting; otherwise, let Rob follow the planned path to Psub(t) move when P is satisfiedR(t+ε)＝Psub(t), turning to the fifth step;
step five, refreshing the content of IS (t), and turning to step two.
It is another object of the present invention to provide a computer-readable storage medium storing a computer program which, when executed by a processor, causes the processor to perform the steps of:
each robot plans a local optimal path by using local information on the premise of not considering other robots;
and collision avoidance between the robot and other robots detected by the vehicle-mounted sensors is realized by utilizing a communication and coordination strategy.
Another object of the present invention is to provide a distributed multi-robot cooperative motion control system for implementing the distributed multi-robot cooperative motion control method, the distributed multi-robot cooperative motion control system including:
the first planning module is used for planning a local optimal path on the premise of not considering other robots by using local information;
and the second planning module is used for realizing collision avoidance between the robot and other robots detected by the vehicle-mounted sensors by utilizing a communication and coordination strategy.
Another object of the present invention is to provide a mobile robot system that operates the distributed multi-robot cooperative motion control method, the mobile robot system including: an unmanned aircraft, an unmanned automobile, or an unmanned underwater vehicle.
By combining all the technical schemes, the invention has the advantages and positive effects that: the invention aims to provide a multi-mobile-robot path planning method based on a rolling window. The method is suitable for cooperative motion control of a limited plurality of robots. The robot cooperative control tasks are as follows: both robots avoid static obstacles and other robots in real time while ultimately reaching the target position with minimal expense.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments of the present application will be briefly described below, and it is obvious that the drawings described below are only some embodiments of the present application, and it is obvious for those skilled in the art that other drawings can be obtained from the drawings without creative efforts.
Fig. 1 is a flowchart of a distributed multi-robot cooperative motion control method according to an embodiment of the present invention.
FIG. 2 is a schematic structural diagram of a distributed multi-robot cooperative motion control system according to an embodiment of the present invention;
in fig. 2: 1. a first planning module; 2. a second planning module.
Fig. 3 is a schematic diagram of a simulation result of a single robot system according to an embodiment of the present invention.
Fig. 4 is a schematic diagram of relative motions of a Robot _ i and a Robot _ j provided in the embodiment of the present invention.
Fig. 5 is a flowchart of a rolling plan according to an embodiment of the present invention.
FIG. 6(a) is a graph showing t according to an embodiment of the present inventionkSchematic view of a local plan of time of day.
FIG. 6(b) is a graph of t provided by an embodiment of the present inventionkSchematic collaboration of time of day.
FIG. 6(c) is a graph of t provided by an embodiment of the present inventionk+1Schematic diagram of the planning of the time of day.
FIG. 6(d) is a graph of t provided by an embodiment of the present inventionk+2Local planning of time of daySchematic representation.
Fig. 6(e) -6 (g) are process diagrams for suspending coordination policy execution according to the embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
In view of the problems in the prior art, the present invention provides a distributed multi-robot cooperative motion control method, system, medium and application, and the present invention is described in detail below with reference to the accompanying drawings.
As shown in fig. 1, the distributed multi-robot cooperative motion control method provided by the present invention includes the following steps:
s101: each robot plans a local optimal path (first planning) by using local information on the premise of not considering other robots;
s102: and (4) collision avoidance (secondary planning) between the robot and other robots detected by the vehicle-mounted sensors is realized by utilizing a communication and coordination strategy.
Those skilled in the art can also implement the distributed multi-robot cooperative motion control method provided by the present invention by using other steps, and the distributed multi-robot cooperative motion control method provided by the present invention in fig. 1 is only a specific embodiment.
As shown in fig. 2, the distributed multi-robot cooperative motion control system provided by the present invention includes:
the first planning module 1 is used for planning a local optimal path on the premise of not considering other robots by using local information.
And the second planning module 2 is used for realizing collision avoidance between the robot and other robots detected by the vehicle-mounted sensors by utilizing a communication and coordination strategy.
The technical solution of the present invention is further described below with reference to the accompanying drawings.
1. The multi-mobile robot path planning problem is defined as follows:
for each robot RobiFinding a path connecting their start and end points requires no collisions between the robot and environmental obstacles, other robots, and requires the following global objective function to be minimal:
in the formula, Ti(i is belonged to 1, 2.. multidot.N) is a robot RiThe time spent moving from the starting point to the end point; i isiIs a robot RiSum of pause times; r is1，r2For weighting, the weight is determined according to the requirements of people on pause time and total moving time, and r is increased if the robot is required to be not paused as much as possible2Otherwise, the opposite is true.
Since each robot moves at a uniform speed at the same rate, equation (1-1) can be transformed into
Wherein L is1，L2，...，LNThe path lengths of the robots are respectively; since the robot is moving at a constant speed, the pause time can be assigned to the path length, i.e./i＝Vi×Ii，ViIs the travel speed of each robot, IiIs the time each robot pauses in place; r is1，r2Is the weight of the objective function, r is more than or equal to 01，r2R is less than or equal to 11+r2The weight is 1, and is determined according to the requirements of people on pause time and overall moving time: if more emphasis is placed on optimizing the path length of the robot, r is increased1(ii) a If the robot is required not to pause as much as possible, r is increased2The value of (c).
