CN116038688A - Mechanical arm joint space path planning method based on probability virtual potential field guided bidirectional RRT algorithm - Google Patents
Mechanical arm joint space path planning method based on probability virtual potential field guided bidirectional RRT algorithm Download PDFInfo
- Publication number
- CN116038688A CN116038688A CN202211418931.4A CN202211418931A CN116038688A CN 116038688 A CN116038688 A CN 116038688A CN 202211418931 A CN202211418931 A CN 202211418931A CN 116038688 A CN116038688 A CN 116038688A
- Authority
- CN
- China
- Prior art keywords
- node
- path
- new
- random
- target
- 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 46
- 230000002457 bidirectional effect Effects 0.000 title claims abstract description 14
- 238000005070 sampling Methods 0.000 claims abstract description 29
- 238000013138 pruning Methods 0.000 claims description 15
- 238000001514 detection method Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 8
- 230000008859 change Effects 0.000 claims description 7
- 238000005457 optimization Methods 0.000 claims description 7
- 230000009471 action Effects 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000012952 Resampling Methods 0.000 claims description 2
- 238000007726 management method Methods 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 5
- 230000004888 barrier function Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/20—Instruments for performing navigational calculations
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J18/00—Arms
-
- 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
- B25J9/1666—Avoiding collision or forbidden zones
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Data Mining & Analysis (AREA)
- Mechanical Engineering (AREA)
- Robotics (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Geometry (AREA)
- Computational Mathematics (AREA)
- Algebra (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Automation & Control Theory (AREA)
- Pure & Applied Mathematics (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computer Hardware Design (AREA)
- Feedback Control In General (AREA)
Abstract
The invention provides a mechanical arm joint space path planning method based on a probability virtual potential field guiding bidirectional RRT algorithm, which integrates a target virtual potential field guiding strategy and a historical node biasing strategy before generating an initial collision-free path on the basis of a traditional bidirectional RRT algorithm, and improves path searching efficiency by utilizing potential field guiding of a target point and historical node information in a target random tree; and integrating a traditional RRT algorithm frame for optimizing the initial path after the initial collision-free path is generated, and respectively performing reselection father node operation and child node operation on nodes in two random trees based on sampling in a continuous variation interval near the nodes in the initial path to improve the path quality. The method is based on the frame of the RRT algorithm, ensures that the search is not in a local minimum state of a virtual potential field, simultaneously considers the smoothness of the track of the mechanical arm in running to the greatest extent, and the improved algorithm enables the mechanical arm to find a collision-free asymptotic optimal path in a working space on the premise of not sacrificing the search efficiency and ensuring the path quality.
Description
Technical Field
The invention belongs to the field of path planning of industrial mechanical arms, and particularly relates to a mechanical arm joint space planning method based on a probability virtual potential field guiding bidirectional RRT (PVFB-RRT) algorithm.
Background
The path planning of the mechanical arm is based on collision detection, and a collision-free path from an initial pose to a target pose is planned by using a discrete structure diagram G (V, E), wherein V is a set of nodes in the diagram, and each node corresponds to a configuration q; e is the collection of edges in the figure, each edge corresponds to a collision-free path between two configurations. The current common path planning algorithms mainly comprise the following three types: a combined path planning algorithm, represented by Dijkstra, astar based on a grid search strategy, etc.; local path planning methods, such as artificial potential field methods; sample-based path planning algorithms, representative are RRT, PRM, etc. algorithms. The combined path planning algorithm is not applicable because of extremely high time complexity and space complexity brought by the display representation of the obstacle space aiming at the object with high freedom degree such as the mechanical arm; the artificial potential field method provides an algorithm framework aiming at a high-dimensional object, but is extremely easy to fall into a local optimal state, so that the actual requirement is difficult to meet; sampling-based path planning algorithms, such as RRT, have the advantages of being very compact, not requiring the display of obstacle space, etc., but may also be inefficient in certain specific environments or objects, and the optimization process for path quality is time consuming. Therefore, it is significant to design a robot path planning method that improves the fast extended random tree (RRT) algorithm.
Aiming at the defects of the prior art, the invention improves the corresponding links of the traditional RRT from the two angles of quality and efficiency, greatly reduces the efficiency of random tree expansion and improves the quality of a final path, and enables the mechanical arm to find a collision-free path in a working space on the premise of not sacrificing search efficiency and ensuring path quality.
