CN117841006A - Track optimization method and device for multiple optimization targets of grabbing manipulator - Google Patents
Track optimization method and device for multiple optimization targets of grabbing manipulator Download PDFInfo
- Publication number
- CN117841006A CN117841006A CN202410252171.7A CN202410252171A CN117841006A CN 117841006 A CN117841006 A CN 117841006A CN 202410252171 A CN202410252171 A CN 202410252171A CN 117841006 A CN117841006 A CN 117841006A
- Authority
- CN
- China
- Prior art keywords
- track
- grabbing
- initial
- optimization
- manipulator
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000005457 optimization Methods 0.000 title claims abstract description 202
- 238000000034 method Methods 0.000 title claims abstract description 78
- 230000006870 function Effects 0.000 claims description 52
- 238000001514 detection method Methods 0.000 claims description 9
- 238000003860 storage Methods 0.000 claims description 9
- 238000005070 sampling Methods 0.000 claims description 6
- 238000000354 decomposition reaction Methods 0.000 claims description 5
- 238000004088 simulation Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 abstract description 5
- 238000004519 manufacturing process Methods 0.000 description 16
- 230000002068 genetic effect Effects 0.000 description 12
- 238000005265 energy consumption Methods 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000011451 sequencing strategy Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000035772 mutation Effects 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
Landscapes
- Manipulator (AREA)
Abstract
The invention relates to the field of data processing, and discloses a track optimization method and a track optimization device for a plurality of optimized targets of a grabbing manipulator. Determining an initial track sub-population of the grabbing subtask according to the starting point and the ending point of the grabbing subtask; co-evolving initial tracks in the initial track sub-population based on a non-dominant ordering strategy, and determining non-dominant solutions in the initial track sub-population; and carrying out global updating according to the non-dominant solution of each grabbing subtask to obtain the target track of the multi-manipulator collaborative grabbing task. Because the initial tracks in the initial track sub-population are subjected to co-evolution based on the non-dominant sorting strategy, and global updating is performed according to the obtained non-dominant solution, the track optimization method can perform the global cooperation of the mechanical arm, the accuracy of track optimization is improved, and the performance of track optimization is improved.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a track optimization method and device for multiple optimization targets of a grabbing manipulator.
Background
Along with the rapid development of industrial automation, the collaborative grabbing of multiple manipulators becomes a key link in modern production. In the fields of manufacturing, logistics and the like, a plurality of manipulators are required to cooperatively complete complex tasks such as component assembly, material handling and the like. However, due to the limitation of the working space, the complex environmental layout and the motion characteristics of the manipulators, how to reasonably plan the paths of the manipulators to achieve efficient collaborative gripping is a problem to be solved.
Existing implementations process complex environmental information through fuzzy logic to achieve optimization of trajectories. However, the method is difficult to model and design in practical application, is easy to fall into a local optimal solution, and has certain defects when being applied to complex multi-manipulator collaborative grabbing scenes.
The foregoing is provided merely for the purpose of facilitating understanding of the technical solutions of the present invention and is not intended to represent an admission that the foregoing is prior art.
Disclosure of Invention
The invention mainly aims to provide a track optimization method and device for multiple optimization targets of a grabbing manipulator, and aims to solve the technical problem that the complex multi-manipulator collaborative grabbing environment has certain defects in the prior art.
In order to achieve the above object, the present invention provides a track optimization method for multiple optimization targets of a grasping manipulator, the method comprising the steps of:
acquiring a multi-manipulator collaborative grabbing task, and dividing the multi-manipulator collaborative grabbing task into grabbing subtasks of a plurality of single manipulators;
determining an initial track sub-population of the grabbing subtask according to the starting point and the ending point of the grabbing subtask;
co-evolving initial tracks in each initial track sub-population based on a non-dominant ordering strategy, and determining non-dominant solutions in each initial track sub-population;
and carrying out global updating according to the non-dominant solution of each grabbing subtask to obtain the target track of the multi-manipulator collaborative grabbing task.
Optionally, the step of co-evolving the initial trajectories in each of the initial trajectory sub-populations based on the non-dominant ranking strategy to determine non-dominant solutions in each of the initial trajectory sub-populations includes:
determining an objective function corresponding to the grabbing subtask;
according to the objective function and based on a non-dominant ranking strategy, performing non-dominant ranking on initial tracks in the initial track sub-population, and determining an initial non-dominant solution of the grabbing subtask;
And performing co-evolution on the initial non-dominant solution corresponding to each grabbing subtask, and determining the non-dominant solution in each initial track sub-population.
Optionally, the step of determining the objective function corresponding to the grabbing subtask includes:
acquiring a resource consumption weight coefficient;
determining an initial objective function of the grabbing subtask according to the optimization objective of the grabbing subtask;
and determining an objective function corresponding to the grabbing subtask according to the resource consumption weight coefficient and the initial objective function.
Optionally, the step of determining the initial track sub-population of the grabbing subtask according to the starting point and the ending point of the grabbing subtask includes:
acquiring a starting point and an ending point of the grabbing subtask, and adding the starting point serving as a root node into a random tree;
sampling is carried out according to the starting point, and a random point is obtained;
determining a target point with a distance value smaller than a preset difference value from the random point in the random tree;
extending based on the direction from the random point to the target point;
if the target point is extended, adding the target point into the random tree;
when the distance between the target point and the termination point is smaller than a preset distance difference value, the target point is taken as a path end point;
Constructing an initial track between the starting point and the path end point according to the random tree;
and when the number of the path end points is larger than the preset number, determining an initial track sub-population formed by the initial tracks.
Optionally, the step of determining an initial track sub-population consisting of the initial tracks includes:
detecting the initial trajectory of the existing collision points based on a collision detection algorithm;
performing local obstacle avoidance optimization on the initial track with collision points according to an obstacle avoidance path planning algorithm to obtain an obstacle avoidance track;
and forming an initial track sub-population according to the initial track without collision points in the grabbing subtask and the obstacle avoidance track.
Optionally, the step of performing local obstacle avoidance optimization on the initial track with the collision point according to an obstacle avoidance path planning algorithm to obtain an obstacle avoidance track includes:
performing motion simulation on the initial track with collision points by using an artificial potential field method;
and when the repulsive force of the collision point is perceived, carrying out local obstacle avoidance optimization on the repulsive force position to obtain an obstacle avoidance track of the grabbing subtask.
Optionally, the step of globally updating according to the non-dominant solution of each grabbing subtask to obtain the target track of the multi-manipulator collaborative grabbing task includes:
Performing global updating according to non-dominant solutions of each grabbing subtask, and determining a global optimal solution set;
and carrying out iterative updating according to the global optimal solution set to obtain the target track of the multi-manipulator collaborative grabbing task.
In addition, in order to achieve the above object, the present invention also provides a track optimizing apparatus for a multi-optimizing object of a grasping manipulator, the track optimizing apparatus for a multi-optimizing object of a grasping manipulator comprising:
the task decomposition module is used for acquiring a multi-manipulator collaborative grabbing task and dividing the multi-manipulator collaborative grabbing task into grabbing subtasks of a plurality of single manipulators;
the track determining module is used for determining an initial track sub-population of the grabbing subtask according to the starting point and the ending point of the grabbing subtask;
the collaborative module is used for carrying out collaborative evolution on the initial tracks in the initial track sub-populations based on a non-dominant ordering strategy, and determining non-dominant solutions in the initial track sub-populations;
and the track updating module is used for carrying out global updating according to the non-dominant solution of each grabbing subtask to obtain the target track of the multi-manipulator collaborative grabbing task.