2. Real-time path planning in two-dimensional unknown environments
2.1 route planning Algorithm based on Rolling Window
The path planning algorithm flow based on the rolling window is as follows:
step1, starting point PsEnd point PgInitializing a working space Wspace, a vision radius R of the robot and a step length epsilon;
step2, if the target point is reached, the planning is terminated;
step3, updating the map information in the current rolling window;
step4, generating local optimization sub-target P according to the sub-target mapping rulesub(t)；
Step5, according to local sub-target Psub(t) planning a local path in the current window with the known environment information;
step6, moving forward one Step according to the local path, wherein the Step length is epsilon (0< epsilon < R);
step7, return to Step 2.
2.2 sub-target mapping rules
In local planning based on rolling windows, due to the global final target PgoalNot necessarily contained within the rolling window, a sub-target of the local plan, which may be considered as P, has to be selectedgoalMapping within a rolling window. The local sub-targets are generated in the following way: at time t, the robot Rob has a scroll window Win (P)R(t)) having a boundary of
If Pgoal∈Win(PR(t)), then P is takensub(t)＝Pgoal(ii) a Otherwise, using heuristic function f (P) g (P) h (P) to select window boundary point P with minimum f (P) as sub-target Psub(t) that is
Wherein g (P) is the cost of the robot moving from the current position to the window boundary P, and the value of g (P) can be estimated according to the position of P and the environmental information in the current window; h (P) represents the generation of movement from the P node to the target pointPrice, since the information outside the rolling window is unknown, order
For simplicity, let g (P) be 0, + ∞, i.e. if P is in free space WfreeIf g (p) ═ 0, otherwise g (p) +∞. The simplified sub-target mapping is determined by formula (2-2), and any one of the sub-targets satisfying the condition can be selected if the calculation is obtained. The mapping rules embody a tradeoff of limited local information constraints with global optimization objectives.
The formula (2-2) represents that the node closest to the target node is selected as the child target point.
The child target point is not set to finally reach the point, but to feedback the position information of the global target point, and ensure the global optimization of the obtained path.
2.3 local Path planning
After the local target is determined, the invention resolves the local path planning problem in the rolling window into P-th order under the condition of known environment informationR(t) to Psub(t) planning problem. The invention adopts A*The algorithm solves this planning problem. A. the*The specific steps of the algorithm are not described in detail here.
2.4 triggering of local planning
The triggering of the local planning, that is, how much distance the mobile robot travels each time and then the next step of the rolling planning, is also one of the important factors to be considered in the path planning.
The idea is that after a local path is obtained every time and a fixed step length is advanced, the environmental information in a rolling window of the robot is refreshed, and then the local planning of the next round is carried out. The strategy has the characteristics that the smaller the step length of each travel, the stronger the sudden change adaptability to the environment, but meanwhile, for the fixed starting point and the fixed end point, the more the number of local planning times is, and the total planning efficiency is low.
2.5 local Path search Algorithm
A real-time heuristic algorithm: sort (near)k)
Eight point coordinates around the current position of the robot are added to near _ k, where k represents the robot number (k is 1,2,3,4)
Inputting: near _ k
And (3) outputting: sorted near _ k
ifnext_node_k＝U or D or L or R:
G＝10
ifnext_node_k＝RU or RD orLD or LU:
G＝14
# G is the cost of moving from the current state to the next state
# H is the predicted cost to the end point from the next state
H＝abs(end_k.row-next_node_k.row)*10+abs(end_k.col-next_node_k.col)*10，k＝1,2,3,4
F＝G+H
And sorting from large to small according to the F value.
A*The evaluation function of the search algorithm is defined as f (n) ═ g (n) + h (n), wherein g (n) is the actual cost from the node where the robot is located to the node n; h (n) is the estimated cost of the best path from node n to the target node, A*The heuristic of the algorithm search is mainly embodied in h (n), which is called a heuristic function. At the beginning of planning, two lists, namely an open list (open list) and a close list (close list) are established, and then the selection of the extended nodes in the open list is guided through an evaluation function f (n).
The feasible path is described as a grid sequence from a starting point S to a target point G, and adjacent squares are searched from S according to the following steps and are expanded to the target point G step by step.
Step1, establishing an Open list and a Close list, wherein the initial states are all empty lists;
step2, add S to the Open list, and search for 8 square nodes adjacent to S. If the square is in a pass state, it is added to the Open list, and the parent node of the 8 square nodes is defined as S. The present invention points "→" to a node and its parent in the drawing. The parent node is used for the last backtracking path.
Step3, remove S from the Open list table and add to the Close list table.
Step4, introducing an evaluation function f (n) ═ g (n) + h (n), and the candidate node is the existing node in the Open list. The node most likely to lead to the target point (the node with the smallest evaluation function value) is found by the evaluation function f (n).
Step5, removing the current node (the node found by the evaluation function in the last Step) from the Open list, and adding the current node into the Close list;
step6, search for 8 nodes adjacent to the current square, ignore the impassable squares and the squares already stored in the Close list. For the other grids:
if the grid does not exist in the Open list, it is added to the Open list, and its parent node is pointed to the current grid, and the estimated function value f (n) of this grid is calculated.