Disclosure of Invention
The invention aims to provide a mechanical arm joint space planning method based on a probability virtual potential field guided bidirectional RRT (PVFB-RRT) algorithm. The method aims at enabling the mechanical arm to find a gradually optimal collision-free path in the working space on the premise of not sacrificing searching efficiency and guaranteeing path quality.
In order to achieve the above objective, the present invention provides a robot arm joint space path planning method based on a probabilistic virtual potential field guiding bi-directional RRT (Probabilistic Virtual Potential Field Bidirectional-RRT, PVFB-RRT) algorithm, which is characterized by comprising the following steps:
step 1: determining joint configuration X of mechanical arm in initial state and target state s And X g Mechanical arm D-H parameter table, specific structural parameters, environment information and two random trees T respectively expanding from initial state and target state s And T g . Wherein x= { θ 1 ，θ 2 ,...，θ n And n represents the degree of freedom of the robot arm.
Step 2: setting a proper joint search step delta theta and a probability interval of a sampling mode as the basis of expansion of each step, setting a proper termination threshold as the final connection range of two random trees, and setting pruning radius and the maximum iteration number NUM of path updating to optimize the current path.
Step 3: and setting joint attraction coefficients, rejection coefficients and rejection field action ranges according to a sampling mode of target virtual potential field guidance.
Step 4: based on the traditional bidirectional RRT algorithm, the mechanical arm generates two random trees T from an initial state and a target state according to the strategy s And T g In the expansion process, a sampling mode is selected by using a method for generating random numbers between 0 and 1, and a new node is further generated through mechanical arm forward kinematics and collision detection.
Step 5: judging the two random trees T according to the termination threshold s And T g Whether the path is in a range capable of being directly connected or not, if the path is lower than a termination threshold value, stopping searching, and communicating an initial path; otherwise, the mechanical arm continues to be in accordance with the original strategy until the condition of the termination threshold is met.
Step 6: after the initial path is generated, a reselection father node and child node strategy of an RRT algorithm is used in the path optimization process, sampling is carried out in a continuous change interval near the node in the initial path, operations of reselecting father nodes and child nodes are respectively carried out on the nodes in the two random trees, and the path is updated once after repeated 50 times of execution. If the path cannot be communicated, restarting 50 times of parent node and child node reselection operations; if the paths can be communicated, outputting the updated paths as current paths, and continuing to repeat the operation.
Step 7: and outputting the current path after the maximum iteration number NUM of the path update is reached.
Furthermore, the step 1 aims to clearly plan the information such as the position, the shape and the size of the obstacle in the environment, and the like, and establishes a kinematic model of the corresponding mechanical arm by using the mechanical arm D-H parameter table.
Furthermore, the initialization parameter of the PVFB-RRT algorithm in step 2 plays a vital role in the selection of each step of node in the path search of the mechanical arm and the optimization of the node connecting line after the initial path generation.
Further, the purpose of step 3 is to initialize parameters for the sampling mode of virtual potential field guidance, so as to more fully utilize the historical nodes in the target random tree as guidance, thereby improving the path searching efficiency.
Further, the policy principles involved in the steps 4 to 6 are expressed as follows:
(1) Principle of bidirectional RRT algorithm: building two random trees T from initial state and target state s And T g In each step of expansion, T is used s And T g Respectively add a new node X new Up to the point, the following description will be given by T s Extended to example, T g The same applies to the extension. Random sampling in search space (excluding obstacle space) to obtain random node X rand As the target point X d The method comprises the steps of carrying out a first treatment on the surface of the Traversal T s Finding out T s Intermediate and target point X d Nearest neighbor node X near The method comprises the steps of carrying out a first treatment on the surface of the Along X near →X d Is extended by a prescribed step length to determine a new node X new The method comprises the steps of carrying out a first treatment on the surface of the Forward motion of mechanical arm based on joint angle in current stateLearning to obtain the positions of all joints and connecting rods; checking new node X for collision detection new Whether collision with an obstacle in the environment occurs, if not, X is determined new Deposit T s In which X is connected to near And X is new Obtaining new branches, switching random tree to repeat the above steps, otherwise resampling until no collision X exists new Is generated; the two random trees T are updated continuously in the above way s And T g Until T s And T g The nodes in the network are in a range capable of being directly connected, namely, the nodes are lower than a termination threshold value, and the initial path is generated.