In addition, in order to achieve the above object, the present invention also proposes a trajectory optimization device for grasping a plurality of optimization targets of a manipulator, the device comprising: the system comprises a memory, a processor and a track optimization program of a grabbing manipulator multi-optimization target, wherein the track optimization program of the grabbing manipulator multi-optimization target is stored on the memory and can run on the processor, and the track optimization program of the grabbing manipulator multi-optimization target is configured to realize the steps of the track optimization method of the grabbing manipulator multi-optimization target.
In addition, in order to achieve the above object, the present invention further provides a storage medium, on which a track optimization program of a multi-optimization target of a grasping manipulator is stored, the track optimization program of the multi-optimization target of the grasping manipulator implementing the steps of the track optimization method of the multi-optimization target of the grasping manipulator as described above when executed by a processor.
According to the invention, the multi-manipulator collaborative grabbing task is divided into a plurality of grabbing subtasks of single manipulators by acquiring the multi-manipulator collaborative grabbing task; determining an initial track sub-population of the grabbing subtask according to the starting point and the ending point of the grabbing subtask; co-evolving initial tracks in the initial track sub-population based on a non-dominant ordering strategy, and determining non-dominant solutions in the initial track sub-population; and carrying out global updating according to the non-dominant solution of each grabbing subtask to obtain the target track of the multi-manipulator collaborative grabbing task. The initial tracks in the initial track sub-population are subjected to co-evolution based on the non-dominant sorting strategy, so that global updating is performed according to the obtained non-dominant solution, and the target track of the multi-manipulator collaborative grabbing task is obtained, so that the track optimization method can perform manipulator global collaboration, the track optimization accuracy is improved, the information exchange and collaboration in the global range are ensured, the track optimization method is prevented from falling into a local optimal solution, and the track optimization performance is improved.
Drawings
FIG. 1 is a schematic structural diagram of a track optimizing device for grasping multiple optimizing targets of a manipulator in a hardware running environment according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of a first embodiment of a track optimization method for grasping multiple optimization targets of a manipulator according to the present invention;
FIG. 3 is a schematic flow chart of a second embodiment of a track optimization method for grasping multiple optimization targets of a manipulator according to the invention;
FIG. 4 is a schematic flow chart of a third embodiment of a track optimization method for grasping multiple optimization targets of a manipulator according to the invention;
FIG. 5 is a flowchart of an application scenario of the track optimization method for grasping multiple optimization targets of a manipulator according to the present invention;
fig. 6 is a block diagram of a track optimizing apparatus for grasping multiple optimizing targets of a manipulator according to a first embodiment of the invention.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Referring to fig. 1, fig. 1 is a schematic diagram of a track optimizing device of a grasping manipulator with multiple optimizing targets in a hardware running environment according to an embodiment of the present invention.
As shown in fig. 1, the track optimizing apparatus for grasping a plurality of optimizing targets of a manipulator may include: a processor 1001, such as a central processing unit (Central Processing Unit, CPU), a communication bus 1002, a user interface 1003, a network interface 1004, a memory 1005. Wherein the communication bus 1002 is used to enable connected communication between these components. The user interface 1003 may include a Display, an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may further include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a Wireless interface (e.g., a Wireless-Fidelity (Wi-Fi) interface). The Memory 1005 may be a high-speed random access Memory (Random Access Memory, RAM) or a stable nonvolatile Memory (NVM), such as a disk Memory. The memory 1005 may also optionally be a storage device separate from the processor 1001 described above.
It will be appreciated by those skilled in the art that the configuration shown in fig. 1 does not constitute a limitation of the track optimization apparatus for multiple optimization objectives of the grasping manipulator, and may include more or fewer components than illustrated, or may combine certain components, or may be a different arrangement of components.
As shown in fig. 1, a memory 1005, which is a storage medium, may include an operating system, a network communication module, a user interface module, and a trajectory optimization program that grasps multiple optimization objectives of a manipulator.
In the track optimizing device with multiple optimizing targets of the grasping manipulator shown in fig. 1, the network interface 1004 is mainly used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the track optimizing device for the multi-optimizing targets of the grabbing manipulator can be arranged in the track optimizing device for the multi-optimizing targets of the grabbing manipulator, and the track optimizing device for the multi-optimizing targets of the grabbing manipulator calls the track optimizing program for the multi-optimizing targets of the grabbing manipulator stored in the memory 1005 through the processor 1001 and executes the track optimizing method for the multi-optimizing targets of the grabbing manipulator.
It should be noted that, the optimizing apparatus of the present invention may be applied to a scenario where object grabbing is required by a plurality of grabbing manipulators (simply referred to as manipulators in the embodiment of the present invention). For example, in the fields of inventory management, logistics transportation, production line automation, etc., embodiments of the present invention are not limited in this regard.
In one embodiment of the invention, the optimization apparatus may be applied to track optimization of a production line truss manipulator for production line automation.
In order to realize automation of the production line, a plurality of manipulators may be disposed on the production line truss. The optimization equipment can optimize the collaborative trajectories among a plurality of manipulators, pursue the minimum consumption of various resources, optimize the running trajectories of the manipulators in the aspect of a plurality of task targets, and avoid the problem that the conventional trajectory optimization method only considers a single optimization target and is difficult to realize efficient collaborative grasping.
The present invention will be described by taking this embodiment as an example, and examples of the apparatus of the present invention and examples of the methods, apparatuses, and storage medium described below will be described.
The embodiment of the invention provides a track optimization method for a multi-optimization target of a grabbing manipulator, and referring to fig. 2, fig. 2 is a flow chart of a first embodiment of the track optimization method for the multi-optimization target of the grabbing manipulator.
In this embodiment, the track optimization method for multiple optimization targets of the grabbing manipulator includes the following steps:
Step S10: and acquiring a multi-manipulator collaborative grabbing task, and dividing the multi-manipulator collaborative grabbing task into grabbing subtasks of a plurality of single manipulators.
It should be noted that, the execution body of the method of this embodiment may be a terminal device having functions of data processing, track optimization and program running, for example, an industrial computer, a server, etc., or may be an electronic device having the same or similar functions, for example, a track optimization device with multiple optimization targets for the grasping manipulator. The present embodiment and the following embodiments will be described below by taking a trajectory optimization device (hereinafter referred to as an optimization device) that grips multiple optimization targets of a manipulator as an example.
It will be appreciated that the robotic arm, i.e., the device or apparatus provided on the production line truss, may be used to effect automated operations. Through the manipulator of setting on the line truss, can realize the high-speed high-efficient automation mechanized operation of line.
It should be understood that a plurality of manipulators (i.e., a plurality of manipulators) can be arranged on the production line truss, and the running tracks of the manipulators on the production line truss are optimized through the optimizing equipment, so that the energy consumption, time and other resource consumption of the production line can be reduced, and the expansion of the production line gauge is facilitated.
In order to realize control and track optimization of multiple manipulators, the optimizing device may process the track optimization flow and operation in a task manner. Specifically, the optimizing device may generate/acquire a multi-manipulator collaborative grabbing task, and obtain grabbing subtasks of a plurality of single manipulators by decomposing the task.
It can be understood that, because the multiple manipulators can be regarded as being composed of multiple single manipulators (i.e. single manipulators), the multiple manipulator collaborative grabbing task can also be regarded as grabbing subtasks of multiple single manipulators, and the target track corresponding to the multiple manipulator collaborative grabbing task can be determined by collaborative optimization of the tracks corresponding to the grabbing subtasks of the single manipulators, so that the problem of optimizing the track of the multiple manipulators on the production line truss is converted from global optimization to local optimization, the track optimizing precision is improved, and meanwhile, the possibility of collision among the multiple single manipulators is reduced.
In a specific implementation, the optimizing device acquires a multi-manipulator collaborative grabbing task and divides the multi-manipulator collaborative grabbing task into grabbing subtasks of a plurality of single manipulators. The multi-manipulator collaborative grabbing task is decomposed, so that the global optimization problem is converted into the local optimization problem, the track optimization precision is improved, and the possibility of collision among a plurality of single manipulators is reduced.
