CN115122317A - Redundant manipulator path planning method and system based on optimal target configuration screening - Google Patents
Redundant manipulator path planning method and system based on optimal target configuration screening Download PDFInfo
- Publication number
- CN115122317A CN115122317A CN202210202596.8A CN202210202596A CN115122317A CN 115122317 A CN115122317 A CN 115122317A CN 202210202596 A CN202210202596 A CN 202210202596A CN 115122317 A CN115122317 A CN 115122317A
- Authority
- CN
- China
- Prior art keywords
- path
- target configuration
- target
- optimal
- new
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000012216 screening Methods 0.000 title claims abstract description 41
- 238000005070 sampling Methods 0.000 claims abstract description 63
- 239000012636 effector Substances 0.000 claims description 16
- 230000008569 process Effects 0.000 claims description 11
- 230000009466 transformation Effects 0.000 claims description 9
- 239000011159 matrix material Substances 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 6
- 238000013519 translation Methods 0.000 claims description 3
- 238000009827 uniform distribution Methods 0.000 claims description 3
- 238000005457 optimization Methods 0.000 abstract description 3
- 238000001514 detection method Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1656—Programme controls characterised by programming, planning systems for manipulators
- B25J9/1664—Programme controls characterised by programming, planning systems for manipulators characterised by motion, path, trajectory planning
Landscapes
- Engineering & Computer Science (AREA)
- Robotics (AREA)
- Mechanical Engineering (AREA)
- Feedback Control In General (AREA)
Abstract
The invention provides a redundant mechanical arm path planning method and system based on optimal target configuration screening, wherein sampling points with different characteristics grow according to the shortest distance; growing a new growing point in the direction of the sampling point by the random expansion tree; continuing to grow the new growing point to the target configuration set, and updating the current optimal path and shortest path distance; screening a target configuration set, eliminating target configurations which do not meet preset conditions, and adding target configurations which meet the preset conditions; through sampling and path continuous optimization in the heuristic joint subspace of the mechanical arm, blind exploration of the whole joint space is avoided, and the algorithm convergence speed is improved; and when the path is optimized, a better target configuration is continuously screened, the autonomous online selection of the target state is realized, and the global optimal solution of the planned path is realized.
Description
Technical Field
The invention belongs to the technical field of path planning, and particularly relates to a redundant mechanical arm path planning method and system based on optimal target configuration screening.
Background
The mechanical arm path planning means that a feasible collision-free geometric space path theta(s) (s is more than or equal to 0 and less than or equal to 1) is planned under the constraint of a known environment, wherein s represents a scalar parameterization path coefficient, theta (0) represents an initial state, theta (1) represents a target state, and no collision occurs between the mechanical arm corresponding to any s and theta(s) state and the environment or a connecting rod of the mechanical arm.
Sampling-based algorithms such as PRM (probabilistic road marking), RRT (fast-expanding random tree) and the like generally need to give definite initial state points and target state points, in a redundant mechanical arm task execution scene, an end effector Cartesian space target pose is often given, and due to the characteristics of the redundant mechanical arm, infinite corresponding joint space inverse solutions can be obtained. In the conventional method, a feasible inverse solution is generally selected from the infinite candidate solutions by an offline method as a target configuration for planning, for example, the target configuration is randomly selected according to a collision-free principle or a collision-free target configuration is selected according to a certain decision criterion such as the nearest Euclidean distance from an initial configuration, and then a general path planning method is adopted for planning, for example, an informed RRT algorithm utilizes a heuristic hyper-ellipsoid sampling subspace, so that the over-exploration of the whole mechanical arm configuration space is avoided, and the shortest path can be converged quickly under the condition that the initial and target configurations are known.
The inventor finds that the target configuration is simple in selection mode, but unreasonable target configurations can ignore more optimal target configurations, so that the shortest planning path is too long; if the strategy of fully sampling and searching the whole joint space is adopted, infinite paths reaching all feasible target inverse solutions can be obtained theoretically, and then the optimal target configuration and the optimal path are compared.
Disclosure of Invention
The invention provides a redundant manipulator path planning method and system based on optimal target configuration screening, aiming at the problem that the optimal joint space target configuration needs to be reasonably selected according to the target pose of a Cartesian space at the given end of a redundant manipulator, and the problem of excessive exploration of the joint space is solved.