If this node already exists in the Open list, then a new g' (n) is computed with the current square as its parent. If g '(n) < g (n), then g (n) is replaced by new g' (n), and the value of f (n) is updated. And pointing the father node of the node as the current grid; otherwise, the original state is kept.
The path generation process is to repeatedly select the grid node with the minimum estimation function f (n) from the Open list, update the list and update the function value. g (n) can be defined as: g (n → parent) + Δ g. That is, the actual cost from the starting point S to the node n may be equivalent to the cost of moving from the parent node of the n node (n → parent) to n plus the cost from the starting point S to n → parent. In the algorithm, the node n and the parent node n → parent are certainly adjacent grids, so the value of the delta g is fixed. Now, it is defined that the distance between two adjacent grids along the coordinate axis direction is 10, and Δ g is 10 or 14.
Let h*(n) is the actual cost of the best path from node n to target point G, which is not accurately predictable. h (n) represents the estimated cost of the best path from node n to target point G. The environment map is an 8-neighborhood grid map, and the algorithm example introduces Manhattan distanceAnd performing estimation comparison. The estimated cost of moving from node n to target point G is defined as the manhattan distance between the two nodes: h (n) ═ xn-xg|+|yn-yg|). 10. In the formula (x)n，yn) Is n node coordinates, (x)g，yg) Is the target point G coordinate.
2.6 rolling path planning based on A-x algorithm
Step1, initializing a starting point (P)init) End point (P _ { good }), work environment (WS), and viewing radius (R) of Rob;
step2, if PgoalE.g. IS (t), let local sub-target Psub(t)＝PgoalOtherwise, determining the sub-targets by the above principle;
step3, planning a slave P by using A-star algorithmR(t) to Psub(t) the optimal path;
step4, if Psub(t)＝PgoalLet Rob move to Psub(t), finishing planning and quitting; otherwise, let Rob follow the planned path to Psub(t) move when satisfying
PR(t+ε)＝Psub(t), go to step 5;
step5, refreshing the content of IS (t), and turning to step 2.
The simulation result of the single robot system is shown in fig. 3.
3. A distributed secondary rolling algorithm for path planning of multiple mobile robots refers to the achievement of path planning of a single robot under the condition that a static environment is unknown. Various path planning algorithms may be combined with a rolling strategy, such as A*And an algorithm, an ant colony algorithm and the like plan a suboptimal path for the robot.
In the multi-robot distributed rolling path planning algorithm, each robot firstly uses local information to plan a local optimal path (first planning) on the premise of not considering other robots. Then, collision avoidance between the robot and other robots detected by the vehicle-mounted sensors is achieved by using communication and coordination strategies (second planning)
Algorithm 2.1 Multi-robot Rolling planning Algorithm:
rob robotiThe rolling planning algorithm is described as follows:
step1, initialization starting point SiEnd point GiThe working environment WS, the view radius R and the robot step length delta a (delta a is less than or equal to R);
step2, refresh robot RobiIS information set ofi(tk) That is, the robot RobiThe current robot sensing range is obtained;
step3, under the premise of not considering other robots and dynamic obstacles, selecting local sub-targets: if G isi∈ISi(tk) Let Psubi(tk)＝GiOtherwise, to
And P issubi(tk) The sub-target mapping rule is met; here, the
Refers to the boundary of the field of view, FD (t)k) Refers to the set of feasible points in the robot work environment WS.
Step4, according to A*Algorithm in ISi(tk) In-computation local optimal path FPi(PRi(tk)，Psubi(tk))；
Step5, whether the sensor detects other robots RobjIf yes, Step6 is executed in sequence; if not, turning to Step 8;
step6, estimating robot RobiAnd RobjIf there is a collision between the two, namely the predicted collision, if yes, Step7 is executed in sequence; if not, turning to Step 8;
step7, collision avoidance is carried out by an optimization coordination strategy to obtain a newly planned local path FP'i(PRi(tk)，Psubi(tk) Let FPi(PRi(tk)，Psubi(tk))＝FP′i(PRi(tk)，Psubi(tk) Updating the planned local path;
step8, local path FP according to planning of roboti(PRi(tk)，Psubi(tk) One step with a step size of ε (0)<ε<R)；
If PRi(tk)＝GiWhen the target is reached, quitting; otherwise, k is k +1, go to Step 2.
4. Robot-to-robot collision scenario prediction (predicted collision)
In a dynamic environment, the robot must not only bypass stationary obstacles, but also avoid other moving robots. The following exemplary scenarios are used to analyze the relative movements of the robot and the robot. As shown in fig. 4, where the circle represents a robot. Let the time for the robot to travel one step length epsilon from the current position according to the planned local path be delta t, first, the robot RiThe vehicle-mounted sensor detects the robot RjRobot RjR is also detectedi。RiAnd RjTo exchange respective local paths. From this, R can be judgediAnd RjWhether there is a conflict with the local path of (2). Therefore, the problem of predicting the relative positions of the two tracks is converted into the judgment of whether the 2 track point sets have intersection. The specific process is as follows:
(1) if the two trajectories do not intersect with each other, as shown in fig. 4(a), 4(b), and 4(e), the robot R will be operated within Δ t timeiAnd robot RjNo collision occurs. The robot continues to advance according to the original path;
(2) if there are overlapping links in the path tracks of the two, there is a possibility that the two travel in the overlapping links, and a frontal collision may occur, as shown in fig. 4 (f). Starting a modified local path coordination strategy;
(3) if the path trajectories of the two have 1 intersection or more and the two move laterally, a side collision occurs, as shown in fig. 4(c) and 4 (d). At this time, the robot R is randomly selectediOr RjA pause strategy is initiated. Pause time of one rollThe period is that the robot moves for a step length epsilon and the other robot continues to move according to the original plan
5. Collision avoidance strategy (optimized coordination strategy for collision avoidance) for various predicted situations
(1) If a possibility of a frontal collision is predicted, a modified local path coordination strategy (explained in detail later) is initiated.