Step S20: and determining an initial track sub-population of the grabbing subtask according to the starting point and the ending point of the grabbing subtask.
It should be noted that, for each single manipulator to be controlled, there may be a start point and an end point. The starting point corresponds to the current position of the manipulator before the grabbing subtask is executed, and the ending point corresponds to the target position of the manipulator after the grabbing subtask is executed. For example, the starting point may be the initial object gripping position and the ending point may be the object placement position of the target by performing the gripping operation by the manipulator.
It should be understood that when performing the multi-manipulator collaborative grabbing task, it may be that a part of the manipulators need to cooperate to complete the task, or that all the manipulators need to cooperate and grab for multiple times to complete the task, which is not limited in the embodiment of the present invention. The single manipulator to be controlled, namely the single manipulator involved in the multi-manipulator collaborative grabbing task.
It should be noted that, when determining the starting point and the ending point corresponding to the grabbing subtask of the single manipulator, path planning may be performed according to the starting point and the ending point, so as to determine a possible track, that is, an initial track, for completing the grabbing subtask.
It will be appreciated that a grabbing subtask may correspond to at least one possible track, i.e. the grabbing subtask comprises at least one initial track. By using these initial trajectories as individuals of the sub-population, the initial trajectory sub-population can be constructed.
It should be noted that, for each single manipulator involved in the multi-manipulator collaborative grabbing task, it may correspond to an initial track sub-population. The overall path optimization of the multi-manipulator collaborative grabbing task can be realized by optimizing the initial track sub-population corresponding to the single manipulator involved in the manipulator collaborative grabbing task and performing collaborative optimization on the single manipulator according to the initial track sub-population.
It should be explained that the embodiment of the present invention is not limited to the method for determining the initial trajectory of the present invention, and may be selected according to the actual application, for example, bezier curve method, a-x algorithm, dijkstra algorithm, and the like.
In one embodiment of the present invention, a fast extended random tree (RRT) algorithm is taken as an example, and a method for generating an initial track in the present invention is described.
Specifically, the step of determining the initial track sub-population of the grabbing subtask according to the starting point and the ending point of the grabbing subtask includes:
Acquiring a starting point and an ending point of the grabbing subtask, and adding the starting point serving as a root node into a random tree;
sampling is carried out according to the starting point, and a random point is obtained;
determining a target point with a distance value smaller than a preset difference value from the random point in the random tree;
extending based on the direction from the random point to the target point;
if the target point is extended, adding the target point into the random tree;
when the distance between the target point and the termination point is smaller than a preset distance difference value, the target point is taken as a path end point;
constructing an initial track between the starting point and the path end point according to the random tree;
and when the number of the path end points is larger than the preset number, determining an initial track sub-population formed by the initial tracks.
It will be appreciated that the fast expanding random tree algorithm can efficiently handle high-dimensional, complex environments and find a continuous path for a single manipulator. To facilitate a detailed description of the use of the fast expanding random tree algorithm, the starting point of the grabbing subtask may be set to P start The termination point is set to P end 。
Specifically, the steps of the application of the fast expanding random tree are as follows:
A. Initializing: adding a starting point serving as a root node of a rapid expansion random tree (called a random tree for short) into the random tree;
B. iteration: adding random sampling points in a preset range near the current position (namely the starting point) of the manipulator into a random tree, and marking the random sampling points as random points P sample . Determining from a random tree a point P sample The nearest target point P nearest . Along the slave target point P nearest To a random point P sample Try to extend the tree until a random point P is reached sample Or collision obstacle if P is reached sample Will target point P nearest Adding into a random tree; if an obstacle is hit, the target point is discarded.
For a target point in the random tree, if the target point is not reached, performing next expansion based on the target point; if the destination is reached, the destination is the path destination.
It should be noted that, the distance between the reaching end point, i.e., the target point and the ending point is smaller than the preset distance difference. The preset difference value, the preset distance difference value and the preset number can be selected according to the actual application, which is not limited in the embodiment of the present invention.
It can be understood that when the number of target points reaching the end point meets the requirement, backtracking connection can be performed according to the target points, and a plurality of initial tracks can be obtained. And determining an initial track sub-population corresponding to the grabbing subtask of the single manipulator through the initial tracks.
In a specific implementation, the optimizing device may determine the initial track sub-population of the grabbing subtask according to the starting point and the ending point of the grabbing subtask. The method has the advantages that the initial track sub-population corresponding to each single manipulator is determined, the collaborative grabbing tasks of the multiple manipulators can be processed in a genetic algorithm, a collaborative algorithm and the like, the path planning problem is converted into the multi-objective optimization problem, and therefore the tracks of the multiple manipulators are optimized.
Step S30: and co-evolving the initial tracks in the initial track sub-population based on a non-dominant ordering strategy, and determining non-dominant solutions in the initial track sub-population.
It should be noted that the non-dominant ranking policy is a method or algorithm for performing non-dominant ranking in the multi-objective optimization problem. The priority degree of the initial tracks in the initial track sub-population in the environment can be ordered through a non-dominant ordering strategy, so that the priority relation of the initial tracks is determined.
It should be appreciated that in embodiments of the present invention, the non-dominant ranking may be performed by a non-dominant ranking genetic (NSGA-II) algorithm, or may be performed based on other algorithms, such as Fitness ranking, crowding Distance, etc., which embodiments of the present invention are not limited in this respect.
It should be noted that co-evolution is a method for describing co-evolution or optimization between different individuals or populations by competing and cooperating with each other due to interactions and effects. In the embodiment of the present invention, the co-evolution method may be a co-evolution algorithm based on a particle swarm, a co-evolution algorithm based on a neural network, a modified co-evolution algorithm, or the like, which is not limited in the embodiment of the present invention.
In one implementation mode of the invention, the embodiment of the invention realizes the track optimization of the multi-optimization targets of the grabbing manipulators through a non-dominant sorting genetic algorithm, and further adopts a collaborative algorithm framework to solve the track problems of the multi-manipulators in groups. The tracks of each single manipulator can form a sub-population, and the optimal solutions under a plurality of optimization targets are cooperatively solved by utilizing the co-evolution and non-dominant sequencing strategies, so that the solving efficiency and accuracy of the problem are improved.
It should be noted that, the path planning problem of the grabbing subtask can be converted into a track optimization problem of multiple optimization targets of the grabbing manipulator to solve. There may be one objective function for each optimization objective. Specifically, M objective functions, f respectively, may be provided 1 （boj）、f 2 （boj）、……、f M (boj) wherein obj is used to represent a parameter vector for trajectory planning. The track optimization problem of the multi-optimization target of the grabbing manipulator can be expressed as:
Minimize f i （obj）；
wherein i=1, 2, … …, M; m is the number of optimization targets.
It can be appreciated that for the track optimization problem of multiple optimization targets of the grasping manipulator, the optimization problem can be measured by pareto optimal solution, that is, there is no scheme that can be better than other schemes on all target functions.
In the embodiment of the present invention, the method of the present invention will be described by taking four manipulators to perform a gripping task as an example.
In particular, for different robots of the truss, it is possible to take care of assembly work of different stages, such as picking up, handling and placing different components. According to the invention, the responsibility areas and task targets of the single manipulator of each truss are defined by decomposing the multi-manipulator collaborative grabbing task, and the multi-manipulator collaborative grabbing task is completed by optimizing the track of the single manipulator of each truss.
It will be appreciated that, for each manipulator's grabbing subtask, its corresponding optimization objective may include time, energy consumption, safety, etc., which the embodiments of the present invention do not limit. Because contradiction and conflict can exist between different optimization targets, the embodiment of the invention optimizes the track through the non-dominant sorting genetic algorithm and the cooperative algorithm, and obtains the non-dominant solution of the initial track sub-population by setting different objective functions for different optimization targets, thereby improving the accuracy of track optimization.
