A robot obstacle avoidance method based on piecewise iterative drag curve
By using a segmented iterative towing curve method and employing pure geometric principles for obstacle avoidance path planning for robots with redundant degrees of freedom, the problem of high computational load and high performance requirements of existing algorithms is solved, and efficient and smooth robot motion is achieved.
Patent Information
- Application Number
- CN202411229401.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-03
- Publication Date
- 2025-11-18
- Estimated Expiration
- 2044-09-03
AI Technical Summary
Existing obstacle avoidance algorithms for redundant degrees of freedom robots require large amounts of environmental data sets. They use intelligent algorithms such as neural networks and genetic algorithms for obstacle avoidance path planning, which involves a large amount of computation and places high demands on the system's hardware and software performance.
A piecewise iterative dragline-based obstacle avoidance algorithm is designed by using pure geometric principles for obstacle avoidance path planning. The algorithm derives the dragline equation, performs recursive calculations, and sets the obstacle collision detection area using an artificial potential field.
It reduces computational load, improves computational efficiency, meets the real-time requirements of robot systems, makes robot movements smoother and more natural, and consumes less energy.
Smart Images
Figure CN119126794B_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of robot obstacle avoidance path planning, in particular to a robot obstacle avoidance method based on segmented iterative drag curve. BACKGROUND
[0002] The narrow and obstacle-laden working environment such as aero-engine internal inspection, post-disaster rescue and tracheal intubation medical operation is completed by a robot with redundant degrees of freedom, which is characterized by the ability to change the pose of the robot arm autonomously according to environmental information to complete obstacle avoidance and finally to work inside the complex environment. The traditional obstacle avoidance algorithm for robots with redundant degrees of freedom generally requires a large number of environmental data sets, and then uses intelligent algorithms such as neural networks and genetic algorithms to plan the obstacle avoidance path, which has a large amount of calculation and requires high performance of the system hardware and software. It is a technical problem to be solved urgently to explore a pure geometric obstacle avoidance algorithm to improve the calculation efficiency, reduce unnecessary movement of the robot and improve the stability of the movement.
[0003] The existing obstacle avoidance algorithm for robots with redundant degrees of freedom generally requires a large number of environmental data sets, and then uses intelligent algorithms such as neural networks and genetic algorithms to plan the obstacle avoidance path, which has a large amount of calculation and requires high performance of the system hardware and software.
[0004] In view of the above problems, a robot obstacle avoidance method based on segmented iterative drag curve is proposed. SUMMARY
[0005] The present application aims to provide a robot obstacle avoidance method based on segmented iterative drag curve, which uses pure geometric principles to work, thereby solving the problem of the obstacle avoidance algorithm for robots with redundant degrees of freedom requiring a large number of environmental data sets, and then using intelligent algorithms such as neural networks and genetic algorithms to plan the obstacle avoidance path, which has a large amount of calculation and requires high performance of the system hardware and software.
[0006] To achieve the above object, the present application provides the following technical scheme: a robot obstacle avoidance method based on segmented iterative drag curve, comprising the following implementation steps:
[0007] S1: deriving a drag curve equation;
[0008] S2: performing recursive operation on the drag curve equation;
[0009] S3: setting the obstacle collision detection area by using artificial potential field, and proposing a segmented iterative drag curve obstacle avoidance method.
[0010] Further, the specific steps of deriving the drag curve equation in S1 are as follows:
[0011] S101: Given the length of the connecting rod L, along the z-axis, the end B of the connecting rod is translated along the y-axis, and the trajectory of the other end A is the tractrix. According to the tractrix formation condition, the velocity vector of the end A always points to the B along the connecting rod L, that is, it is aligned with the tangent of the tractrix. The tractrix equation can be derived from the tangent differential equation of the end A of the connecting rod L as follows:
[0012]
[0013] Further, the specific steps of recursive operation on the tractrix equation in S2 are as follows:
[0014] S201: (y, z) is the coordinate of point A, L is the length of the connecting rod, and the closed form calculation is used for the tractrix equation to solve the inverse kinematics problem of multi-joint super-redundant degree of freedom robot, so as to realize the more natural and smooth motion of super-redundant robot. The formula obtained is as follows:
[0015]
[0016] Further, in S3, the obstacle collision detection area is set by using artificial potential field, and a segmented iterative tractrix obstacle avoidance method is proposed. The specific steps are as follows:
[0017] S301: Virtually separate the super-redundant robot joint that does not meet the obstacle avoidance condition;
[0018] S302: By discretizing the virtual joint, the super-redundant robot meets the obstacle avoidance condition, and the end effector P1 can track the target point;
[0019] S303: The discrete joint is combined again by the bidirectional iterative tractrix method.