In order to realize the purpose, the invention is realized by the following technical scheme:
in a first aspect, the invention provides a redundant manipulator path planning method based on optimal target configuration screening, which includes:
acquiring a joint value of the multi-degree-of-freedom mechanical arm and a target pose of an end effector; establishing a random expansion tree by taking the joint value of the multi-degree-of-freedom mechanical arm as a root node, and initializing the optimal path and the shortest path distance in the random expansion tree; establishing a target configuration set according to the target pose of the end effector;
growing sampling points with different characteristics according to the shortest distance;
growing a new growing point in the direction of the sampling point by the random expansion tree;
continuing to grow the new growing point to the target configuration set, and updating the current optimal path and shortest path distance;
screening a target configuration set, eliminating target configurations which do not meet preset conditions, and adding target configurations which meet the preset conditions;
circularly growing sampling points with different characteristics to a process of screening a target configuration set; and stopping circulation when a preset condition is reached to obtain a global optimal path.
Further, a target configuration set established according to the target pose of the end effector is a candidate optimal target configuration set, the element form in the candidate optimal target configuration set is (x, flag), x is a random inverse kinematics solution of the target pose of the end effector of the mechanical arm, and a flag logic bit indicates whether the x target configuration is added into a node set of a random expansion tree or not, namely whether a feasible path to x is planned or not.
Further, the process of growing the sampling points with different characteristics according to the shortest distance is as follows:
when the shortest path distance is infinite, a sampling point is obtained by configuring a space uniform sampling and matching with a target configuration direct sampling mode, and the method specifically comprises the following steps: generating a random number, uniformly sampling in a joint space of the mechanical arm to obtain a sampling point when the random number is smaller than a set threshold, and randomly selecting a certain target configuration in a target configuration set to be directly used as the sampling point when the random number is not smaller than the set threshold;
when the shortest path distance is less than infinity, sampling is performed in the superimposed hyperellipsoid sampling subspace, specifically: randomly selecting a certain element (x _ coarse, flag) in a target configuration set in a uniform probability mode, taking multi-degree-of-freedom mechanical arm joint values x _ begin and x _ coarse as focuses, and taking shortest path distance as the length of a long axis to sample a single hyperellipsoid sampling subspace according to uniform distribution; the sampling points were: x _ sample ═ CLx unit +x center ，
Wherein C represents a rotation transformation matrix from a hyper-ellipsoid coordinate system to a world coordinate system, L represents a scaling transformation from a unit hyper-sphere to a hyper-ellipsoid, and x unit Representing uniform sampling, x, within a multidimensional unit hypersphere center Representing a translation transformation.
Further, a new growing point grows in the direction of the sampling point by the random expansion tree, and nodes in the field of the new growing point are optimized by selecting an optimal father node and a reconnection mode.
Further, the way for the new growth point to continue growing to the target configuration set is as follows: traversing all elements of a target configuration set, if the elements (x, flag) meet the condition that flag is false, | | x _ new-x | | < step and x _ new is not more than x, and direct local paths from x _ new to x have no collision, adding a new node x into a node set, and adding edges (x _ new, x) into a connection relation set between nodes; step represents a constant of the maximum growth step of the node, x _ new represents that a new growth point is set, and false represents a feasible path;
the mode of updating the distance between the current optimal path and the shortest path is as follows: traversing the target configuration set, if the flag in the element (x, flag) is true, obtaining temp _ path by x in the random expansion tree edge set in a backtracking mode, and obtaining the value of temp _ path by x
Calculating the path length temp _ cbest of temp _ path, wherein d represents the element number of the temp _ path sequence, comparing the minimum values of the path lengths corresponding to all the elements meeting the condition that flag is true, assigning the minimum values to the shortest path distance, and obtaining the elementAnd assigning the obtained path to the optimal path.
Further, screening the set of target configurations comprises:
when the shortest path ratio is smaller than the value of the last cycle, traversing all elements of the target configuration set, and if a certain element meets the condition that | | | x _ begin-x | | > is more than or equal to cbest, deleting the element;
when the total number of elements in the target configuration set is less than the maximum number of elements in the target configuration set, determining whether to add a new candidate configuration through a random strategy, specifically: generating a random number, and when the random number is not less than a set threshold, not adding a new target element to the target configuration set; when the random number is smaller than a set threshold value, calling a mechanical arm inverse kinematics algorithm to generate a random inverse kinematics solution x corresponding to a collision-free mechanical arm target pose, if | | | x _ begin-x | < cbest, adding the solution into a target configuration set, and if not, continuously generating a random joint inverse solution until an adding condition is met; wherein cbest is the shortest distance.