(2) If the situation that side collision is about to occur is predicted, the robot only needs to wait for the time T required by moving one step length epsilon in situ and then moves according to the original planned path.
(3) And if the robot is predicted not to collide with the other robot, directly traveling according to the original planned path.
5.1 amend the local path coordination strategy, as shown in fig. 5, the basic idea of the strategy is: firstly, the robot plans a local optimal path by using local information on the premise of not considering other robots and dynamic obstacles, and the planning performed at the moment is recorded as the 1 st planning; then, the robot Rob is predictediAnd RobjWhen positive conflict possibly occurs, each robot plans a new local optimal path under the condition of considering the other robot, and determines which robot travels along the new path by comparing the influence of the new local paths on the global objective function, wherein the planning performed at this time is regarded as 2 nd planning. Let RobiIs longer than the original path by delta i, RobjIs longer than the original path by Δ j. If Δ i < Δ j, RobiSelecting a new path, RobjProceeding along the originally planned path; if Δ i > Δ j, RobjSelecting a new path, RobiTravel along the originally planned path; if Δ i ═ Δ j, Rob is chosen randomlyiOr RobjTravel along the newly planned path.
Considering the complexity and the particularity of the modified local path coordination strategy, the situation that the robot trajectory planning is not solved may occur, and therefore, the modified local path coordination strategy needs to be improved. For example, in the collision avoidance process, if the robot RiAnd robot RjOne of them fails to plan outSetting the track increment to + ∞forthe optimal track from the current position to the sub-target position, and then carrying out R operation on the robotiAnd robot RjComparing the trace increments of (a); if the two robots can not plan the optimal track from the current position to the sub-target position, one robot is randomly selected, the robot is retreated by a step length epsilon from the current position, information interaction is carried out with other robots to ensure that the retreating operation can not collide with other robots, and the other robot continues to advance along the original track.
This strategy is described in detail below.
With a robot RobiFor example. Rob learning by information interaction between robotsjCurrently planned local path, predicting robot RobiAnd RobjPath conflict of tkThe coordination strategy of the time is as follows:
step 1: at robot RiIS ofi(tk) In (1), the robot RjPropagates along its path as a static obstacle, giving IS'i(tk)；
Step 2: according to A*Algorithm IS'i(tk) Calculating a local optimal path FP'ki(PRi(tk)，P′subi(tk) There may be two cases for this step: FP 'can be planned out'ki(PRi(tk)，P′subi(tk) Let Δ i be FP'ki(PRi(tk)，P′subi(tk))-FPki(PRi(tk)，Psubi(tk) FP) hereki(PRi(tk)，Psubi(tk) And FP'ki(PRi(tk)，P′subi(tk) Respectively, the local optimal paths planned by the layer 1 planning and the layer 2 planning respectively correspond to the P according to the sub-target mapping rulesubi(tk)、P′subi(tk)；
No regulations marking FP'ki(PRi(tk)，P′subi(tk))；
Step 3: with robot RobjCommunication, knowing FP'kj(PRj(tk)，P′subj(tk) Whether it is programmable;
step 4: according to RobiAnd RobjAnd (3) determining a coordination result according to the planning condition of the new local path:
if RobiNew local paths can be planned, and RobjNot, then RobiStep size epsilon of travel along new local pathiInstant FPki(PRi(tk)，Psubi(tk))＝FP′ki(PRi(tk)，P′subi(tk))。
If RobjNew local paths can be planned, and RobiNot, then RobiThe step length epsilon is finished along the original local pathi。
If RobiAnd RobjA new local path can be planned, comparing Δ i and Δ j:
if Δ i < Δ j, RobiStep size epsilon of travel along new local pathiInstant FPki(PRi(tk)Psubi(tk))＝FP′ki(PRi(tk)，P′subi(tk))。
If delta i is equal to delta j, then a random number RandomNum is randomly selected from [0, 1 ], if 0.5 ≦ RandomNum < 1, then RobiStep size ε i is traveled along the original local path, otherwise, RobiStep size epsilon of travel along new local pathiInstant FPki(PRi(tk)，Psubi(tk))＝FP′ki(PRi(tk)，P′subi(tk))。
If RobiAnd RobjIf no new local path can be planned, d (P) is comparedRi(tk)，Gi) And d (P)Rj(tk)，Gj)：
If d (P)Ri(tk)，Gi)＜d(PRj(tk)，Gj) Then RobiBack by one step epsilon along the path traversedi。
If d (P)Ri(tk)Gi)＝d(PRj(tk)，Gj) Randomly selecting a number RandomNum in [0, 1), if the RandomNum is more than or equal to 0 and less than 0.5, then RobiBack by one step epsilon along the path traversedi。
The technical solution of the present invention is further described with reference to the following specific examples.