[0020] Further, in S303, the discrete joint is combined again by the bidirectional iterative tractrix method. The specific steps are as follows:
[0021] S3031: Redefine the position update direction and iteration step;
[0022] S3032: Calculate whether the super-redundant robot meets the obstacle avoidance condition, and the formula is as follows:
[0023] ||P tar -P obs ||≤ε0.
[0024] Further, in S3032, the specific steps of calculating whether the super-redundant robot meets the obstacle avoidance condition are as follows:
[0025] S30321: Assume that the virtual discrete joint P 3a Without considering the obstacle avoidance condition, the position update direction is:
[0026]
[0027] S30322: When the virtual discrete joint P 3a The repulsive force F rep and the attractive force F att are defined according to the artificial potential field, the update direction D 3a of the iterative trajectory of the hyper-redundant robot in the segment P 3a 1 to P 3a is recalculated, and the repulsive force of the P rep3a point update iterative motion is F r3a =(y r3a , z att3a ), the attractive force is F a3a =(y a3a , z 3a ), and the resultant force F 3a is:
[0028] F rep3a +F att3a =(y r3a +y a3a , z r3a +z a3a );
[0029]
[0030] Further, the linear equation in the update direction is given by the point-slope equation:
[0031]
[0032] Further, in order to achieve fast convergence, the iterative step length is calculated by using the closest distance point (vertical foot) of the virtual joint P3b(y3b, z3b) to the point-slope equation:
[0033]
[0034] Where j is the iteration number, and the calculation formula of the iterative step length is:
[0035] d 3aj =sqrt((y 3a -y 3aj ) 2 +(z 3a -z 3aj ) 2 )。
[0036] Further, it is assumed that P 3a and P 3b overlap after the jth iteration, and the iteration termination condition is set as:
[0037]
[0038] Further, when the distance Δl3 between the two virtual discrete joints is less than a set value δ3 and the obstacle avoidance condition is met, the iteration ends, at this time, the inverse kinematics solution (α i ,β i ) is obtained, and the next obstacle avoidance motion planning is started, and finally the target position is reached.
[0039] Compared with the prior art, the present application has the following beneficial effects:
[0040] 1. The present application extends and innovates the geometric characteristics of the tractrix, proposes a bidirectional segmented iterative tractrix principle, and designs a new obstacle avoidance algorithm for multi-joint redundant degree robots, realizing adaptive obstacle avoidance motion planning for multi-joint redundant degree robots.
[0041] 2. The obstacle avoidance algorithm uses a purely geometric calculation method, which does not require a large number of iterative calculations like neural networks, improves the calculation efficiency, meets the real-time requirements of the robot system, and the geometric characteristics of the tractrix gradual convergence also make the robot motion form more smooth and natural, with low energy consumption. BRIEF DESCRIPTION OF DRAWINGS
[0042] Figure 1 is the iterative tractrix traction motion principle of the present application;
[0043] Figure 2 is the discrete bidirectional iterative tractrix obstacle avoidance principle of the present application (a) the i-th trajectory position (b) the j-th position update iteration (c) the (j+1)-th position update iteration (d) the (i+1)-th trajectory position;
[0044] Figure 3 is the bidirectional iterative tractrix flowchart of the present application;
[0045] Figure 4 is the overall flowchart of the present application;
[0046] Figure 5 is the specific step flowchart for deriving the tractrix equation of the present application;
[0047] Figure 6 is the specific step flowchart for performing recursive operation on the tractrix equation of the present application;
[0048] Figure 7 is the specific step flowchart for setting the obstacle collision detection area using artificial potential field and proposing a segmented iterative tractrix obstacle avoidance method of the present application;
[0049] Figure 8The specific step flow chart for recombining the discrete joints by the bidirectional iterative tractrix method of the application;
[0050] Figure 9 The specific step flow chart for calculating whether the super-redundant robot meets the obstacle avoidance condition. DETAILED DESCRIPTION
[0051] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by a person of ordinary skill in the art without creative effort fall within the protection scope of the application.