(2) Target virtual potential field guiding strategy principle: firstly, it is clear that each joint of the mechanical arm has three different change trends of +delta theta and-delta theta at a certain moment, so that the joint has 3 total from the current joint configuration state to the next joint configuration state n A combination of species and configuration; secondly, the virtual force sources are respectively the attraction force of the target and the repulsion force of the obstacle; last traverse 3 in the collection n The combination of species and configurations and the calculation of the potential force values of all the possible configuration combinations, and the joint configuration group with the minimum potential force value is regarded as a new node X new 。
(3) Historical node bias policy principle: regarding the node with the minimum cost value with the current expanding random tree root node in the target random tree as a target point X d The method is used as a target guide to guide the random tree which is currently expanding to be rapidly expanded to the target random tree.
(4) Random sampling mode: randomly generating target points X d＝{θ d1 ，θ d2 ,...，θ dn Calculating the target point X d Euclidean distance between the node with the minimum distance and all nodes in the current random tree is taken as X near Each joint component of the mechanical arm is along X near →X d The direction of the corresponding component in the model is extended by a prescribed step length delta theta to obtain X new 。
(5) Based on the principle of a probabilistic virtual potential field guiding strategy: mainly solves the problem of new node X new Is a problem with the generation of (a). At the extended new node X new A random number between 0 and 1 is generated before,determining a sampling mode according to the interval where the random number is located, and further determining a new node X new The method can enable the expansion process of the random tree to have target guidance while having blind randomness, so that the search efficiency is accelerated. The sampling modes are random sampling, history node bias and target virtual potential field guiding as described in the above (2) to (4), namely each sampling mode comprises the following steps of
Is selected.
(6) Sampling in a continuous variation interval near an initial path: the change trend of each joint of the mechanical arm at a certain moment is regarded as a continuous interval [ -2delta theta, +2delta theta]Namely, the executable change trend of the configuration of a certain node in the generated path is a continuous interval, each joint randomly takes a value in the continuous interval, and the collision-free node after collision detection is taken as a new node X new And performing the operations of reselecting the father and the son, then attempting to update the path once after 50 operations of reselecting the father and the son, and optimizing the path within the specified iteration times.
(7) Reselection parent and child node policy principles: the new node is taken as the center, the pruning radius demarcates a re-planning range (namely, all nodes in the current tree are traversed, the distance value d between all nodes and the current new node is calculated, the node with the distance value smaller than the pruning radius is selected), the cost value of the nodes is added with d, and the node with the shortest path is selected as X new Of the new parent node, i.e. calculate the distance from the start point through the selected node to the new node X new Cost value of (2); similarly, the new node is used as the center, the pruning radius is used for defining a re-planning range (namely traversing all nodes in the current tree, calculating the distance value d between all nodes and the current new node, and selecting the node with the distance value smaller than the pruning radius), and the new node X is used for new Adding the cost value of (2) to d, comparing with the current cost value of each node, and if the cost value is smaller than the current cost value, adding a new node X new As a new parent node of the node, i.e. calculating the distance from the start point through the new node X new Cost value to the selected node.
Further, the step 7 is used as a final complete collision-free path from the initial pose to the target pose.
Compared with the prior art, the invention has the advantages that: the method is characterized in that the corresponding links of the traditional RRT are improved from the two angles of quality and efficiency, the angle of efficiency corresponds to the mode of generating random numbers between 0 and 1 before the initial path is generated, the random of the traditional RRT algorithm is reserved, the sampling mode is optimized by referring to the thought of an artificial potential field method, and the initial path is communicated with the optimal quality rapidly. And after the quality optimization angle corresponds to the initial path generation, the strategy of reselecting the father node and the child node by utilizing the RRT algorithm is utilized to optimize the communicated paths, so that the aim of gradual optimization is fulfilled in the process of limited iterations. The improved algorithm enables the mechanical arm to find a gradually optimal collision-free path in the working space on the premise of not sacrificing searching efficiency and guaranteeing path quality.