In a specific implementation, the optimization device co-evolves the initial trajectories in the initial trajectory sub-population based on a non-dominant ordering strategy, and determines non-dominant solutions in the initial trajectory sub-population. As non-dominant solutions in the track sub-population are determined, the diversity of multi-manipulator running track planning is improved, the selection space of the system is increased, and the accuracy of track optimization is improved.
Step S40: and carrying out global updating according to the non-dominant solution of each grabbing subtask to obtain the target track of the multi-manipulator collaborative grabbing task.
It can be appreciated that when determining the non-dominant solution of the initial track sub-population corresponding to each single manipulator, the non-dominant solution of each initial track sub-population can be globally summarized. By globally summarizing non-dominant solutions corresponding to each single manipulator, information exchange and cooperation in a global scope are guaranteed, quality of solutions is further optimized, and more optimization choices are provided. By collecting non-dominant solutions from different sub-populations, a global pareto optimal front is formed, high-quality solutions of the whole multi-manipulator system are covered, the track optimization method is prevented from being trapped into a local optimal solution, and the track optimization performance is improved.
It should be noted that, for each single manipulator, the updated non-dominant solution may be used as a new population to perform further iterative updating, and through continuous iteration, an effective information sharing and cooperation are formed between the global and local, and when the iteration termination condition is satisfied, the target track of the multi-manipulator collaborative grabbing task may be obtained.
Specifically, the step of globally updating according to the non-dominant solution of each grabbing subtask to obtain the target track of the multi-manipulator collaborative grabbing task includes:
performing global updating according to non-dominant solutions of each grabbing subtask, and determining a global optimal solution set;
and carrying out iterative updating according to the global optimal solution set to obtain the target track of the multi-manipulator collaborative grabbing task.
It will be appreciated that after a global update, the resulting set of globally optimal solutions may not be the true globally optimal solution. In order to enhance the accuracy of the trajectory optimization, the method may return to step S20 to perform co-evolution based on the non-dominant ranking strategy and according to the trajectories in the sub-population corresponding to each manipulator in the global optimal solution set, so as to implement iterative update. And when the iterative updating of the target is completed, the target track of the multi-manipulator collaborative grabbing task can be obtained.
In a specific implementation, the optimization device may perform global update according to the non-dominant solution of each grabbing subtask, so as to determine the target track of the manipulator collaborative grabbing task. The global updating is carried out according to the non-dominant solution, so that the information exchange and cooperation in the global range are ensured, the quality of the solution is further optimized, more optimization choices are provided, the track optimization method is prevented from falling into the local optimal solution, and the track optimization performance is improved.
According to the embodiment of the invention, the multi-manipulator collaborative grabbing task is divided into a plurality of grabbing subtasks of a single manipulator by acquiring the multi-manipulator collaborative grabbing task; determining an initial track sub-population of the grabbing subtask according to the starting point and the ending point of the grabbing subtask; co-evolving initial tracks in the initial track sub-population based on a non-dominant ordering strategy, and determining non-dominant solutions in the initial track sub-population; and carrying out global updating according to the non-dominant solution of each grabbing subtask to obtain the target track of the multi-manipulator collaborative grabbing task. The initial tracks in the initial track sub-population are subjected to co-evolution based on the non-dominant sorting strategy, so that global updating is performed according to the obtained non-dominant solution, and the target track of the multi-manipulator collaborative grabbing task is obtained, so that the track optimization method can perform manipulator global collaboration, the track optimization accuracy is improved, the information exchange and collaboration in the global range are ensured, the track optimization method is prevented from falling into a local optimal solution, and the track optimization performance is improved.
Based on the first embodiment of the trajectory optimization method of the grasping manipulator multi-optimization target of the present invention as described above, a second embodiment of the trajectory optimization method of the grasping manipulator multi-optimization target of the present invention is proposed.
Referring to fig. 3, fig. 3 is a flowchart of a second embodiment of a track optimization method for grasping multiple optimization targets of a manipulator according to the present invention.
In this embodiment, the step of determining the non-dominant solution in the initial trajectory sub-population by co-evolving the initial trajectory in the initial trajectory sub-population based on the non-dominant ranking strategy includes:
step S31: and determining an objective function corresponding to the grabbing subtask.
It will be appreciated that for different optimization objectives of the grabbing subtasks, the corresponding objective functions may be the same or different. In one embodiment of the present invention, the objective function of the grabbing subtask may be expressed as: minimum fi (obj). Wherein i=1, 2, … …, M; m is the number of optimization targets.
Specifically, the step of determining the objective function corresponding to the grabbing subtask includes:
acquiring a resource consumption weight coefficient;
determining an initial objective function of the grabbing subtask according to the optimization objective of the grabbing subtask;
And determining an objective function corresponding to the grabbing subtask according to the resource consumption weight coefficient and the initial objective function.
It should be noted that, according to the embodiment of the present invention, the initial objective function corresponding to the optimization objective may be determined according to the resource consumption weight coefficient corresponding to the optimization objective. And determining the target function of the grabbing subtask based on each initial target function in the grabbing subtask and the resource consumption weight coefficient corresponding to the initial target function.
It is understood that the resources may be time resources, energy resources, path length resources, etc., which are not limited by the embodiment of the present invention.
In one embodiment of the present invention, the comprehensive utilization of key resources such as energy and time is realized. An energy consumption model and a time model can be introduced to find the best balance point between energy consumption and time cost. For example, the objective function may be expressed using the following formula:
Minimize （α·fenergy（obj）+（1+α）·ftime（obj））。
where α is a trade-off coefficient of energy and time, fernergy (obj) is used to represent an objective function of energy, and ftime (boj) is used to represent an objective function of time.
It can be understood that the invention considers the comprehensive utilization of key resources such as robot energy, time and the like, introduces an energy consumption model and a time model, combines a multi-objective optimization algorithm, searches an optimal balance point between energy consumption and time cost, and improves the grabbing efficiency and the resource utilization rate.
Step S32: and determining an initial non-dominant solution of the grabbing subtask according to the objective function and based on a non-dominant ranking strategy to perform non-dominant ranking on the initial tracks in the initial track sub-population.
When determining the objective function corresponding to the grabbing subtask, the method can evaluate the corresponding initial track in the initial track sub-population of the grabbing subtask based on the objective function, thereby realizing the quantification of the performance of each initial track. According to the evaluation result, an evolution strategy of a non-dominant ordering genetic algorithm is adopted in each sub-population, and the operation such as crossing, mutation and the like is carried out on individuals in the sub-population corresponding to each single manipulator, so that new individuals (namely, an evolution track) are formed, and the new individuals are added into the atomic population (namely, an initial track sub-population) to obtain the evolution track sub-population.
It will be appreciated that the evolution strategy of the non-dominant ranking genetic algorithm may be a plurality of group strategies, a multi-objective comparison strategy, a co-evolution strategy, etc., and the embodiments of the present invention are not limited thereto.
In one embodiment of the present invention, multiple group strategies may be employed to evolve a sub-population corresponding to a single manipulator. Specifically, the sub-population corresponding to a single manipulator may be subdivided into multiple sub-populations (i.e., sun Chongqun), each grandchild population may be responsible for optimizing a particular objective function. Information sharing or competition can be carried out among grandchild populations, so that the effect of jointly optimizing multiple optimization targets is achieved.
In another embodiment of the invention, elite strategies can be introduced, a part of individuals with higher fitness are selected for genetic operation before genetic operation, and then other individuals are subjected to crossover, mutation and other genetic operations, so that the individuals with higher fitness have more opportunities to be stored, and the algorithm complexity of the system is reduced.