Example 1: one example of predictive front impact multi-robot path planning
FIG. 6(a) tkLocal planning of the time of day, FIG. 6(b) tkCoordination of time, FIG. 6(c) tk+1Planning of time of day, FIG. 6(d) tk+2Local planning of the time of day.
Example 2: one example of predictive side impact multi-robot path planning
Fig. 6(e), 6(f), and 6(g) are processes of suspending the coordination policy execution.
Experimental data for example 1
Experimental data for example 2
It should be noted that the embodiments of the present invention can be realized by hardware, software, or a combination of software and hardware. The hardware portion may be implemented using dedicated logic; the software portions may be stored in a memory and executed by a suitable instruction execution system, such as a microprocessor or specially designed hardware. Those skilled in the art will appreciate that the apparatus and methods described above may be implemented using computer executable instructions and/or embodied in processor control code, such code being provided on a carrier medium such as a disk, CD-or DVD-ROM, programmable memory such as read only memory (firmware), or a data carrier such as an optical or electronic signal carrier, for example. The apparatus and its modules of the present invention may be implemented by hardware circuits such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., or by software executed by various types of processors, or by a combination of hardware circuits and software, e.g., firmware.
The above description is only for the purpose of illustrating the present invention and the appended claims are not to be construed as limiting the scope of the invention, which is intended to cover all modifications, equivalents and improvements that are within the spirit and scope of the invention as defined by the appended claims.
Claims (10)
1. A distributed multi-robot cooperative motion control method is characterized by comprising the following steps:
each robot plans a local optimal path by using local information on the premise of not considering other robots;
and collision avoidance between the robot and other robots detected by the vehicle-mounted sensors is realized by utilizing a communication and coordination strategy.
2. The distributed multi-robot cooperative motion control method according to claim 1, wherein the robot Rob of the distributed multi-robot cooperative motion control methodiThe rolling planning algorithm comprises the following specific steps:
step one, initializing a starting point SiEnd point GiThe working environment WS, the view radius R and the robot step length delta a, wherein the delta a is less than or equal to R;
step two, refreshing robot RobiIS information set ofi(tk) That is, the robot RobiThe view field or the rolling window is used for acquiring the environmental information in the sensing range of the current robot;
thirdly, selecting local sub-targets on the premise of not considering other robots and dynamic obstacles: if G isi∈ISi(tk) Let Psubi(tk)＝GiOtherwise, to
And P issubi(tk) The sub-target mapping rule is met;
refers to the boundary of the field of view, FD (t)k) Refers to a feasible point set in the robot working environment WS;
step four, according to A*Algorithm in ISi(tk) In-computation local optimal path FPi(PRi(tk)，Psubi(tk))；
Step five, whether the sensor detects other robots RobjIf yes, Step6 is executed in sequence; if not, turning to Step 8;
step six, estimating Rob of robotiAnd RobjIf so, executing the step seven in sequence; if not, turning to the step eight;
step seven, collision avoidance is carried out through an optimization coordination strategy to obtain a newly planned local path FP'i(PRi(tk)，Psubi(tk) Let FPi(PRi(tk)，Psubi(tk))＝FP′i(PRi(tk)，Psubi(tk) Updating the planned local path;
step eight, the robot follows the planned local path FPi(PRi(tk)，Psubi(tk) One step with a step size of ε, 0< ε < R;
if PRi(tk)＝GiWhen the target is reached, quitting; otherwise, k is k +1, and step two is carried out.
3. The distributed multi-robot cooperative motion control method according to claim 1, wherein the time taken for the robot to travel one step size epsilon from the current position along the planned local path is Δ t, first, the robot RiThe vehicle-mounted sensor detects the robot RjRobot RjR is also detectedi，RiAnd RjCommunicate with each other, exchange respective local paths, and determine RiAnd RjIf the local paths have conflict, the problem of predicting the relative positions of the two tracks is converted into the judgment of whether the 2 track point sets have intersection; the specific process is as follows:
(1) if the two tracks do not intersect, the robot R is in the delta t timeiAnd robot RjThe robot continues to move forward according to the original path without collision;
(2) if the path tracks of the two paths have overlapped road sections, the two paths may run in the same direction on the overlapped road sections, so that a frontal collision occurs, and a local path coordination correction strategy is started;
(3) if the path tracks of the robot R and the robot R have 1 or more intersection points and the robot R move laterally and generate side collision, the robot R is randomly selectediOr RjStarting a pause strategy, wherein the pause time is a rolling period, namely the time T required by the robot to move one step epsilon, and the other robot moves according to the original timeThe planning continues to move.
4. The distributed multi-robot cooperative motion control method according to claim 1, wherein the collision avoidance strategies for various prediction situations of the distributed multi-robot cooperative motion control method include:
(1) if the situation that the front collision is possible is predicted, starting a local path correction coordination strategy;
(2) if the situation that side collision is about to occur is predicted, the robot only needs to wait for the time T required by moving one step length epsilon in situ and then moves according to the original planned path;
(3) and if the robot is predicted not to collide with the other robot, directly traveling according to the original planned path.