[0052] To solve the technical problem that the obstacle avoidance algorithm of the robot with redundant degrees of freedom needs a large amount of environment data set, and then the obstacle avoidance path is planned by using intelligent algorithms such as a neural network and a genetic algorithm, and the calculation amount is large, and the system software and hardware performance requirements are high, such as Figures 1-9 As shown in the following preferred technical solutions: The tractrix was first proposed by Huygens in 1692, also known as "tracing curve". As shown in the following preferred technical solutions: Figure 1 As shown in the following preferred technical solutions: The tractrix was first proposed by Huygens in 1692, also known as "tracing curve". As shown in the following preferred technical solutions: A single link on a plane is taken as an example to introduce the tractrix. A link with a length of L is placed along the z axis. One end B of the link is translated along the y axis, and then the trajectory of the other end A is the tractrix. According to the tractrix formation condition, the velocity vector of the end A always points to the B along the link L, that is, aligns with the tangent of the tractrix. The tangent differential equation of the end A of the link L can be used to derive the tractrix equation as follows:
[0053]
[0054] In the formula, (y, z) is the coordinate of point A, and L is the length of the link. The closed form calculation of formula (1) can obtain:
[0055]
[0056] The recursive operation on the tractrix can be used to solve the inverse kinematics problem of the multi-joint super-redundant degree of freedom robot, so that the super-redundant robot can move more naturally and smoothly.
[0057] In order to realize the adaptive obstacle avoidance control of the super-redundant robot, the patent uses the obstacle collision detection area set by the artificial potential field, and proposes a segmented iterative tractrix obstacle avoidance method. It is assumed that the initial position of the super-redundant robot is as shown in the following preferred technical solutions: Figure 2(a) shown, the known target point H, obstacle center O, dangerous area radius r1 and r2, and the hyper-redundant robot end effector point P. Due to the existence of obstacles on the motion path, the simple drag curve traction motion, the robot cannot safely reach the target position. In order to achieve obstacle avoidance during the motion of the hyper-redundant robot, the hyper-redundant robot joint point that does not meet the obstacle avoidance condition can be virtually separated, such as Figure 2 (b) shown. By discrete virtual joints, the hyper-redundant robot meets the obstacle avoidance condition, and the end effector P1 can track the target point. However, in fact, the hyper-redundant robot is not discrete, so the present application proposes a bidirectional iterative drag curve method to combine the discrete joints again. As shown in Figure 2 (b) shown, assuming that the P3 joint does not meet the obstacle avoidance condition at the jth position update, it is actually separated into two points P 3a and P 3b respectively. Then the discrete joint is approximated using the bidirectional iterative drag curve method.
[0058] The key problem of discrete virtual joint overlap is to redefine the position update direction and iteration step size. For this purpose, the specific flow chart of the discrete bidirectional iterative drag curve is designed as shown in Figure 3 First, calculate whether the hyper-redundant robot meets the obstacle avoidance condition:
[0059] ||P tar -P obs ||≤ε0 (3).
[0060] Where P tar is the point on the robot joint and hinge, P obs is the center point of the obstacle, and ε0 is the obstacle detection radius. If the distance between P ta and P obs is less than ε0, the segmented iterative drag curve obstacle avoidance operation is started. Otherwise, the bidirectional iterative drag curve motion planning is started.
[0061] Assuming that the virtual discrete joint P 3a does not consider the obstacle avoidance condition, the position update direction is:
[0062]
[0063] That is, the hyper-redundant robot discrete joint point P 3a moves along the line connecting P 3a and P 3b to P 3b , and the iteration step size d 3aj is:
[0064] d 3aj =sqrt((y 3a -y3b ) 2 +(z 3a -z 3b ) 2 ) (5)。
[0065] When the virtual discrete joint P 3a needs to consider the obstacle avoidance condition. According to the repulsive force F rep and attractive force F att defined by the artificial potential field, the update direction D 3a of the iterative trajectory of the hyper-redundant robot from P1 to P 3a is recalculated. According to the artificial potential field method, the repulsive force of the P 3a point update iteration motion is F rep3a =(y r3a , z r3a ), and the attractive force is F att3a =(y a3a , z a3a ), and the resultant force F 3a is:
[0066] F 3a =F rep3a +F att3a =(y r3a +y a3a , z r3a +z a3a ) (6);
[0067]
[0068] The linear equation in the update direction is given by the point-slope equation:
[0069]
[0070] In order to realize fast convergence, the iterative step length is calculated by using the closest distance point (vertical foot) of the virtual joint P3b (y3b, z3b) to the linear equation formula (8):
[0071]
[0072] Where j is the iteration number. The calculation formula of the iterative step length is:
[0073] d 3aj =sqrt((y 3a -y 3aj ) 2 +(z 3a -z 3aj ) 2 (10)。
[0074] In the hyper-redundant robot from P1 to P 3aAfter the position is updated, the super-redundant robot will then be moved from P. 3b Update to P5, such as Figure 2 As shown in (c), the update process is related to the robot's movement from P1 to P... 3a The update process is similar and will not be repeated here. Assume that after the j-th iteration, P... 3a and P 3b Overlap, and set the iteration termination condition to:
[0075]
[0076] In other words, the iteration ends when the distance Δl3 between the two virtual discrete joints is less than the set value δ3 and the obstacle avoidance condition is met, at which point the inverse kinematic solution (α) is obtained. i ,β i Then, it begins planning the next obstacle avoidance maneuver, eventually reaching the target location.
[0077] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.
[0078] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. A robot obstacle avoidance method based on piecewise iterative torsion lines, characterized in that: The implementation steps include the following: S1: Derive the equation of the drag line; S2: Perform recursive calculations on the equation of the drag line; S3: An obstacle collision detection area is set using an artificial potential field, and a piecewise iterative drag line obstacle avoidance method is proposed. S3 utilizes an artificial potential field to define the obstacle collision detection region and proposes a piecewise iterative parabolic obstacle avoidance method, the specific steps of which are as follows: S301: Virtually separate the super-redundant robot joints that do not meet the obstacle avoidance conditions; S302: Through discrete virtual joints, the super-redundant robot meets the obstacle avoidance conditions, and the end effector P1 can track the target point; S303: Discrete joints are recombined using the bidirectional iterative dragline method; In S303, the discrete joints are recombined using the bidirectional iterative dragline method. The specific steps are as follows: S3031: Redefine the position update direction and iteration step size; S3032: Calculate whether the hyper-redundant robot meets the obstacle avoidance condition, using the following formula: ||P tar -P obs ||≤ε0; Among them, P tar Let P be a point on the robot's joints and hinges. obs Let P be the center point of the obstacle, and ε0 be the obstacle detection radius; if P ta With P obs If the distance between them is less than ε0, then the segmented iterative drag line obstacle avoidance operation will be initiated; otherwise, the bidirectional iterative drag line motion planning will be initiated. The specific steps for calculating whether the super-redundant robot meets the obstacle avoidance conditions in S3032 are as follows: S30321: Assume a virtual discrete joint P 3a If obstacle avoidance is not considered, the position update direction is: S30322: When the virtual discrete joint P 3a When obstacle avoidance conditions need to be considered, the repulsive force F is defined according to the artificial potential field. rep and attraction F att Recalculate the redundant robot from P1 to P 3a The update direction D of the segment's iterative trajectory 3a According to the artificial potential field method, P 3a The repulsive force of the point's iterative motion is F. rep3a =(y r3a ,z r3a The attractiveness is F. att3a =(y a3a ,z a3a If the resultant force F is... 3a for: F 3a =F rep3a +F att3a =(y r3a +y a3a ,z r3a +z a3a ); 2. The robot obstacle avoidance method based on piecewise iterative torsion lines according to claim 1, characterized in that: The specific steps for deriving the equation of the drag line in S1 are as follows: S101: Given a link of length L, placed along the z-axis, one end B of the link is moved along the y-axis. The trajectory traversed by the other end A is the drag line. From the conditions for the formation of the drag line, we know that the velocity vector of the end A always points along the link L towards point B, that is, it is aligned with the tangent of the drag line. From the differential equation of the tangent at end A of the link L, we can derive the equation of the drag line as follows:
3. The robot obstacle avoidance method based on a piecewise iterative torsion line according to claim 2, characterized in that: The specific steps for recursively calculating the equation of the drag line in S2 are as follows: S201: (y,z) are the coordinates of point A, and L is the length of the link. The equation of the drag line is calculated using a closed-form method to solve the inverse kinematics problem of a multi-joint, super-redundant degree-of-freedom robot, thereby achieving more natural and smooth motion of the super-redundant robot. The resulting formula is as follows:
4. The robot obstacle avoidance method based on a piecewise iterative torsion line according to claim 3, characterized in that: The linear equation in the update direction is given by the point-slope equation:
5. A robot obstacle avoidance method based on a piecewise iterative torsion line according to claim 4, characterized in that: To achieve fast convergence, a virtual joint P is used. 3b (y 3b ,z 3b The iteration step size is calculated by finding the nearest point (vertical foot) to the point's slope equation. Where j is the iteration number, and the formula for calculating the iteration step size is: d 3aj =sqrt((y 3a -y 3aj ) 2 +(z 3a -z 3aj ) 2 )。 6. The robot obstacle avoidance method based on a piecewise iterative torsion line according to claim 5, characterized in that: Suppose that after the j-th iteration, P 3a and P 3b Overlap, and set the iteration termination condition to:
7. The robot obstacle avoidance method based on a piecewise iterative torsion line according to claim 6, characterized in that: The iteration ends when the distance Δl3 between the two virtual discrete joints is less than the set value δ3 and the obstacle avoidance condition is met. At this time, the inverse kinematic solution (α) is obtained. i ,β i Then, it begins planning the next obstacle avoidance maneuver, eventually reaching the target location.