It should be explained that, by evaluating the sub-population of the evolution track through a preset fitness function (for example, the objective function, or a method based on weight and constraint, etc.), a set of non-dominant excellent tracks can be obtained by performing non-dominant sorting according to the evaluation result, and the non-dominant excellent tracks are the initial non-dominant solutions of the grabbing subtasks.
In a specific implementation, the optimizing device performs non-dominant sorting on the initial tracks in the initial track sub-population according to the objective function and based on a non-dominant sorting strategy, and determines an initial non-dominant solution of the grabbing subtask. The initial tracks in the initial track sub-population corresponding to each manipulator are subjected to non-dominant sequencing according to the objective function and based on the non-dominant sequencing strategy, so that a non-dominant solution is determined, quality optimization of the track population corresponding to each single manipulator is realized, and robustness and accuracy of an algorithm are improved. The method effectively solves the problem of weight determination in the traditional multi-objective optimization, improves the diversity and global convergence of knowledge, and realizes the optimal track solution under the multi-optimization objective through non-dominant sorting and global optimization.
Step S33: and performing co-evolution on the initial non-dominant solution corresponding to each grabbing subtask, and determining the non-dominant solution in each initial track sub-population.
It should be noted that, in the track optimization process, the track optimization problem between a plurality of single manipulators can be regarded as a multi-manipulator track population, wherein the track sub-population of each single manipulator is used as an individual of the multi-manipulator track population; the problem of trajectory optimization for a single manipulator may be seen as a sub-population of trajectories, wherein the trajectories of the single manipulator are the individuals of the sub-population of trajectories.
It should be explained that, for the initial track sub-population corresponding to each single manipulator, the initial track sub-population can be subjected to non-dominant ranking by a non-dominant ranking genetic algorithm, so as to obtain non-dominant solutions in each initial track sub-population. For the multi-manipulator track population, the non-dominant solution in each initial track sub-population can be co-evolved through a cooperative type cooperative algorithm, so that the non-dominant solution corresponding to each single manipulator in cooperation and the global optimal solution formed by combining the non-dominant solutions corresponding to each single manipulator are determined.
It will be appreciated that with the scheme of the present invention, the individuals in each sub-population are divided into different levels according to non-dominant relationships, thereby forming a plurality of non-dominant layers. The individuals in each non-dominant layer are optimal for multiple objectives, providing a solution for diversity in trajectory optimization.
Note that, for both trajectories a and B, a can be considered to dominate B if a is not less than B on all objective functions. The solutions in the sub-population are divided into a plurality of non-dominant layers by non-dominant ordering, wherein the more front the non-dominant layer is, the more effective the corresponding solution is.
In a specific implementation, the optimizing device performs co-evolution on initial non-dominant solutions corresponding to each grabbing subtask, and determines non-dominant solutions in each initial track sub-population. The path optimization problem of the multiple manipulators is decomposed into the optimization problem of multiple optimization targets in the single manipulator and the optimization problem of multiple optimization targets among the single manipulators, the optimization problem of the multiple optimization targets in the single manipulator is solved through the non-dominant sequencing genetic algorithm, the optimization problem of the multiple optimization targets among the single manipulators is solved through the cooperative collaborative algorithm, and therefore non-dominant solutions corresponding to the single manipulators are determined, multi-target optimization of the truss manipulator of the production line is achieved, and the problem solving efficiency and accuracy are improved by utilizing the collaborative evolution and non-dominant sequencing strategies, and the robustness of the system is improved.
According to the embodiment of the invention, the objective function corresponding to the grabbing subtask is determined; according to the objective function and based on a non-dominant sorting strategy, performing non-dominant sorting on initial tracks in the initial track sub-population, and determining an initial non-dominant solution of the grabbing subtask; and co-evolving initial non-dominant solutions corresponding to the grabbing subtasks to determine non-dominant solutions in the initial track sub-populations. The path optimization problem of the multiple manipulators is decomposed into the optimization problem of multiple optimization targets in the single manipulator and the optimization problem of multiple optimization targets among the single manipulators, the optimization problem of the multiple optimization targets in the single manipulator is solved through the non-dominant sorting genetic algorithm, and the optimization problem of the multiple optimization targets among the single manipulators is solved through the cooperative collaborative algorithm, so that non-dominant solutions corresponding to the single manipulators are determined, the multi-target optimization of the truss manipulator of the production line is realized, and the robustness of the system and the accuracy of track optimization are improved.
Based on the above embodiments, in order to implement obstacle avoidance and expansion detection of multiple manipulators, a third embodiment of the method of the present invention is provided, and referring to fig. 4, fig. 4 is a schematic flow chart of a third embodiment of a track optimization method of multiple optimization targets of a grabbing manipulator of the present invention.
In this embodiment, the step of determining an initial track sub-population consisting of the initial tracks includes:
step S210: the initial trajectory of the existing collision points is detected based on a collision detection algorithm.
In an actual production environment, image acquisition equipment such as a camera can be further arranged, and the operation and grabbing track of the manipulator can be predicted and judged through the image acquisition equipment. Based on the image acquisition device, the trajectory optimization for each manipulator can be further achieved. In particular, a collision detection algorithm may be introduced. The algorithm can detect whether a possible collision point exists on the trajectory path of the manipulator. Once the collision point is detected, corresponding obstacle avoidance path planning measures, such as path adjustment or robot path stopping, can be taken.
It will be appreciated that the collision point may be the point where the robot arm intersects or other objects, and the embodiments of the present invention are not limited in this respect.
Step S220: and carrying out local obstacle avoidance optimization on the initial track with collision points according to an obstacle avoidance path planning algorithm to obtain an obstacle avoidance track.
It can be appreciated that an algorithm for obstacle avoidance path planning is designed in the embodiment of the invention, so as to ensure that the manipulator can bypass an obstacle in the motion process. Specifically, the step of performing local obstacle avoidance optimization on the initial track with collision points according to an obstacle avoidance path planning algorithm to obtain an obstacle avoidance track includes:
performing motion simulation on the initial track with collision points by using an artificial potential field method;
and when the repulsive force of the collision point is perceived, carrying out local obstacle avoidance optimization on the repulsive force position to obtain an obstacle avoidance track of the grabbing subtask.
The invention simulates the movement of the manipulator in the potential field by using an artificial potential field method, so that the manipulator receives repulsive force from an obstacle and receives attractive force of a target, thereby planning a path for avoiding the obstacle.
Step S230: and forming an initial track sub-population according to the initial track without collision points in the grabbing subtask and the obstacle avoidance track.
It can be understood that in the sub-population corresponding to the grabbing sub-task, the initial track without collision points does not need to be optimized, and the initial track sub-population can be formed according to the initial track without collision points and the obstacle avoidance track.
The embodiment of the invention detects the initial track of the existing collision point based on a collision detection algorithm; carrying out local obstacle avoidance optimization on an initial track with collision points according to an obstacle avoidance path planning algorithm to obtain an obstacle avoidance track; and forming an initial track sub-population according to the initial track without collision points in the grabbing subtask and the obstacle avoidance track. By introducing a collision detection algorithm and an obstacle avoidance path planning algorithm, the manipulator is ensured not to collide with other objects or the manipulator when the manipulator executes the grabbing task, and the safety and stability of the system are improved.
In one embodiment of the present invention, as shown in fig. 5, fig. 5 is a flowchart of an application scenario of the track optimization method for grasping multiple optimization targets of a manipulator according to the present invention.