5. The distributed multi-robot cooperative motion control method according to claim 4, wherein the basic idea of modifying the local path coordination strategy is: firstly, planning a local optimal path by using local information on the premise of not considering other robots and dynamic obstacles, and recording the planning performed at the moment as the 1 st planning; then, the robot Rob is predictediAnd RobjWhen positive conflict possibly occurs, planning a new local optimal path by each robot under the condition of considering the opposite side, determining which robot travels along the new path by comparing the influence of the new local path on the global objective function, and recording the planning performed at the moment as the 2 nd planning; robiIs longer than the original path by delta i, RobjIs longer than the original path by delta j, if delta i < delta j, RobiSelecting a new path, RobjProceeding along the originally planned path; if Δ i > Δ j, RobjSelecting a new path, RobiTravel along the originally planned path; if Δ i ═ Δ j, Rob is chosen randomlyiOr RobjTravel along the newly planned path;
in the collision avoidance process, if the robot RiAnd robot RjOne of which fails to plan the current location to the sub-target locationIf the track is optimal, setting the track increment to be + ∞, and then carrying out R operation on the robotiAnd robot RjComparing the trace increments of (a); if the two robots can not plan the optimal track from the current position to the sub-target position, one robot is randomly selected, the robot is retreated by a step length epsilon from the current position, information interaction is carried out with other robots to ensure that the retreating operation can not collide with other robots, and the other robot continues to advance along the original track.
6. The distributed multi-robot cooperative motion control method according to claim 5, further comprising: rob learning by information interaction between robotsjCurrently planned local path, predicting robot RobiAnd RobjPath conflict of tkThe coordination strategy of the time is as follows:
the method comprises the following steps: at robot RiIS ofi(tk) In (1), the robot RjPropagates along its path as a static obstacle, giving IS'i(tk)；
Step two: according to A*Algorithm IS'i(tk) Calculating a local optimal path FP'ki(PRi(tk)，P′subi(tk) There may be two cases for this step: FP 'can be planned out'ki(PRi(tk)，P′subi(tk) Let Δ i be FP'ki(PRi(tk)，P′subi(tk))-FPki(PRi(tk)，Psubi(tk) FP) hereki(PRi(tk)，Fsubi(tk) And FP'ki(PRi(tk)，P′subi(tk) Respectively, the local optimal paths planned by the layer 1 planning and the layer 2 planning respectively correspond to the P according to the sub-target mapping rulesubi(tk)、P′subi(tk)；
No regulations marking FP'ki(PRi(tk)，P′subi(tk))；
Step three: with robot RobjCommunication, knowing FP'kj(PRj(tk)，P′subj(tk) Whether it is programmable;
step four: according to RobiAnd RobjAnd (3) determining a coordination result according to the planning condition of the new local path:
if RobiNew local paths can be planned, and RobjNot, then RobiStep size epsilon of travel along new local pathiInstant FPki(PRi(tk)，Psubi(tk))＝FP′ki(PRi(tk)，P′subi(tk))；
If RobjNew local paths can be planned, and RobiNot, then RobiThe step length epsilon is finished along the original local pathi；
If RobiAnd RobjA new local path can be planned, comparing Δ i and Δ j:
if Δ i < Δ j, RobiStep size epsilon of travel along new local pathiInstant FPki(PRi(tk)，Psubi(tk))＝FP′ki(PRi(tk)，P′subi(tk))；
If delta i is equal to delta j, then a random number RandomNum is randomly selected from [0, 1 ], if 0.5 ≦ RandomNum < 1, then RobiStep size ε i is traveled along the original local path, otherwise, RobiStep size epsilon of travel along new local pathiInstant FPki(PRi(tk)，Psubi(tk))＝FP′ki(PRi(tk)，P′subi(tk))；
If RobiAnd RobjIf no new local path can be planned, d (P) is comparedRi(tk)，Gi) And d (P)Rj(tk)，Gj)：
If d (P)Ri(tk)，Gi)＜d(PRj(tk)，Gj) Then RobiBack by one step epsilon along the path traversedi；
If d (P)Ri(tk)，Gi)＝d(PRj(tk)，Gj) Randomly selecting a number RandomNum in [0, 1), if the RandomNum is more than or equal to 0 and less than 0.5, then RobiBack by one step epsilon along the path traversedi。
7. The distributed multi-robot coordinated motion control method of claim 1, wherein after the local target is determined, the local path planning problem in the rolling window is resolved from P under the known environment informationR(t) to Psub(t) planning problem, using A*Solving the planning problem by an algorithm;
the path planning algorithm flow based on the rolling window is as follows:
step one, aligning with a starting point PsEnd point PgA work space WspaceInitializing the vision radius R and the step length epsilon of the robot;
step two, if the target point is reached, the planning is terminated;
step three, updating the map information in the current rolling window;
step four, generating a local optimization sub-target F according to the sub-target mapping rulesub(t)；
Step five, according to the local sub-target Psub(t) planning a local path in the current window with the known environment information;
step six, moving forward one step according to the local path, wherein the step length is epsilon, and epsilon is more than 0 and less than or equal to R;
step seven, returning to the step two;
the rolling path planning based on the A-algorithm comprises the following steps:
step one, initializing a starting point PinitEnd point P _ { good }, field radius R of working environment WS and Rob;
step two, if PgoalE.g. lS (t), let local sub-target Psub(t)＝PgoalOtherwise, determining the sub-targets by the above principle;
step three, planning a slave P by using A-star algorithmR(t) to Fsub(t) the optimal path;
step four, if Psub(t)＝PgoalLet Rob move to Psub(t), finishing planning and quitting; otherwise, let Rob follow the planned path to Psub(t) move when P is satisfiedR(t+ε)＝Psub(t), turning to the fifth step;
step five, refreshing the content of IS (t), and turning to step two.
8. A computer-readable storage medium storing a computer program which, when executed by a processor, causes the processor to perform the steps of:
each robot plans a local optimal path by using local information on the premise of not considering other robots;
and collision avoidance between the robot and other robots detected by the vehicle-mounted sensors is realized by utilizing a communication and coordination strategy.
9. A distributed multi-robot cooperative motion control system for implementing the distributed multi-robot cooperative motion control method according to any one of claims 1 to 7, wherein the distributed multi-robot cooperative motion control system comprises:
the first planning module is used for planning a local optimal path on the premise of not considering other robots by using local information;
and the second planning module is used for realizing collision avoidance between the robot and other robots detected by the vehicle-mounted sensors by utilizing a communication and coordination strategy.
10. A mobile robot system, wherein the mobile robot system runs the distributed multi-robot cooperative motion control method according to any one of claims 1 to 7, and the mobile robot system comprises: an unmanned aircraft, an unmanned automobile, or an unmanned underwater vehicle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011103507.1A CN112327829A (en) | 2020-10-15 | 2020-10-15 | Distributed multi-robot cooperative motion control method, system, medium and application |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011103507.1A CN112327829A (en) | 2020-10-15 | 2020-10-15 | Distributed multi-robot cooperative motion control method, system, medium and application |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112327829A true CN112327829A (en) | 2021-02-05 |
Family
ID=74314242
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011103507.1A Pending CN112327829A (en) | 2020-10-15 | 2020-10-15 | Distributed multi-robot cooperative motion control method, system, medium and application |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112327829A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113408784A (en) * | 2021-05-18 | 2021-09-17 | 华中科技大学 | Multi-robot transfer cooperative assembly method and equipment |
| CN113741424A (en) * | 2021-08-04 | 2021-12-03 | 深圳市普渡科技有限公司 | Robot cooperative obstacle avoidance system, method, robot and storage medium |
| CN114442636A (en) * | 2022-02-10 | 2022-05-06 | 上海擎朗智能科技有限公司 | Control method and device for following robot, robot and storage medium |
| CN115097816A (en) * | 2022-05-20 | 2022-09-23 | 深圳市大族机器人有限公司 | Modularized multi-robot cooperation control method |
| CN115576332A (en) * | 2022-12-07 | 2023-01-06 | 广东省科学院智能制造研究所 | Task-level multi-robot collaborative motion planning system and method |
| CN115657463A (en) * | 2022-05-27 | 2023-01-31 | 安徽大学 | Multi-robot distributed optimal cooperative control algorithm based on energy difference |
| EP4141599A1 (en) * | 2021-08-30 | 2023-03-01 | Rapyuta Robotics Co., Ltd. | Multi-robot route planning |
| CN116638528A (en) * | 2023-07-26 | 2023-08-25 | 深圳墨影科技有限公司 | Hybrid scheduling method of robot mobile collaboration system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030137268A1 (en) * | 1999-11-19 | 2003-07-24 | Regents Of The University Of Minnesota | Miniature robotic vehicles and methods of controlling same |
| CN110209167A (en) * | 2019-05-27 | 2019-09-06 | 西安电子科技大学 | A method of fully distributed multi-robot system is formed into columns in real time |
| CN110398967A (en) * | 2019-07-24 | 2019-11-01 | 西安电子科技大学 | A kind of multirobot collaboration trace information processing method using discretization method |
| CN111263335A (en) * | 2020-01-09 | 2020-06-09 | 上海交通大学 | Robot formation communication recovery method based on distributed cooperation and robot equipment |
| CN111708370A (en) * | 2020-07-21 | 2020-09-25 | 四川大学 | Multi-robot collaborative path planning method and system based on artificial potential field |
-
2020
- 2020-10-15 CN CN202011103507.1A patent/CN112327829A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030137268A1 (en) * | 1999-11-19 | 2003-07-24 | Regents Of The University Of Minnesota | Miniature robotic vehicles and methods of controlling same |
| CN110209167A (en) * | 2019-05-27 | 2019-09-06 | 西安电子科技大学 | A method of fully distributed multi-robot system is formed into columns in real time |
| CN110398967A (en) * | 2019-07-24 | 2019-11-01 | 西安电子科技大学 | A kind of multirobot collaboration trace information processing method using discretization method |
| CN111263335A (en) * | 2020-01-09 | 2020-06-09 | 上海交通大学 | Robot formation communication recovery method based on distributed cooperation and robot equipment |
| CN111708370A (en) * | 2020-07-21 | 2020-09-25 | 四川大学 | Multi-robot collaborative path planning method and system based on artificial potential field |
Non-Patent Citations (3)
| Title |
|---|