Drawings
Fig. 1 is a flowchart of a robot arm path planning method of PVFB-RRT algorithm.
Fig. 2 is a schematic diagram of a target virtual potential field guiding strategy in PVFB-RRT algorithm.
Fig. 3 is a schematic diagram of a policy principle of a parent node reselection in the PVFB-RRT algorithm.
Fig. 4 is a schematic diagram of a reselection sub-node policy in PVFB-RRT algorithm.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be further described below.
The invention aims to provide a mechanical arm joint space planning method based on a probability virtual potential field guided bidirectional RRT (PVFB-RRT) algorithm, which comprises the following steps as shown in figure 1:
step 1: determining joint configuration X of mechanical arm in initial state and target state s And X g Mechanical arm D-H parameter table, specific structural parameters, environment information and two random trees T respectively expanding from initial state and target state s And T g . Wherein x= { θ 1 ，θ 2 ,...，θ n And n represents the degree of freedom of the robot arm.
Step 2: setting a proper joint search step delta theta and a probability interval of a sampling mode as the basis of expansion of each step, setting a proper termination threshold as the final connection range of two random trees, and setting pruning radius and the maximum iteration number NUM of path updating to optimize the current path.
Step 3: and setting joint attraction coefficients, rejection coefficients and rejection field action ranges according to a sampling mode of target virtual potential field guidance.
Step 4: building two random trees T from initial state and target state s And T g The following description is given of T s Extended to example, T g The same applies to the extension. Generating a random number between 0 and 1, and judging the interval in which the random number is located: if in a state of
Then the new node X is directly selected by adopting the potential field guiding strategy of the target point new The method comprises the steps of carrying out a first treatment on the surface of the If at->
Then adopting the principle of a historical node bias strategy to enable the node X with the minimum cost value with the current random tree root node in the target random tree nearest As the target point X d Further get new node X new The method comprises the steps of carrying out a first treatment on the surface of the If at->
Then a random node is obtained as target point X in a random sampling manner d Further get new node X new . The potential value calculation formula in the potential field guiding strategy of the target point is explained as follows:
attraction potential force V a And (3) calculating:
V a ＝η a ·|X-X target | 2
wherein eta a Represents the suction coefficient, X represents a group of joint configurations, X target Representing the target configuration, i.e. T s And T g Joint configuration at root node of (a), along with the random tree T s And T g Is switched by the switching of (a).
Repulsive potential V r And (3) calculating:
wherein eta r Represents the rejection coefficient, d 0 Indicating the repulsive field action range, d st Representing the minimum distance between the surface of the s-th joint arm and the surface of the t-th obstacle, V rst Represents the repulsive force of the t barrier to the s joint, V r Representing the sum of the repulsive forces of all m obstacles against a set of configurations.
Total potential V of a set of joint configurations X:
V＝ω 1 V a +ω 2 V r
wherein omega 1 And omega 2 Representing the weight coefficient.
Step 5: for newly generated new node X new Forward kinematics and collision detection are performed, and if no collision occurs, the new node X is obtained new The communication path is stored in T s In (a) and (b); if collision occurs, returning to the step 4.
Step 6: checking whether the current state meets the termination threshold condition, if so, outputting an initial path, and continuously executing the step 7; if not, switching the random tree, and returning to the step 4.
Step 7: after the initial path is generated, path optimization is performed, and the path update cycle variable j=1.
Step 8: the parent and child node loop variable k=1 is reselected.
Step 9: sampling in the continuous variation interval near the node in the current path to obtain a new node X knew That is, the executable variation trend of the configuration of a certain node in the current path is continuous interval [ -2Δθ, +2Δθ]。
Step 10: reselecting the father node: as shown in fig. 3, the new node is used as the center, the pruning radius demarcates the re-planning range (i.e. all nodes in the current tree are traversed, the distance value d between all nodes and the current new node is calculated, the node with the distance value smaller than the pruning radius is selected), the cost value of the nodes is added with d, and the node with the shortest path is selected as X knew Of the new parent node of (a), i.e. calculate the distance from the start point through the selected node to the new node X knew Cost value of (a).