In this embodiment, the track optimization method of the optimizing device for executing the multiple optimization targets of the grabbing manipulator includes the following steps:
determining a target: determining a multi-manipulator collaborative grabbing task;
task decomposition: performing task decomposition on the multi-manipulator collaborative grabbing task to obtain grabbing subtasks of a plurality of single manipulators;
determining a starting point and an end point and an optimal path planning: for each grabbing subtask of a single manipulator, determining a starting point and an ending point of the grabbing subtask; and performing optimal path planning based on the starting point and the end point; during path planning, a collision detection algorithm and an obstacle avoidance path planning algorithm are combined to improve the safety of the system.
The path planning problem is converted into a multi-objective optimization problem: based on an initial track sub-population obtained by optimal path planning, converting a path planning problem into a multi-objective optimization problem, and optimizing a plurality of objective functions through NSGA-II; at the time of optimization, key resource utilization is performed to determine the objective function of the optimization.
Specifically, the initial tracks in the initial track sub-population can be subjected to non-dominant sorting to obtain a non-dominant sorting relation; and carrying out sub-population evolution according to the non-dominant ordering relation, and introducing elite strategies to realize collaborative solving of optimal solutions under a plurality of targets.
In addition, the embodiment of the invention also provides a storage medium, wherein the storage medium is stored with a track optimization program of the multi-optimization targets of the grabbing manipulator, and the track optimization program of the multi-optimization targets of the grabbing manipulator realizes the steps of the track optimization method of the multi-optimization targets of the grabbing manipulator when being executed by a processor.
Based on the first embodiment of the track optimization method for the multi-optimization targets of the grabbing manipulator, the first embodiment of the track optimization device for the multi-optimization targets of the grabbing manipulator is provided, and referring to fig. 6, fig. 6 is a structural block diagram of the first embodiment of the track optimization device for the multi-optimization targets of the grabbing manipulator.
As shown in fig. 6, a track optimizing device for grasping multiple optimization targets of a manipulator according to an embodiment of the present invention includes:
the task decomposition module 601 is configured to obtain a multi-manipulator collaborative grabbing task, and divide the multi-manipulator collaborative grabbing task into grabbing subtasks of multiple single manipulators;
the track determining module 602 is configured to determine an initial track sub-population of the grabbing subtask according to a start point and an end point of the grabbing subtask;
a co-module 603, configured to co-evolve initial trajectories in each of the initial trajectory sub-populations based on a non-dominant ranking policy, and determine non-dominant solutions in each of the initial trajectory sub-populations;
and a track updating module 604, configured to perform global updating according to the non-dominant solution of each grabbing subtask, and obtain a target track of the multi-manipulator collaborative grabbing task.
According to the embodiment of the invention, the multi-manipulator collaborative grabbing task is divided into a plurality of grabbing subtasks of a single manipulator by acquiring the multi-manipulator collaborative grabbing task; determining an initial track sub-population of the grabbing subtask according to the starting point and the ending point of the grabbing subtask; co-evolving initial tracks in the initial track sub-population based on a non-dominant ordering strategy, and determining non-dominant solutions in the initial track sub-population; and carrying out global updating according to the non-dominant solution of each grabbing subtask to obtain the target track of the multi-manipulator collaborative grabbing task. The initial tracks in the initial track sub-population are subjected to co-evolution based on the non-dominant sorting strategy, so that global updating is performed according to the obtained non-dominant solution, and the target track of the multi-manipulator collaborative grabbing task is obtained, so that the track optimization method can perform manipulator global collaboration, the track optimization accuracy is improved, the information exchange and collaboration in the global range are ensured, the track optimization method is prevented from falling into a local optimal solution, and the track optimization performance is improved.
Further, the collaboration module 603 is further configured to determine an objective function corresponding to the grabbing subtask; according to the objective function and based on a non-dominant ranking strategy, performing non-dominant ranking on initial tracks in the initial track sub-population, and determining an initial non-dominant solution of the grabbing subtask; and performing co-evolution on the initial non-dominant solution corresponding to each grabbing subtask, and determining the non-dominant solution in each initial track sub-population.
Further, the coordination module 603 is further configured to obtain a resource consumption weight coefficient; determining an initial objective function of the grabbing subtask according to the optimization objective of the grabbing subtask; and determining an objective function corresponding to the grabbing subtask according to the resource consumption weight coefficient and the initial objective function.
Further, the track determining module 602 is further configured to obtain a start point and an end point of the grabbing subtask, and add the start point as a root node to a random tree; sampling is carried out according to the starting point, and a random point is obtained; determining a target point with a distance value smaller than a preset difference value from the random point in the random tree; extending based on the direction from the random point to the target point; if the target point is extended, adding the target point into the random tree; when the distance between the target point and the termination point is smaller than a preset distance difference value, the target point is taken as a path end point; constructing an initial track between the starting point and the path end point according to the random tree; and when the number of the path end points is larger than the preset number, determining an initial track sub-population formed by the initial tracks.
Further, the trajectory determination module 602 is further configured to detect the initial trajectory of the existing collision point based on a collision detection algorithm; performing local obstacle avoidance optimization on the initial track with collision points according to an obstacle avoidance path planning algorithm to obtain an obstacle avoidance track; and forming an initial track sub-population according to the initial track without collision points in the grabbing subtask and the obstacle avoidance track.
Further, the trajectory determination module 602 is further configured to perform motion simulation on the initial trajectory with the collision point by using an artificial potential field method; and when the repulsive force of the collision point is perceived, carrying out local obstacle avoidance optimization on the repulsive force position to obtain an obstacle avoidance track of the grabbing subtask.
Further, the track updating module 604 is further configured to perform global updating according to the non-dominant solution of each grabbing subtask, and determine a global optimal solution set; and carrying out iterative updating according to the global optimal solution set to obtain the target track of the multi-manipulator collaborative grabbing task.
Other embodiments or specific implementation manners of the track optimizing device for grasping multiple optimized targets of the manipulator according to the present invention may refer to the above method embodiments, and will not be described herein.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. read-only memory/random-access memory, magnetic disk, optical disk), comprising instructions for causing a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the method according to the embodiments of the present invention.
The foregoing description is only of the preferred embodiments of the present invention, and is not intended to limit the scope of the invention, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein or in the alternative, which may be employed directly or indirectly in other related arts.
Claims (9)
1. A track optimization method for a multi-optimization target of a grabbing manipulator, which is characterized by comprising the following steps:
acquiring a multi-manipulator collaborative grabbing task, and dividing the multi-manipulator collaborative grabbing task into grabbing subtasks of a plurality of single manipulators;
determining an initial track sub-population of the grabbing subtask according to the starting point and the ending point of the grabbing subtask;
co-evolving initial tracks in each initial track sub-population based on a non-dominant ordering strategy, and determining non-dominant solutions in each initial track sub-population;
performing global updating according to non-dominant solutions of all the grabbing subtasks to obtain target tracks of the multi-manipulator collaborative grabbing tasks;
the step of determining the non-dominant solution in each initial trajectory sub-population by co-evolving the initial trajectory in each initial trajectory sub-population based on the non-dominant ordering strategy comprises the following steps:
Determining an objective function corresponding to the grabbing subtask;
according to the objective function and based on a non-dominant ranking strategy, performing non-dominant ranking on initial tracks in the initial track sub-population, and determining an initial non-dominant solution of the grabbing subtask;
and performing co-evolution on the initial non-dominant solution corresponding to each grabbing subtask, and determining the non-dominant solution in each initial track sub-population.
2. The method for optimizing the trajectory of multiple optimization targets of a grasping manipulator according to claim 1, wherein the step of determining the objective function corresponding to the grasping subtask includes:
acquiring a resource consumption weight coefficient;
determining an initial objective function of the grabbing subtask according to the optimization objective of the grabbing subtask;
and determining an objective function corresponding to the grabbing subtask according to the resource consumption weight coefficient and the initial objective function.