| 孙玉娇等: "基于领航者的多机器人系统编队控制研究", 《鲁东大学学报(自然科学版)》 * |
| 李静等: "一种改进多机器人分布式滚动路径规划算法", 《控制工程》 * |
| 王一可等: "一种多移动机器人的分布式滚动路径规划算法", 《微型电脑应用》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113408784A (en) * | 2021-05-18 | 2021-09-17 | 华中科技大学 | Multi-robot transfer cooperative assembly method and equipment |
| CN113408784B (en) * | 2021-05-18 | 2024-01-12 | 华中科技大学 | Multi-robot transfer cooperative assembly method and equipment |
| CN113741424A (en) * | 2021-08-04 | 2021-12-03 | 深圳市普渡科技有限公司 | Robot cooperative obstacle avoidance system, method, robot and storage medium |
| EP4141599A1 (en) * | 2021-08-30 | 2023-03-01 | Rapyuta Robotics Co., Ltd. | Multi-robot route planning |
| CN114442636A (en) * | 2022-02-10 | 2022-05-06 | 上海擎朗智能科技有限公司 | Control method and device for following robot, robot and storage medium |
| CN114442636B (en) * | 2022-02-10 | 2024-03-29 | 上海擎朗智能科技有限公司 | Control method and device of following robot, robot and storage medium |
| CN115097816A (en) * | 2022-05-20 | 2022-09-23 | 深圳市大族机器人有限公司 | Modularized multi-robot cooperation control method |
| CN115657463B (en) * | 2022-05-27 | 2023-10-13 | 安徽大学 | Multi-robot distributed optimal cooperative control algorithm based on energy difference |
| CN115657463A (en) * | 2022-05-27 | 2023-01-31 | 安徽大学 | Multi-robot distributed optimal cooperative control algorithm based on energy difference |
| CN115576332B (en) * | 2022-12-07 | 2023-03-24 | 广东省科学院智能制造研究所 | Task-level multi-robot collaborative motion planning system and method |
| CN115576332A (en) * | 2022-12-07 | 2023-01-06 | 广东省科学院智能制造研究所 | Task-level multi-robot collaborative motion planning system and method |
| CN116638528A (en) * | 2023-07-26 | 2023-08-25 | 深圳墨影科技有限公司 | Hybrid scheduling method of robot mobile collaboration system |
| CN116638528B (en) * | 2023-07-26 | 2023-09-22 | 深圳墨影科技有限公司 | Hybrid scheduling method of robot mobile collaboration system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112327829A (en) | Distributed multi-robot cooperative motion control method, system, medium and application | |
| CN111780777B (en) | Unmanned vehicle route planning method based on improved A-star algorithm and deep reinforcement learning | |
| Sun et al. | Probabilistic prediction of interactive driving behavior via hierarchical inverse reinforcement learning | |
| US11970161B2 (en) | Apparatus, method and article to facilitate motion planning of an autonomous vehicle in an environment having dynamic objects | |
| CN110398967B (en) | Multi-robot cooperative track information processing method adopting discretization method | |
| Parhi et al. | Navigational control of several mobile robotic agents using Petri-potential-fuzzy hybrid controller | |
| Fulgenzi et al. | Probabilistic motion planning among moving obstacles following typical motion patterns | |
| CN113110478A (en) | Method, system and storage medium for multi-robot motion planning | |
| Zhao et al. | A path planning method based on multi-objective cauchy mutation cat swarm optimization algorithm for navigation system of intelligent patrol car | |
| Mouhagir et al. | A markov decision process-based approach for trajectory planning with clothoid tentacles | |
| CN111176276A (en) | Development and application of intelligent warehousing robot | |
| Li et al. | Simulation analysis of a deep reinforcement learning approach for task selection by autonomous material handling vehicles | |
| Aizat et al. | A survey on navigation approaches for automated guided vehicle robots in dynamic surrounding | |
| Gu et al. | Path planning for mobile robot in a 2.5‐dimensional grid‐based map | |
| Kim et al. | Learning forward dynamics model and informed trajectory sampler for safe quadruped navigation | |
| Lin et al. | An improved fault-tolerant cultural-PSO with probability for multi-AGV path planning | |
| Guo et al. | Optimal navigation for AGVs: A soft actor–critic-based reinforcement learning approach with composite auxiliary rewards | |
| CN114428499A (en) | Astar and DWA algorithm fused mobile trolley path planning method | |
| Huy et al. | A practical and optimal path planning for autonomous parking using fast marching algorithm and support vector machine | |
| CN116551703A (en) | Motion planning method based on machine learning in complex environment | |
| Palacios-Morocho et al. | Multipath planning acceleration method with double deep R-learning based on a genetic algorithm | |
| Pandey et al. | Trajectory Planning and Collision Control of a Mobile Robot: A Penalty-Based PSO Approach | |
| Vladareanu et al. | The navigation mobile robot systems using Bayesian approach through the virtual projection method | |
| Zhang et al. | An adaptive artificial potential function approach for geometric sensing | |
| Han et al. | Reinforce learning-based collision avoidance in network assisted automated driving |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210205 |
|
| RJ01 | Rejection of invention patent application after publication |