And further, judging whether the process from the growth of the sampling points with different characteristics to the screening of the target configuration set circularly runs to a preset number of times or meets a preset circulating time, if not, continuously circulating, if so, the output optimal path is the optimal path of the whole situation, the first element in the optimal path sequence is a given initial configuration, and the tail element is the screened optimal target configuration.
In a second aspect, the present invention further provides a redundant manipulator path planning system based on an optimal target configuration, including:
a data processing module configured to: acquiring a joint value of the multi-degree-of-freedom mechanical arm and a target pose of an end effector; establishing a random expansion tree by taking the joint value of the multi-degree-of-freedom mechanical arm as a root node, and initializing the optimal path and the shortest path distance in the random expansion tree; establishing a target configuration set according to the target pose of the end effector;
a sample point acquisition module configured to: growing sampling points with different characteristics according to the shortest distance;
a new growth point acquisition module configured to: growing a new growing point in the direction of the sampling point by the random expansion tree;
an update module configured to: continuously growing the new growing point to the target configuration set, and updating the current optimal path and the shortest path distance;
a screening module configured to: screening a target configuration set, eliminating target configurations which do not meet preset conditions, and adding target configurations which meet the preset conditions;
a loop processing module configured to: circularly growing sampling points with different characteristics to a process of screening a target configuration set; and stopping circulation when a preset condition is reached to obtain a global optimal path.
In a third aspect, the present invention further provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor, implements the steps of the redundant manipulator path planning method based on the filtered optimal target configuration according to the first aspect.
In a fourth aspect, the present invention further provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor executes the program to implement the steps of the method for planning a path of a redundant manipulator based on a configuration of a screened optimal target according to the first aspect.
Compared with the prior art, the invention has the beneficial effects that:
according to the method, sampling points with different characteristics grow according to the shortest distance; growing a new growing point in the direction of the sampling point by the random expansion tree; continuing to grow the new growing point to the target configuration set, and updating the current optimal path and shortest path distance; screening a target configuration set, eliminating target configurations which do not meet preset conditions, and adding target configurations which meet the preset conditions; by sampling and path continuous optimization in the heuristic joint subspace of the mechanical arm, blind exploration on the whole joint space is avoided, and the algorithm convergence speed is improved; and when the path is optimized, a better target configuration is continuously screened, the autonomous online selection of the target state is realized, and the global optimal solution of the planned path is realized.
Drawings
The accompanying drawings, which form a part hereof, are included to provide a further understanding of the present embodiments, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the present embodiments and together with the description serve to explain the present embodiments without unduly limiting the present embodiments.
FIG. 1 is a flow chart of example 1 of the present invention.
The specific implementation mode is as follows:
the invention is further described with reference to the following figures and examples.
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure herein. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.
Example 1:
aiming at the problems that the optimal joint space target configuration needs to be reasonably selected and the excessive exploration of the joint space is avoided when the target pose of the Cartesian space at the given tail end of the redundant manipulator is given, the embodiment provides a redundant manipulator path planning method based on the optimal target configuration screening, in particular to a redundant manipulator heuristic joint subspace path planning method for online screening of the optimal target configuration in the planning process, and the method specifically comprises the following steps:
s1, setting obstacle environment information in a mechanical arm working space, establishing a redundant mechanical arm kinematics model and a collision model, and setting the following variable initial values:
in the embodiment, a simulation platform of the seven-degree-of-freedom mechanical arm can be built on an open-source robot operating system ROS; obtaining obstacle information of a working space by scanning with a depth camera, and obtaining the size and position information of all obstacles by simplifying environmental information with an AABB bounding box; specifically, an initial state is given as x _ begin, and the initial state is an n-dimensional vector and represents a joint value of the mechanical arm with n degrees of freedom; giving an end effector target pose P; growing a fast random expansion tree T by taking an initial state x _ begin as a root node, wherein T is (V, E), V represents a node set in the random expansion tree T, E represents a connection relation set between nodes in V, and an element form (x _ parent, x) in E represents that a node x _ parent is a parent node of a node x; setting a planned optimal path initial value as an empty sequence, sequentially storing turning points of the path by the path sequence, and connecting adjacent turning points in a PTP (Point-to-point) mode, namely an n-dimensional space linear motion mode; the shortest path distance cbest is a scalar quantity, the initial value is infinite, and the distance used in the embodiment is an Euclidean distance; setting a candidate optimal target configuration set goalset, wherein the element form in the goalset is (x, flag), wherein x is a random inverse kinematics solution of a target pose P at the tail end of a mechanical arm, and the logical bit of the flag indicates whether the target configuration x is added into a node set V of a random extended tree T, namely whether a feasible path to x is planned, wherein the default is false, the goalset contains an initial element, and the maximum element quantity of the goalset set is eta.