3. The method for optimizing a trajectory of a multi-optimization target of a grasping manipulator according to claim 1, wherein the step of determining an initial trajectory sub-population of the grasping subtask according to a start point and an end point of the grasping subtask comprises:
acquiring a starting point and an ending point of the grabbing subtask, and adding the starting point serving as a root node into a random tree;
Sampling is carried out according to the starting point, and a random point is obtained;
determining a target point with a distance value smaller than a preset difference value from the random point in the random tree;
extending based on the direction from the random point to the target point;
if the target point is extended, adding the target point into the random tree;
when the distance between the target point and the termination point is smaller than a preset distance difference value, the target point is taken as a path end point;
constructing an initial track between the starting point and the path end point according to the random tree;
and when the number of the path end points is larger than the preset number, determining an initial track sub-population formed by the initial tracks.
4. A method of track optimization for multiple optimization objects of a grasping manipulator according to claim 3, wherein the step of determining an initial track sub-population composed of the initial tracks comprises:
detecting the initial trajectory of the existing collision points based on a collision detection algorithm;
performing local obstacle avoidance optimization on the initial track with collision points according to an obstacle avoidance path planning algorithm to obtain an obstacle avoidance track;
and forming an initial track sub-population according to the initial track without collision points in the grabbing subtask and the obstacle avoidance track.
5. The method for optimizing the track of multiple optimized targets of a grasping manipulator according to claim 4, wherein the step of performing local obstacle avoidance optimization on the initial track with collision points according to an obstacle avoidance path planning algorithm to obtain an obstacle avoidance track comprises the steps of:
performing motion simulation on the initial track with collision points by using an artificial potential field method;
and when the repulsive force of the collision point is perceived, carrying out local obstacle avoidance optimization on the repulsive force position to obtain an obstacle avoidance track of the grabbing subtask.
6. The method for optimizing the track of the multi-optimization target of the grabbing manipulator according to claim 1, wherein the step of obtaining the target track of the multi-manipulator collaborative grabbing task by performing global update according to the non-dominant solution of each grabbing subtask comprises the following steps:
performing global updating according to non-dominant solutions of each grabbing subtask, and determining a global optimal solution set;
and carrying out iterative updating according to the global optimal solution set to obtain the target track of the multi-manipulator collaborative grabbing task.
7. The track optimizing device of many optimizing target of manipulator is snatched, its characterized in that, the track optimizing device of many optimizing target of manipulator includes:
The task decomposition module is used for acquiring a multi-manipulator collaborative grabbing task and dividing the multi-manipulator collaborative grabbing task into grabbing subtasks of a plurality of single manipulators;
the track determining module is used for determining an initial track sub-population of the grabbing subtask according to the starting point and the ending point of the grabbing subtask;
the collaborative module is used for carrying out collaborative evolution on the initial tracks in the initial track sub-population based on a non-dominant sorting strategy, and determining non-dominant solutions in the initial track sub-population;
the track updating module is used for carrying out global updating according to the non-dominant solution of each grabbing subtask to obtain a target track of the multi-manipulator collaborative grabbing task;
the operation of co-evolving the initial trajectories in each of the initial trajectory sub-populations based on the non-dominant ranking strategy to determine non-dominant solutions in each of the initial trajectory sub-populations includes:
determining an objective function corresponding to the grabbing subtask;
according to the objective function and based on a non-dominant ranking strategy, performing non-dominant ranking on initial tracks in the initial track sub-population, and determining an initial non-dominant solution of the grabbing subtask;
and performing co-evolution on the initial non-dominant solution corresponding to each grabbing subtask, and determining the non-dominant solution in each initial track sub-population.
8. A trajectory optimization device for grasping multiple optimization targets of a manipulator, the device comprising: a memory, a processor, and a trajectory optimization program of a grasping manipulator multi-optimization target stored on the memory and operable on the processor, the trajectory optimization program of the grasping manipulator multi-optimization target configured to implement the steps of the trajectory optimization method of the grasping manipulator multi-optimization target according to any one of claims 1 to 6.
9. A storage medium, wherein a track optimization program of a multi-optimization target of a grasping manipulator is stored on the storage medium, and the track optimization program of the multi-optimization target of the grasping manipulator realizes the steps of the track optimization method of the multi-optimization target of the grasping manipulator according to any one of claims 1 to 6 when executed by a processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410252171.7A CN117841006B (en) | 2024-03-06 | 2024-03-06 | Track optimization method and device for multiple optimization targets of grabbing manipulator |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410252171.7A CN117841006B (en) | 2024-03-06 | 2024-03-06 | Track optimization method and device for multiple optimization targets of grabbing manipulator |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117841006A true CN117841006A (en) | 2024-04-09 |
| CN117841006B CN117841006B (en) | 2024-05-28 |
Family
ID=90542193
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410252171.7A Active CN117841006B (en) | 2024-03-06 | 2024-03-06 | Track optimization method and device for multiple optimization targets of grabbing manipulator |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117841006B (en) |
Citations (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105033520A (en) * | 2015-08-05 | 2015-11-11 | 柳州职业技术学院 | Multi-welding robot cooperative control system for improving particle swarm algorithm |
| CN105069530A (en) * | 2015-08-13 | 2015-11-18 | 肇庆学院 | Multi-robot task allocation method based on multi-objective optimization |
| CN204934904U (en) * | 2015-08-05 | 2016-01-06 | 柳州职业技术学院 | A kind of multiple stage welding robot cooperative control system of improve PSO algorithm |
| CN107390684A (en) * | 2017-07-14 | 2017-11-24 | 华南理工大学 | A kind of auto-parts welding optimum path planning method of multirobot collaboration |
| CN107443373A (en) * | 2017-07-20 | 2017-12-08 | 广东工业大学 | Collision prevention method for planning track and device based on articulated arm robots |
| US20180161986A1 (en) * | 2016-12-12 | 2018-06-14 | The Charles Stark Draper Laboratory, Inc. | System and method for semantic simultaneous localization and mapping of static and dynamic objects |
| CN110216673A (en) * | 2019-06-10 | 2019-09-10 | 成都理工大学 | The non-dominant neighborhood immune genetic Multipurpose Optimal Method of electro-hydraulic joint of robot track |
| US20200101608A1 (en) * | 2018-09-30 | 2020-04-02 | Xiangtan University | Method and system for determining motion path of mechanical arm |
| CN111046559A (en) * | 2019-12-12 | 2020-04-21 | 南京邮电大学通达学院 | Multi-objective optimization method based on double-layer elite co-evolution |
| CN111113409A (en) * | 2019-11-21 | 2020-05-08 | 东南大学 | Multi-robot multi-station cooperative spot welding operation planning method based on step-by-step optimization |