Step 11: reselecting the child node: the new node is taken as the center, the pruning radius demarcates a re-planning range (namely, all nodes in the current tree are traversed, the distance value d between all nodes and the current new node is calculated, and the node with the distance value smaller than the pruning radius is selected), and the new node X is selected knew Adding the cost value of (2) to d, comparing with the current cost value of each node, and if the cost value is smaller than the current cost value, adding a new node X knew As a new parent node of the node, i.e. calculating the distance from the start point through the new node X knew Cost value to the selected node.
Step 12: after executing the steps 10 to 11, k++, if k is less than 50, returning to the step 9; otherwise j++, go to step 13.
Step 13: attempting to update the current path, if the path cannot be communicated, restarting 50 times of parent and child node reselection operations, namely returning to the step 8; if the path can be communicated, updating the path and returning to the step 8 after being used as the current path.
Step 14: and (3) repeatedly executing the steps 8-13, continuously updating the nodes in the path until the nodes in the path are traversed, and finally outputting a progressively optimal collision-free path.
The foregoing is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Any person skilled in the art will make any equivalent substitution or modification to the technical solution and technical content disclosed in the invention without departing from the scope of the technical solution of the invention, and the technical solution of the invention is not departing from the scope of the invention.
Claims (9)
1. The mechanical arm joint space path planning method based on the probability virtual potential field guided bidirectional RRT algorithm is characterized by comprising the following steps of:
step 1: determining joint configuration X of mechanical arm in initial state and target state s And X g Mechanical arm D-H parameter table, specific structural parameters, environment information and two random trees T respectively expanding from initial state and target state s And T g Wherein x= { θ 1 ，θ 2 ，...，θ n N represents the degree of freedom of the mechanical arm;
step 2: setting a proper joint search step delta theta and a probability interval of a sampling mode as the basis of expansion of each step, setting a proper termination threshold as the final connection range of two random trees, and setting pruning radius and the maximum iteration number NUM of path updating to optimize the current path;
step 3: setting joint attraction coefficients, rejection coefficients and rejection field action ranges according to a sampling mode of target virtual potential field guidance;
step 4: based on the traditional bidirectional RRT algorithm, the mechanical arm generates two random trees T from an initial state and a target state according to the strategy s And T g Selecting a sampling mode by using a method for generating random numbers between 0 and 1 in the expansion process, and further generating new nodes through mechanical arm forward kinematics and collision detection;
step 5: judging the two random trees T according to the termination threshold s And T g Whether the path is in a range capable of being directly connected or not, if the path is lower than a termination threshold value, stopping searching, and communicating an initial path; otherwise, the mechanical arm continues to be in accordance with the original strategy until the condition of the termination threshold is met;
step 6: after the initial path is generated, a reselection father node and child node strategy of an RRT algorithm is used in the path optimization process, sampling is carried out in a continuous change interval near the node in the initial path, operations of reselecting father nodes and child nodes are respectively carried out on the nodes in the two random trees, and the path is updated once after repeated 50 times of execution; if the paths cannot be communicated, restarting 50 times of parent node and child node reselection operations; if the paths can be communicated, outputting the updated paths as current paths, and continuing to repeat the operation;
step 7: and outputting the current path after the maximum iteration number NUM of the path update is reached.
2. The robot arm joint space path planning method based on the probabilistic virtual potential field guided bidirectional RRT algorithm of claim 1, wherein the bidirectional RRT algorithm principle: building two random trees T from initial state and target state s And T g In each step of expansion, T is used s And T g Respectively add a new node X new Up to the point, the following description will be given by T s Extended to example, T g Expansion and management; random sampling in search space (excluding obstacle space) to obtain random node X rand As the target point X d The method comprises the steps of carrying out a first treatment on the surface of the Traversal T s Finding out T s Intermediate and target point X d Nearest neighbor node X near The method comprises the steps of carrying out a first treatment on the surface of the Along X near →X d Is extended by a prescribed step length to determine a new node X new The method comprises the steps of carrying out a first treatment on the surface of the Performing forward kinematics of the mechanical arm based on the joint angle in the current state to obtain the positions of all joints and connecting rods; checking new node X for collision detection new Whether collision with an obstacle in the environment occurs, if not, X is determined new Deposit T s In which X is connected to near And X is new Obtaining new branches, switching random tree to repeat the above steps, otherwise resampling until no collision X exists new Is generated; the two random trees T are updated continuously in the above way s And T g Until T s And T g The nodes in the network are in a range capable of being directly connected, namely, the nodes are lower than a termination threshold value, and the initial path is generated.
3. The robot arm joint space path planning method based on the probabilistic virtual potential field guiding bi-directional RRT algorithm of claim 1, wherein the target virtual potential field guiding strategy principle: firstly, it is clear that each joint of the mechanical arm has three different changes at a certain momentTrends are +Δθ, hold motionless, - Δθ, so there are a total of 3n configuration combinations from the current joint configuration state to the next joint configuration state; secondly, the virtual force sources are respectively the attraction force of the target and the repulsion force of the obstacle; last traverse 3 in the collection n The combination of species and configurations and the calculation of the potential force values of all the possible configuration combinations, and the joint configuration group with the minimum potential force value is regarded as a new node X new 。
4. The method for planning a space path of a robot joint based on a probabilistic virtual potential field guided bi-directional RRT algorithm according to claim 1, wherein the historical node bias policy principle: regarding the node with the minimum cost value with the current expanding random tree root node in the target random tree as a target point X d The method is used as a target guide to guide the random tree which is currently expanding to be rapidly expanded to the target random tree.
5. The method for planning a space path of a robot joint based on a probabilistic virtual potential field guided bi-directional RRT algorithm according to claim 1, wherein the random sampling manner is as follows: randomly generating target point X d ＝{θ d1 ，θ d2 ，...，θ dn Calculating the target point X d Euclidean distance between the node with the minimum distance and all nodes in the current random tree is taken as X near Each joint component of the mechanical arm is along X near →X d The direction of the corresponding component in the model is extended by a prescribed step length delta theta to obtain X new 。
6. The robot arm joint space path planning method based on the probabilistic virtual potential field guiding bi-directional RRT algorithm of claim 1, wherein the probabilistic virtual potential field guiding policy principle is based on: mainly solves the problem of new node X new A generation problem of (2); at the extended new node X new Generating random numbers between 0 and 1 before, determining a sampling mode according to the interval where the random numbers are located, and further determining a new node X new The expanding process of the random tree can be enabled to have blind randomness and simultaneously haveThe target guidance is achieved so as to accelerate the searching efficiency; the sampling modes are random sampling, historical node bias and target virtual potential field guidance, and each sampling mode is provided with
Is selected.
7. The method for planning a spatial path of a robot joint based on a probabilistic virtual potential field guided bi-directional RRT algorithm according to claim 1, wherein the continuously variable interval around the initial path is sampled: the change trend of each joint of the mechanical arm at a certain moment is regarded as a continuous interval [ -2delta theta, +2delta theta]Namely, the executable change trend of the configuration of a certain node in the generated path is a continuous interval, each joint randomly takes a value in the continuous interval, and the collision-free node after collision detection is taken as a new node X new And performing the operations of reselecting the father and the son, then attempting to update the path once after 50 operations of reselecting the father and the son, and optimizing the path within the specified iteration times.
8. The robot arm joint space path planning method based on the probabilistic virtual potential field guided bidirectional RRT algorithm of claim 1, wherein the reselection parent-child node policy principle: the new node is taken as the center, the pruning radius demarcates a re-planning range (namely, all nodes in the current tree are traversed, the distance value d between all nodes and the current new node is calculated, the node with the distance value smaller than the pruning radius is selected), the cost value of the nodes is added with d, and the node with the shortest path is selected as X new Of the new parent node, i.e. calculate the distance from the start point through the selected node to the new node X new Cost value of (2); similarly, the new node is used as the center, the pruning radius is used for defining a re-planning range (namely traversing all nodes in the current tree, calculating the distance value d between all nodes and the current new node, and selecting the node with the distance value smaller than the pruning radius), and the new node X is used for new The cost value of (c) is added to d, and compared with the current cost value of each node, if less than the current cost value,then new node X will be new As a new parent node of the node, i.e. calculating the distance from the start point through the new node X new Cost value to the selected node.
9. The method for planning a spatial path of a robot joint based on a probabilistic virtual potential field guided bi-directional RRT algorithm according to claim 1, wherein the step 7 is performed as a final complete collision-free path from the starting pose to the target pose.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211418931.4A CN116038688A (en) | 2022-11-14 | 2022-11-14 | Mechanical arm joint space path planning method based on probability virtual potential field guided bidirectional RRT algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211418931.4A CN116038688A (en) | 2022-11-14 | 2022-11-14 | Mechanical arm joint space path planning method based on probability virtual potential field guided bidirectional RRT algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116038688A true CN116038688A (en) | 2023-05-02 |
Family
ID=86130299
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211418931.4A Pending CN116038688A (en) | 2022-11-14 | 2022-11-14 | Mechanical arm joint space path planning method based on probability virtual potential field guided bidirectional RRT algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116038688A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117168483A (en) * | 2023-09-01 | 2023-12-05 | 哈尔滨理工大学 | Unmanned vehicle path planning method considering map complexity |
| CN117470252A (en) * | 2023-12-28 | 2024-01-30 | 中闽(福清)风电有限公司 | RRT algorithm-based wind farm booster station robot inspection local path planning method |
-
2022
- 2022-11-14 CN CN202211418931.4A patent/CN116038688A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117168483A (en) * | 2023-09-01 | 2023-12-05 | 哈尔滨理工大学 | Unmanned vehicle path planning method considering map complexity |
| CN117168483B (en) * | 2023-09-01 | 2024-05-14 | 哈尔滨理工大学 | Unmanned vehicle path planning method considering map complexity |
| CN117470252A (en) * | 2023-12-28 | 2024-01-30 | 中闽(福清)风电有限公司 | RRT algorithm-based wind farm booster station robot inspection local path planning method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116038688A (en) | Mechanical arm joint space path planning method based on probability virtual potential field guided bidirectional RRT algorithm | |
| CN111546347B (en) | Mechanical arm path planning method suitable for dynamic environment | |
| CN110228069B (en) | Online obstacle avoidance motion planning method for mechanical arm | |
| CN110497403A (en) | A kind of manipulator motion planning method improving two-way RRT algorithm | |
| CN110609547B (en) | Mobile robot planning method based on visual map guidance | |
| CN109343345B (en) | Mechanical arm polynomial interpolation track planning method based on QPSO algorithm | |
| CN110515094B (en) | Robot point cloud map path planning method and system based on improved RRT | |
| CN113103236A (en) | Rapid and gradual optimal mechanical arm obstacle avoidance path planning method | |
| CN111678523B (en) | Rapid BI _ RRT obstacle avoidance trajectory planning method based on STAR algorithm optimization | |
| Nieuwenhuisen et al. | An effective framework for path planning amidst movable obstacles | |
| CN109990787B (en) | Method for avoiding dynamic obstacle in complex scene by robot | |
| CN115723129B (en) | Continuous operation motion planning method for mechanical arm | |
| CN110275528B (en) | Improved path optimization method for RRT algorithm | |
| CN113858210B (en) | Mechanical arm path planning method based on hybrid algorithm | |
| CN109211242B (en) | Three-dimensional space multi-target path planning method integrating RRT and ant colony algorithm | |
| Przybylski et al. | D* Extra Lite: A dynamic A* with search–tree cutting and frontier–gap repairing | |
| CN115958590A (en) | RRT-based mechanical arm deep frame obstacle avoidance motion planning method and device | |
| CN110705803B (en) | Route planning method based on triangle inner center guide RRT algorithm | |
| Xue et al. | Hybrid bidirectional rapidly-exploring random trees algorithm with heuristic target graviton | |
| CN116652932A (en) | Mechanical arm autonomous obstacle avoidance method based on improved RRT algorithm | |
| CN113204238B (en) | Path planning method, equipment and storage medium for mobile robot | |
| CN115167388A (en) | RRT multi-robot formation path planning algorithm based on target guidance | |
| Zhang et al. | Improve RRT algorithm for path planning in complex environments | |
| CN113400303A (en) | Six-axis robot fruit and vegetable picking path planning method based on RRT (recursive least squares) algorithm | |
| Yan et al. | Weighted multi-tree RRT algorithm for efficient path-planning of mobile robots |
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 |