S2, growing sampling points x _ sample with different characteristics according to the shortest distance cbest:
when cbest is infinite, x _ sample is obtained by means of configuring space uniform sampling and matching target configuration direct sampling, and the specific method comprises the following steps: generating a random number rand, uniformly sampling in a joint space of the mechanical arm to obtain x _ sample when the rand is smaller than a set threshold value gamma, and randomly selecting a certain target configuration in a goleset set to be directly used as a sampling point when the rand is not smaller than the set threshold value gamma;
when cbest is smaller than infinity, sampling is carried out in the superimposed hyperellipsoid sampling subspace, and the specific method comprises the following steps:
randomly selecting a certain element (x _ goal, flag) in the golset in a uniform probability manner, and sampling according to uniform distribution in a single hyperellipsoid sampling subspace taking x _ begin and x _ goal as focuses and cbest as the long axis length. Sample point x _ sample ═ CLx unit +x center Wherein C ═ Udiag {1,.., 1, det (u) det (V) } V T Representing a rotation transformation matrix from a super-ellipsoid coordinate system to a world coordinate system, wherein the diagonal element number of the diag { } diagonal matrix is the degree of freedom n of the mechanical arm,
and
is to satisfy U sigma V T Of [ identical to ] MUnitary matrix obtainable by singular value decomposition of matrix M
Wherein
Representing the transpose of the first column of the identity matrix I. x is the number of center And (x _ good + x _ begin)/2, representing a translation transformation. L represents the scaling transformation from a unit hypersphere to a hypersphere,
wherein cmin | | x _ good-x _ init | |, and the number of elements in the diagonalmatrix is the degree of freedom n of the mechanical arm. x is the number of unit Representing uniform sampling, x, within an n-dimensional unit hypersphere unit ＝(x 1 ,x 2 ,...x n )，x i Is [ -1, 1 [ ]]Random floating point numbers evenly distributed over the interval, and | | x unit ||≤1。
S3, growing a new node x _ new in the x _ sample direction by the random expansion tree T, and optimizing the nodes in the x _ new field by selecting an optimal father node and a reconnection mode:
the new node growing mode is as follows: finding a node x _ nearest to x _ sample in the node set V, wherein x belongs to V, if | | | x _ sample-x | | | | is less than or equal to step, let x _ new ═ x _ sample, wherein step is a constant representing the maximum growth step of the node, and otherwise, let x _ nearest ═ argmin (| | | x _ sample-x | | | | | |), and let x _ new | x _ sample
Judging whether the local path between x _ new and x _ nearest is collided, wherein delta represents the detection step length of collision detection, and collecting
Wherein
Representing the maximum integer not greater than (·), if the mechanical arm configuration represented by any element in the set Π collides with the environment or between the self connecting rods, the process goes to step S5, otherwise, a new node x _ new is added to the set V.
The node mode in the field of optimizing x _ new is as follows: selecting a node omega near x _ new ═ x | x ∈ V, | | | | x-x _ new | ≦ min (α · (log (s))/s) 1/n β step), noCollision (x, x _ new), where s represents the number of elements in the set V, n represents the degree of freedom of the robot arm, noCollision (x, x _ new) indicates that no collision is required for the local path between x and x _ new, and α and β are adjustment coefficients. Selecting x _ new optimal parent node x _ min ═ argmin (cost (x) + | | | x-x _ new | |) in the set { x _ new | >, wherein cost (x) returns the length of the path from the root node x _ begin to the node x |, and then adding edges (x _ min, x _ new) into the set E, wherein the optimal parent node representing x _ new is set to x _ min. Then, the set { x _ nearest ^ U.OMEGA \ x _ min } is traversed, and \ "represents difference set operation, if a certain element x in the set meets cost (x _ new) + | x-x _ new |<cost (x), then (x _ parent, x) is deleted from the edge set E, and (x _ new, x) is added, i.e. the parent node of x is changed from x _ parent to x _ new by means of reconnection.
The function cost (x) is implemented as: and setting the length of a return path as l and the initial value as 0, because other nodes except the root node x _ begin in the random expansion tree T all have unique father nodes, searching for a unique element (x _ parent, x) from the edge set E, making l | + | | x _ parent-x |, continuously searching the father node at the upper stage of x and accumulating the length of the local path until the local path length added to the start node x _ begin, and returning to l.
S4, the new growth point x _ new tries to continue growing to the target set goalset, and the current shortest distance cbest and the optimal path are updated:
the way in which x _ new tries to grow to the target set goleset is: traversing all elements of the goleset set, if the elements (x, flag) meet the conditions that flag is false, | x _ new-x | < step |, and no collision exists in the direct local path from x _ new to x, adding a new node x into the set V, adding an edge (x _ new, x) into the set E, and updating the flag to true, wherein step is a constant representing the maximum growth step of the node, and judging whether the local path collides in the same step S3.
The method for updating the shortest distance cbest and the optimal path is as follows: traversing the golset set, if the flag in the element (x, flag) is true, obtaining temp _ path by x in the random expansion tree edge set E in a backtracking mode, and obtaining the temp _ path by x
Calculating the path length temp _ cbest of temp _ path, wherein d represents the element number of the temp _ path sequence, comparing the minimum values of the path lengths corresponding to all elements meeting the condition that flag is true, assigning the minimum values to cbest, and assigning the path obtained by the elements to path; cbest and path are not updated if the golset set is empty or all elements flag are false.
The method for obtaining temp _ path by backtracking x is as follows: firstly, the initial element of the temp _ path sequence only has x, a unique element (x _ parent, x) is searched in the edge set E, the x _ parent is added into the header of the temp _ path sequence, and the parent node of the head element in the temp _ path sequence is continuously searched and the header of the sequence is added in the same way until the added node is x _ begin.
S5, screening a target configuration set goleset, and adding a potential more optimal candidate target configuration:
when the shortest path cbest is smaller than the value of the last loop, all elements of the goleset are traversed, and if a certain element (x, flag) satisfies | | | x _ begin-x | | | | > cbest, the element is deleted from the goleset.
When the total number of elements in the goalset is less than the maximum number eta, whether to add a new candidate configuration is determined by a random strategy, and the specific method is as follows: generating a random number rand, and when rand is not less than the set threshold μ, not adding a new target element to the golset, and going directly to step S6; and when rand is smaller than a set threshold value mu, calling a mechanical arm inverse kinematics algorithm to generate a random inverse kinematics solution x corresponding to a collision-free mechanical arm target pose, if | | | x _ begin-x | < cbest, adding the solution into the set goalset, and otherwise, continuously generating a random joint inverse solution until the addition condition is met.
S6, judging whether the loop operation N times or the program set maximum operation time is reached in the steps S2-S5, if not, jumping to the step S2, if so, the finally output variable path is the required global optimal path, the first element in the path sequence is a given initial configuration, and the tail element is the screened optimal target configuration.
The embodiment has the advantages that: sampling and path continuous optimization are carried out in heuristic joint subspace of the mechanical arm, blind exploration on the whole joint space is avoided, and the algorithm convergence speed is improved; and continuously screening more excellent target configurations while optimizing the path, realizing the autonomous online selection of the target state and realizing the global optimal solution of the planned path.
Example 2:
the embodiment provides a redundant manipulator path planning system based on an optimal target configuration, which includes:
a data processing module configured to: acquiring a joint value of the multi-degree-of-freedom mechanical arm and a target pose of an end effector; establishing a random expansion tree by taking the joint value of the multi-degree-of-freedom mechanical arm as a root node, and initializing the optimal path and the shortest path distance in the random expansion tree; establishing a target configuration set according to the target pose of the end effector;
a sample point acquisition module configured to: growing sampling points with different characteristics according to the shortest distance;
a new growth point acquisition module configured to: growing a new growing point in the direction of the sampling point by the random expansion tree;
an update module configured to: continuing to grow the new growing point to the target configuration set, and updating the current optimal path and shortest path distance;
a screening module configured to: screening a target configuration set, eliminating target configurations which do not meet preset conditions, and adding target configurations which meet the preset conditions;
a loop processing module configured to: circularly growing sampling points with different characteristics to a process of screening a target configuration set; and stopping circulation when a preset condition is reached to obtain a global optimal path.
The working method of the system is the same as the redundant mechanical arm path planning method based on the optimal target configuration screening in embodiment 1, and is not described herein again.
Example 3:
the present embodiment provides a computer-readable storage medium, on which a computer program is stored, and the program, when executed by a processor, implements the steps of the redundant robotic arm path planning method based on screening optimal target configurations described in embodiment 1.
Example 4:
the embodiment provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the steps of the redundant manipulator path planning method based on screening optimal target configurations according to embodiment 1 when executing the program.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and those skilled in the art can make various modifications and variations. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present embodiment should be included in the protection scope of the present embodiment.
Claims (10)
1. The redundant manipulator path planning method based on the screened optimal target configuration is characterized by comprising the following steps:
acquiring a joint value of the multi-degree-of-freedom mechanical arm and a target pose of an end effector; establishing a random expansion tree by taking the joint value of the multi-degree-of-freedom mechanical arm as a root node, and initializing the optimal path and the shortest path distance in the random expansion tree; establishing a target configuration set according to the target pose of the end effector;
growing sampling points with different characteristics according to the shortest distance;
growing a new growing point in the direction of the sampling point by the random expansion tree;
continuing to grow the new growing point to the target configuration set, and updating the current optimal path and shortest path distance;
screening a target configuration set, eliminating target configurations which do not meet preset conditions, and adding target configurations which meet the preset conditions;
circularly growing sampling points with different characteristics to a process of screening a target configuration set; and stopping circulation when a preset condition is reached to obtain a global optimal path.
2. The redundant manipulator path planning method based on screening of optimal target configurations as claimed in claim 1, wherein a target configuration set established according to the target pose of the end effector is a candidate optimal target configuration set, the element form in the candidate optimal target configuration set is (x, flag), x is a random inverse kinematics solution of the target pose of the end effector of the manipulator, and the flag logical bit indicates whether the x target configuration has been added into the node set of the random extended tree.
3. The method for planning the path of the redundant manipulator based on the screening of the optimal target configuration as claimed in claim 1, wherein the process of growing the sampling points of different characteristics according to the shortest distance comprises the following steps:
when the shortest path distance is infinite, a sampling point is obtained by configuring a space uniform sampling and matching with a target configuration direct sampling mode, and the method specifically comprises the following steps: generating a random number, uniformly sampling in a joint space of the mechanical arm to obtain a sampling point when the random number is smaller than a set threshold, and randomly selecting a certain target configuration in a target configuration set to be directly used as the sampling point when the random number is not smaller than the set threshold;
when the shortest path distance is less than infinity, sampling is performed in the superimposed hyperellipsoid sampling subspace, specifically: randomly selecting a certain element (x _ coarse, flag) in a target configuration set in a uniform probability mode, taking multi-degree-of-freedom mechanical arm joint values x _ begin and x _ coarse as focuses, and taking shortest path distance as the length of a long axis to sample a single hyperellipsoid sampling subspace according to uniform distribution; the sampling points are as follows: x _ sample ═ CLx unit +x center ，
Wherein C represents a rotation transformation matrix from a hyper-ellipsoid coordinate system to a world coordinate system, L represents a scaling transformation from a unit hyper-sphere to a hyper-ellipsoid, and x unit All expressed in a multi-dimensional unit hypersphereUniform sampling, x center Representing a translation transformation.
4. The method for planning the path of the redundant manipulator based on the screened optimal target configuration as claimed in claim 1, wherein a new growing point grows from a random expansion tree to a sampling point direction, and nodes in the field of the new growing point are optimized by selecting an optimal father node and a reconnection mode.
5. The method for planning the path of the redundant manipulator based on the screening of the optimal target configuration as claimed in claim 1, wherein the mode of continuing the growth of the new growth point to the target configuration set is as follows: traversing all elements of a target configuration set, if the elements (x, flag) meet the condition that flag is false, | | x _ new-x | | < step and x _ new is not more than x, and direct local paths from x _ new to x have no collision, adding a new node x into a node set, and adding edges (x _ new, x) into a connection relation set between nodes; step represents a constant of the maximum growth step of the node, x _ new represents that a new growth point is set, and false represents a feasible path;
the method for updating the distance between the current optimal path and the shortest path comprises the following steps: traversing the target configuration set, if the flag in the element (x, flag) is true, obtaining temp _ path by x in the random expansion tree edge set in a backtracking mode, and obtaining the value of temp _ path by x
And calculating the path length temp _ cbest of the temp _ path, wherein d represents the element number of the temp _ path sequence, comparing the minimum values of the path lengths corresponding to all the elements meeting the condition that the flag is true, assigning the minimum values to the shortest path distance, and assigning the path obtained by the elements to the optimal path.
6. The method for redundant robotic arm path planning based on screening optimal target configurations of claim 5, wherein screening the set of target configurations comprises:
when the shortest path ratio is smaller than the value of the last cycle, traversing all elements of the target configuration set, and if a certain element meets the condition that | | | x _ begin-x | | > is more than or equal to cbest, deleting the element;
when the total number of elements in the target configuration set is less than the maximum number of elements in the target configuration set, determining whether to add a new candidate configuration through a random strategy, specifically: generating a random number, and when the random number is not less than a set threshold, not adding a new target element to the target configuration set; when the random number is smaller than a set threshold value, calling a mechanical arm inverse kinematics algorithm to generate a random inverse kinematics solution x corresponding to a collision-free mechanical arm target pose, if | | | x _ begin-x | < cbest is met, adding the solution into a target configuration set, and if not, continuously generating a random joint inverse solution until meeting an adding condition; wherein cbest is the shortest distance.
7. The method for planning the path of the redundant manipulator based on the optimal target configuration screening as claimed in claim 1, wherein the method comprises the steps of judging whether the process from the sampling points with different characteristics to the target configuration screening set is operated circularly for a preset number of times or meets a preset circulating time, if not, continuing to circulate, if so, the output optimal path is the optimal path of the whole project, the head element in the optimal path sequence is a given initial configuration, and the tail element is the optimal target configuration after screening.
8. Redundant mechanical arm path planning system based on screening optimal target configuration, its characterized in that includes:
a data processing module configured to: acquiring a joint value of the multi-degree-of-freedom mechanical arm and a target pose of an end effector; establishing a random expansion tree by taking the joint value of the multi-degree-of-freedom mechanical arm as a root node, and initializing the optimal path and the shortest path distance in the random expansion tree; establishing a target configuration set according to the target pose of the end effector;
a sample point acquisition module configured to: growing sampling points with different characteristics according to the shortest distance;
a new growth point acquisition module configured to: growing a new growing point in the direction of the sampling point by the random expansion tree;
an update module configured to: continuing to grow the new growing point to the target configuration set, and updating the current optimal path and shortest path distance;
a screening module configured to: screening a target configuration set, eliminating target configurations which do not meet preset conditions, and adding target configurations which meet the preset conditions;
a loop processing module configured to: circularly growing sampling points with different characteristics to a process of screening a target configuration set; and stopping circulation when a preset condition is reached to obtain a global optimal path.
9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the redundant robotic arm path planning method based on screening optimal target configurations according to any one of claims 1-7.
10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program performs the steps of the redundant robotic arm path planning method based on screening optimal target configurations according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210202596.8A CN115122317A (en) | 2022-03-02 | 2022-03-02 | Redundant manipulator path planning method and system based on optimal target configuration screening |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210202596.8A CN115122317A (en) | 2022-03-02 | 2022-03-02 | Redundant manipulator path planning method and system based on optimal target configuration screening |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115122317A true CN115122317A (en) | 2022-09-30 |
Family
ID=83376455
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210202596.8A Pending CN115122317A (en) | 2022-03-02 | 2022-03-02 | Redundant manipulator path planning method and system based on optimal target configuration screening |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115122317A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115990877A (en) * | 2022-12-08 | 2023-04-21 | 北京航天飞行控制中心 | Multi-target operation sequence planning method and device for space manipulator and electronic equipment |
| CN116901086A (en) * | 2023-09-13 | 2023-10-20 | 千里眼(广州)人工智能科技有限公司 | Intelligent algorithm-based control method for optimizing industrial robot track |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113103236A (en) * | 2021-04-22 | 2021-07-13 | 山东大学 | Rapid and gradual optimal mechanical arm obstacle avoidance path planning method |
| CN113352319A (en) * | 2021-05-08 | 2021-09-07 | 上海工程技术大学 | Redundant mechanical arm obstacle avoidance trajectory planning method based on improved fast expansion random tree |
| CN113485367A (en) * | 2021-08-06 | 2021-10-08 | 浙江工业大学 | Path planning method of multifunctional stage mobile robot |
-
2022
- 2022-03-02 CN CN202210202596.8A patent/CN115122317A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113103236A (en) * | 2021-04-22 | 2021-07-13 | 山东大学 | Rapid and gradual optimal mechanical arm obstacle avoidance path planning method |
| CN113352319A (en) * | 2021-05-08 | 2021-09-07 | 上海工程技术大学 | Redundant mechanical arm obstacle avoidance trajectory planning method based on improved fast expansion random tree |
| CN113485367A (en) * | 2021-08-06 | 2021-10-08 | 浙江工业大学 | Path planning method of multifunctional stage mobile robot |
Non-Patent Citations (2)
| Title |
|---|
| R. MASHAYEKHI, M. Y. I. IDRIS, M. H. ANISI, I. AHMEDY AND I. ALI: "Informed RRT*-Connect: An Asymptotically Optimal Single-Query Path Planning Method", 《IEEE ACCESS》, vol. 8, 24 January 2020 (2020-01-24), XP011770426, DOI: 10.1109/ACCESS.2020.2969316 * |
| 戴冰: "基于改进RRT算法的机械臂路径规划方法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》, 15 February 2022 (2022-02-15) * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115990877A (en) * | 2022-12-08 | 2023-04-21 | 北京航天飞行控制中心 | Multi-target operation sequence planning method and device for space manipulator and electronic equipment |
| CN115990877B (en) * | 2022-12-08 | 2024-06-07 | 北京航天飞行控制中心 | Multi-target operation sequence planning method and device for space manipulator and electronic equipment |
| CN116901086A (en) * | 2023-09-13 | 2023-10-20 | 千里眼(广州)人工智能科技有限公司 | Intelligent algorithm-based control method for optimizing industrial robot track |
| CN116901086B (en) * | 2023-09-13 | 2023-11-28 | 千里眼(广州)人工智能科技有限公司 | Intelligent algorithm-based control method for optimizing industrial robot track |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113103236B (en) | Rapid and gradual optimal mechanical arm obstacle avoidance path planning method | |
| CN115122317A (en) | Redundant manipulator path planning method and system based on optimal target configuration screening | |
| CN110962130B (en) | Heuristic RRT mechanical arm motion planning method based on target deviation optimization | |
| CN107234617B (en) | Obstacle avoidance path planning method guided by artificial potential field irrelevant to obstacle avoidance task | |
| Li et al. | An adaptive rapidly-exploring random tree | |
| CN112987799B (en) | Unmanned aerial vehicle path planning method based on improved RRT algorithm | |
| CN113885535B (en) | Impact constraint robot obstacle avoidance and time optimal track planning method | |
| CN109990787B (en) | Method for avoiding dynamic obstacle in complex scene by robot | |
| Shi et al. | Spark PRM: Using RRTs within PRMs to efficiently explore narrow passages | |
| Lindemann et al. | Steps toward derandomizing RRTs | |
| WO2023155923A1 (en) | Path planning method and apparatus and readable storage medium | |
| WO2019154944A1 (en) | Distributed machine learning system | |
| CN115248592A (en) | Multi-robot autonomous exploration method and system based on improved rapid exploration random tree | |
| CN113848889A (en) | Path planning method based on combination of whale optimization algorithm and artificial potential field method | |
| CN116852367A (en) | Six-axis mechanical arm obstacle avoidance path planning method based on improved RRTstar | |
| Golluccio et al. | Robotic weight-based object relocation in clutter via tree-based q-learning approach using breadth and depth search techniques | |
| Golluccio et al. | Objects relocation in clutter with robot manipulators via tree-based q-learning algorithm: Analysis and experiments | |
| CN115056222A (en) | Mechanical arm path planning method based on improved RRT algorithm | |
| CN112393739B (en) | Improved rapid search random tree path planning method in large-area environment | |
| Steffen et al. | Reducing the Dimension of the Configuration Space with Self Organizing Neural Networks | |
| Neumann et al. | Efficient continuous-time reinforcement learning with adaptive state graphs | |
| CN113011516A (en) | Three-dimensional mesh model classification method and device based on graph topology and storage medium | |
| Feng et al. | Accelerated RRT* by local directional visibility | |
| Saberifar et al. | Inconsequential improprieties: Filter reduction in probabilistic worlds | |
| CN118031989A (en) | Robot path planning method based on RRT algorithm improvement |
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 |