| CN111506083A (en) * | 2020-05-19 | 2020-08-07 | 上海应用技术大学 | Industrial robot safety obstacle avoidance method based on artificial potential field method |
| CN113050422A (en) * | 2021-03-09 | 2021-06-29 | 东北大学 | Multi-robot scheduling method based on maximin function multi-objective optimization algorithm |
| CN113084811A (en) * | 2021-04-12 | 2021-07-09 | 贵州大学 | Mechanical arm path planning method |
| CN113296407A (en) * | 2021-05-25 | 2021-08-24 | 南京航空航天大学 | Multi-machine cooperative track optimization method based on 5-time non-uniform rational B-spline |
| CN115122338A (en) * | 2022-08-15 | 2022-09-30 | 西南交通大学 | Multi-robot cooperation arc welding task planning method based on multi-objective optimization |
| CN116300756A (en) * | 2023-03-27 | 2023-06-23 | 华中科技大学 | Double-target optimal scheduling method and system for flexible manufacturing workshop with transportation robot |
| CN116276955A (en) * | 2022-11-30 | 2023-06-23 | 西安科技大学 | Drilling and anchoring robot drill boom track planning method, system and electronic equipment |
| CN116352714A (en) * | 2023-04-11 | 2023-06-30 | 广东工业大学 | Mechanical arm obstacle avoidance path planning method |
| CN116352697A (en) * | 2022-12-06 | 2023-06-30 | 四川大学锦江学院 | Mechanical arm track optimization acquisition method based on improved multi-target PSO algorithm |
| CN116460843A (en) * | 2023-04-04 | 2023-07-21 | 广东省科学院智能制造研究所 | Multi-robot collaborative grabbing method and system based on meta heuristic algorithm |
| CN116880424A (en) * | 2023-08-24 | 2023-10-13 | 遨博(北京)智能科技股份有限公司 | Multi-robot scheduling method and device based on multi-objective optimization |
| CN117140524A (en) * | 2023-09-22 | 2023-12-01 | 中冶赛迪技术研究中心有限公司 | Multi-scene multi-task multi-robot task allocation and collaborative operation method and system |
| CN117325157A (en) * | 2023-09-27 | 2024-01-02 | 上海船舶工艺研究所(中国船舶集团有限公司第十一研究所) | Task allocation method for multi-robot collaborative operation |
-
2024
- 2024-03-06 CN CN202410252171.7A patent/CN117841006B/en active Active
Patent Citations (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105033520A (en) * | 2015-08-05 | 2015-11-11 | 柳州职业技术学院 | Multi-welding robot cooperative control system for improving particle swarm algorithm |
| CN204934904U (en) * | 2015-08-05 | 2016-01-06 | 柳州职业技术学院 | A kind of multiple stage welding robot cooperative control system of improve PSO algorithm |
| CN105069530A (en) * | 2015-08-13 | 2015-11-18 | 肇庆学院 | Multi-robot task allocation method based on multi-objective optimization |
| US20180161986A1 (en) * | 2016-12-12 | 2018-06-14 | The Charles Stark Draper Laboratory, Inc. | System and method for semantic simultaneous localization and mapping of static and dynamic objects |
| CN107390684A (en) * | 2017-07-14 | 2017-11-24 | 华南理工大学 | A kind of auto-parts welding optimum path planning method of multirobot collaboration |
| CN107443373A (en) * | 2017-07-20 | 2017-12-08 | 广东工业大学 | Collision prevention method for planning track and device based on articulated arm robots |
| US20200101608A1 (en) * | 2018-09-30 | 2020-04-02 | Xiangtan University | Method and system for determining motion path of mechanical arm |
| CN110216673A (en) * | 2019-06-10 | 2019-09-10 | 成都理工大学 | The non-dominant neighborhood immune genetic Multipurpose Optimal Method of electro-hydraulic joint of robot track |
| CN111113409A (en) * | 2019-11-21 | 2020-05-08 | 东南大学 | Multi-robot multi-station cooperative spot welding operation planning method based on step-by-step optimization |
| CN111046559A (en) * | 2019-12-12 | 2020-04-21 | 南京邮电大学通达学院 | Multi-objective optimization method based on double-layer elite co-evolution |
| CN111506083A (en) * | 2020-05-19 | 2020-08-07 | 上海应用技术大学 | Industrial robot safety obstacle avoidance method based on artificial potential field method |
| CN113050422A (en) * | 2021-03-09 | 2021-06-29 | 东北大学 | Multi-robot scheduling method based on maximin function multi-objective optimization algorithm |
| CN113084811A (en) * | 2021-04-12 | 2021-07-09 | 贵州大学 | Mechanical arm path planning method |
| CN113296407A (en) * | 2021-05-25 | 2021-08-24 | 南京航空航天大学 | Multi-machine cooperative track optimization method based on 5-time non-uniform rational B-spline |
| CN115122338A (en) * | 2022-08-15 | 2022-09-30 | 西南交通大学 | Multi-robot cooperation arc welding task planning method based on multi-objective optimization |
| CN116276955A (en) * | 2022-11-30 | 2023-06-23 | 西安科技大学 | Drilling and anchoring robot drill boom track planning method, system and electronic equipment |
| CN116352697A (en) * | 2022-12-06 | 2023-06-30 | 四川大学锦江学院 | Mechanical arm track optimization acquisition method based on improved multi-target PSO algorithm |
| CN116300756A (en) * | 2023-03-27 | 2023-06-23 | 华中科技大学 | Double-target optimal scheduling method and system for flexible manufacturing workshop with transportation robot |
| CN116460843A (en) * | 2023-04-04 | 2023-07-21 | 广东省科学院智能制造研究所 | Multi-robot collaborative grabbing method and system based on meta heuristic algorithm |
| CN116352714A (en) * | 2023-04-11 | 2023-06-30 | 广东工业大学 | Mechanical arm obstacle avoidance path planning method |
| CN116880424A (en) * | 2023-08-24 | 2023-10-13 | 遨博(北京)智能科技股份有限公司 | Multi-robot scheduling method and device based on multi-objective optimization |
| CN117140524A (en) * | 2023-09-22 | 2023-12-01 | 中冶赛迪技术研究中心有限公司 | Multi-scene multi-task multi-robot task allocation and collaborative operation method and system |
| CN117325157A (en) * | 2023-09-27 | 2024-01-02 | 上海船舶工艺研究所(中国船舶集团有限公司第十一研究所) | Task allocation method for multi-robot collaborative operation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117841006B (en) | 2024-05-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Gong et al. | An effective memetic algorithm for multi-objective job-shop scheduling | |
| CN110321666B (en) | Multi-robot path planning method based on priori knowledge and DQN algorithm | |
| Hui et al. | A genetic algorithm for product disassembly sequence planning | |
| Wang et al. | Learning scheduling policies for multi-robot coordination with graph attention networks | |
| Niu et al. | An improved intelligent water drops algorithm for solving multi-objective job shop scheduling | |
| CN107895225B (en) | Multi-Agent conflict-free cooperative task allocation method | |
| CN111191934B (en) | Multi-target cloud workflow scheduling method based on reinforcement learning strategy | |
| Elsayed et al. | On an evolutionary approach for constrained optimization problem solving | |
| Fox et al. | Industrial applications of the ant colony optimization algorithm | |
| Huang et al. | Robust multi-robot coordination in pick-and-place tasks based on part-dispatching rules | |
| Zhang et al. | A combinatorial evolutionary algorithm for unrelated parallel machine scheduling problem with sequence and machine-dependent setup times, limited worker resources and learning effect | |
| CN110969362A (en) | Multi-target task scheduling method and system under cloud computing system | |
| CN112286149A (en) | Flexible workshop scheduling optimization method and system considering crane transportation process | |
| Kim et al. | FMS scheduling based on timed Petri Net model and reactive graph search | |
| Elsayed et al. | Integrated strategies differential evolution algorithm with a local search for constrained optimization | |
| CN112348314A (en) | Distributed flexible workshop scheduling method and system with crane | |
| Reddy et al. | Minimizing the total completion time on a multi-machine FMS using flower pollination algorithm | |
| CN115983423A (en) | Feeding and discharging scene scheduling optimization method considering double resource constraints | |
| Nguyen et al. | A PSO-based hyper-heuristic for evolving dispatching rules in job shop scheduling | |
| Aslan | Mathematical model and a variable neighborhood search algorithm for mixed-model robotic two-sided assembly line balancing problems with sequence-dependent setup times | |
| Kamali et al. | An immune-based multi-agent system for flexible job shop scheduling problem in dynamic and multi-objective environments | |
| CN117841006B (en) | Track optimization method and device for multiple optimization targets of grabbing manipulator | |
| Gaggero et al. | When time matters: Predictive mission planning in cyber-physical scenarios | |
| Panda et al. | Autonomous mobile robot path planning using hybridization of particle swarm optimization and Tabu search | |
| Turguner et al. | Solving job shop scheduling problem with Ant Colony Optimization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |