Optimal motion planning method and system for inchworm robot based on improved PSO algorithm

Abstract

The invention discloses an inchworm robot optimal motion planning method and system based on an improved PSO algorithm, and the method comprises the steps: firstly building a kinematics model, and deducing position and speed kinematics forward and inverse solutions; gaits such as wriggling, rotating and overturning and obstacle crossing performance are analyzed. Secondly, based on an error tracking model of a Cartesian space cubic polynomial and a joint space septimal polynomial, an energy consumption optimal planning method with kinematics and dynamics as constraints is designed; mathematical descriptions of robot energy consumption and space curve similar operators are given; a mathematical model of a track and a path is established. And finally, an experiment of robot motion planning under continuous switching climbing gaits is tested. Through a series of modeling and motion planning of the inchworm robot, the inchworm robot can finally quickly obtain an effective path with optimal energy consumption under the condition that the starting point is known in the working process, and the working efficiency is effectively improved while the energy consumption is saved.

Description

Technical Field

[0001] The present invention belongs to the field of robot motion trajectory design, and in particular relates to an optimal motion planning method and system for an inchworm robot based on an improved PSO algorithm. Background Art

[0002] When working at heights, workers rely on equipment such as work vehicles, but hardware wear and human error can easily lead to accidents, and the work intensity is high and the efficiency is low. For this reason, climbing robots came into being. Through pre-set program mechanized operation, they replace humans to enter dangerous areas to work, reducing the error rate and improving safety. This technology has attracted much attention due to its high efficiency and stability, and is gradually becoming a key aid for high-altitude operations.

[0003] The application of climbing robots not only improves the safety of operations, but also promotes the innovation and development of aerial work technology. Climbing robots integrate the frontiers of mechanics, electronics, sensing, communication, automatic control and computer science, and are the result of multidisciplinary innovation. Faced with harsh operating environments, research and development challenges are numerous, but its broad application prospects are driving the rapid advancement of technology.

[0004] According to the operation scenario, climbing robots are divided into two categories: pole climbing and wall climbing. The inchworm robot involved in this application belongs to the pole climbing series. The core focuses on optimizing the robot's action strategy, striving to minimize energy consumption and improve energy efficiency. In essence, this involves the optimal motion planning of the robot's energy consumption from the starting point to the target point. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides an optimal motion planning method and system for an inchworm robot based on an improved PSO algorithm.

[0006] In a first aspect, the present invention provides an optimal motion planning method for an inchworm robot based on an improved PSO algorithm, comprising the following steps:

[0007] Step 1: Use the DH method to construct the kinematic model of the inchworm robot;

[0008] Step 2: Based on the kinematic model, analyzing the obstacle-crossing capability of the inchworm robot;

[0009] Step 3: Based on the kinematic model, a Lagrangian method based on DH parameter representation is used to perform dynamic modeling on the inchworm robot to obtain its dynamic expression;

[0010] Step 4: Perform motion planning for the inchworm robot, specifically:

[0011] Step 4-1: Optimize robot motion energy consumption;

[0012] Step 4-2: Construct an error tracking framework based on the programming of cubic polynomials in Cartesian space and septenary polynomials in joint space;

[0013] Step 4-3: Perform continuous trajectory planning for the robot based on the joint space acceleration curve to obtain the trajectory of its displacement, velocity, acceleration, and jerk;

[0014] Step 4-4: In Cartesian space, plan a path with continuous motion curves to generate a feasible trajectory for efficient operation of the robot;

[0015] Step 4-5: Construct a comprehensive model for optimizing the robot's energy consumption;

[0016] Step 5: Use the improved PSO algorithm to solve the path with the lowest energy consumption;

[0017] The improved PSO algorithm includes an improvement one for the inefficiency caused by the long search time, wherein the improvement one dynamically controls the inertia weight value by an adaptive adjustment strategy;

[0018] as well as,

[0019] A second improvement is made to address the problem of low precision caused by being easily trapped in a local optimal solution. The taboo area in the TS algorithm is introduced to exchange the algorithm space complexity for time complexity.

[0020] In a second aspect, the present invention provides an optimal motion planning system for an inchworm robot based on an improved PSO algorithm, comprising:

[0021] Kinematic model building module, using DH method to build the kinematic model of the inchworm robot;

[0022] An obstacle analysis module, which analyzes the obstacle crossing capability of the inchworm robot based on the kinematic model;

[0023] A dynamic model building module is used to construct a dynamic model of the inchworm robot by using a Lagrangian method based on DH parameter representation on the basis of the kinematic model to obtain its dynamic expression;

[0024] The motion planning module includes optimizing the robot's motion energy consumption; constructing an error tracking framework based on the planning of cubic polynomials in Cartesian space and septad polynomials in joint space; planning the robot's trajectory based on the continuous jerk curve in joint space to obtain its displacement, velocity, acceleration, and jerk trajectory; planning a continuous motion curve path in Cartesian space to generate a feasible trajectory for the robot's efficient operation; and constructing a comprehensive model for optimizing the robot's energy consumption.

[0025] The minimum energy consumption path solving module uses an improved PSO algorithm to solve the minimum energy consumption path; the improved PSO algorithm includes an improvement one for the inefficiency caused by too long search time, and the improvement one uses an adaptive adjustment strategy to dynamically control the inertia weight value;

[0026] as well as,

[0027] A second improvement is made to address the problem of low precision caused by being easily trapped in a local optimal solution. The taboo area in the TS algorithm is introduced to exchange the algorithm space complexity for time complexity.

[0028] In a third aspect, the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and wherein the processor implements the steps of optimal motion planning of a inchworm robot based on an improved PSO algorithm when executing the computer program.

[0029] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the computer program implements the steps of optimal motion planning of a inchworm robot based on an improved PSO algorithm.

[0030] Beneficial effects of the present invention: Through a series of modeling and motion planning of the inchworm robot, the present invention can ultimately enable the inchworm robot to quickly obtain an effective path with optimal energy consumption when a starting point is known during its operation. This not only enables the inchworm robot to save energy and extend its working time during high-altitude operations, but also effectively improves its work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is a simplified structural diagram of the inchworm robot according to an embodiment of the present application;

[0032] Figure 2 This is a parameter diagram of the inchworm robot according to an embodiment of the present application;

[0033] Figure 3 A description of the space curve of an embodiment of the present application in a Cartesian coordinate system;

[0034] Figure 4 The relationship between the energy consumption value and n of the embodiment of the present application;

[0035] Figure 5 This is a flowchart of the algorithm of the embodiment of the present application;

[0036] Figure 6 Schematic diagram of climbing different paths according to an embodiment of the present application. DETAILED DESCRIPTION

[0037] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0038] The present application discloses an optimal motion planning method for an inchworm robot based on a PSO algorithm, and particularly relates to the problem of motion energy consumption for the robot's creeping gait. Based on the "error tracking" model of a cubic polynomial in Cartesian space and a septenary polynomial in joint space, an optimal energy consumption planning method with kinematics and dynamics as constraints is designed to achieve automatic control of climbing.

[0039] In this application, the energy consumption of the inchworm robot in continuous creeping and linear gaits is studied. Considering the kinematic and dynamic constraints of the robot, with the minimum energy consumption as the optimization goal, an attempt is made to solve the optimal trajectory of the robot based on an improved adaptive particle swarm algorithm.

[0040] The optimal motion planning method for the inchworm robot based on the improved PSO algorithm provided in this application includes the following steps:

[0041] Step 1: Use the DH method to construct the kinematic model of the inchworm robot, integrate velocity kinematics and position kinematics analysis, and achieve a comprehensive and detailed kinematic description of the inchworm robot.

[0042] Step 2: After obtaining its kinematic model in step 1, the ability of the inchworm robot in surmounting obstacles is analyzed. By considering parameters such as the longitudinal radius of the gripper, the height from the gripper to the adjacent joint, and the length of the connecting rod between the joints, the intrinsic relationship between the height and width of the obstacles that can be surmounted is revealed.

[0043] Step 3: Based on the kinematic model established in step 1, the Lagrangian method based on DH parameter representation is used to model the dynamics of the inchworm robot and obtain its dynamic expression. The prerequisite for fully obtaining the torque of each joint is to accurately calculate the trajectory of its displacement, angular velocity and angular acceleration.

[0044] Step 4: After establishing the complete kinematic and dynamic model of the robot and analyzing its obstacle crossing performance through the above steps, motion planning is performed for the robot.

[0045] First: Optimize the robot motion energy consumption analysis and introduce a more intuitive mathematical expression;

[0046] Second: Construct an "error tracking" framework based on the programming of Cartesian space cubic polynomials and joint space septad polynomials. To this end, it is necessary to design a spatial curve similarity operator to accurately match the path characteristics;

[0047] Third: Perform continuous trajectory planning on the robot based on the joint space acceleration curve to obtain the trajectory of its displacement, velocity, acceleration, and jerk;

[0048] Fourth: Plan a continuous path of motion curves in Cartesian space to generate a feasible trajectory for efficient operation of the robot;

[0049] Fifth: Integrate the results of steps three and four to build a comprehensive model for optimizing the robot's energy consumption.

[0050] Step 5: After obtaining the motion energy consumption model through step 4, in order to pursue the ideal path with the most optimized energy consumption, it is essentially to explore the optimal solution in the complex high-dimensional nonlinear space. This application uses the PSO algorithm as a solution strategy to accurately locate the path with the lowest energy consumption.

[0051] For solving high-dimensional nonlinear problems, the two major drawbacks of the traditional PSO algorithm are the low efficiency caused by the long search time and the low accuracy caused by being trapped in the local optimal solution. For the former, an adaptive adjustment strategy is used to dynamically control the ω value to overcome the defect; for the latter, the taboo region in the TS (Tabu search) algorithm is introduced to solve this problem by exchanging the algorithm space complexity for time complexity.

[0052] In a preferred example, in step 1, the kinematic model of the inchworm robot is constructed using the DH method. The structural diagram of the inchworm robot is as follows: Figure 1 shown.

[0053] In robot operation, if the front gripper S2 remains stationary, the movement speed of the robot end can be expressed accordingly: Where J is the Jacobian matrix, is the joint angular velocity, are the linear velocity and angular velocity of the robot terminal relative to the base coordinate system, respectively.

[0054] The Jacobian matrix is ​​expressed as: Considering the inverse solution of velocity kinematics, for a general multi-degree-of-freedom robot, the inverse solution of velocity kinematics can be expressed as: When the front gripper S2 of the robot is fixed, the above formula is adopted and the inverse solution of the position kinematics is differentiated to obtain:

[0055]

[0056] Among them, v j1 、v j2 、v j3 、v j4 and v j5 are the rotation speeds of each joint, v j234 It is the sum of the rotation speeds of joints J2 to J4.

[0057] The above equation maps the swing foot velocity to the joint velocity. When the rear gripper S1 is fixed, the robot symmetry is used to transform the target posture to the rear gripper coordinate system, and then solve it according to the above equation. This completes the kinematic modeling of the robot.

[0058] In a preferred example, the step 2 is to analyze the obstacle-crossing capability of the inchworm robot after obtaining its kinematic model in the step 1.

[0059] Obstacles are indicated by gray rectangles. The parameters of the inchworm robot are as follows: Figure 2 As shown. Assume that the obstacle height is h, width is w, the longitudinal radius of the gripper is R, the height from the gripper to the J2 joint is a0, the length of the connecting rod from the J3 joint to the J4 joint is a1, the angle of the J3 joint is q1, and the radius of the joint sleeve is e. The relationship between the height and width of the obstacle that the robot can cross is as follows:

[0060]

[0061] When the robot's trunk joints are fully opened and stretched into a straight line, the robot can cross the widest obstacle. According to the actual robot, a0=265mm, a1=296mm, r=143mm, e=46mm, q1=0mm, substituting into the left formula we get w=306mm, and then substituting into the right formula we get the relationship between h and w.

[0062] In a preferred example, the step three continues the robot dynamics modeling field based on the kinematic model of step one, and in order to coordinate with the kinematic modeling system, a Lagrangian method based on DH parameters is selected to construct a dynamic model.

[0063] If the robot is composed of L joints connected in series, the dynamic expression can be obtained by derivation of the Lagrange equation:

[0064]

[0065] In the formula, the moment is composed of inertia term, centrifugal force, Coriolis force term and gravity term. i ,q i are the generalized torque and displacement of the i-th joint respectively.

[0066] In actual calculations, since there are L joints in total and adjacent joints interact with each other, it is necessary to derive the values ​​of each part one by one. The prerequisite for fully obtaining the torque of each joint is to accurately calculate the trajectory of its displacement, angular velocity and angular acceleration.

[0067] Dynamic modeling needs to distinguish between the upper and lower clamping situations of the robot. In view of the symmetry of the robot design in the embodiment of the present application, the modeling process of the clamping situation at the other end is similar, except that the derivation order of the joints and connecting rods is reversed.

[0068] In a preferred example, the fourth step is to perform motion planning for the robot after establishing the complete kinematic and dynamic models of the robot through the above steps and analyzing its obstacle crossing performance.

[0069] First, optimize the robot motion energy consumption analysis and introduce a more intuitive mathematical expression. The robot motion energy consumption can be roughly expressed as the following three parts: the energy consumption when the motor joint torque is consistent with the robot motion direction is positive work, the energy consumption when the direction is opposite is negative work, and the energy consumption of motor heat generation is heat loss.

[0070] Since the uncertainty of heat loss makes the solution of total energy consumption complicated, this embodiment directly quotes a more intuitive mathematical description that has been proposed to express energy consumption, as shown in the following formula:

[0071]

[0072] Where, t s , t e They represent the start and end time of the robot's motion respectively, and τ is the joint torque.

[0073] Second, under the “error tracking” framework of constructing the planning of cubic polynomials in Cartesian space and septad polynomials in joint space, it is necessary to evaluate the similarity between the planned path and the actual motion trajectory.

[0074] Furthermore, if the error between the two is maintained within a preset threshold, it is assumed that the energy consumption levels are similar, and the algorithm search strategy is guided accordingly. To this end, this application defines a quantitative index of spatial curve similarity. For two curves L1 and L2 composed of N discrete points, their similarity can be characterized by calculating the sum of the squares of the Euclidean distances between these points, thereby accurately measuring the spatial proximity between the curves:

[0075]

[0076] In the formula, is the spatial coordinate of the i-th point on the curve L1, is the spatial coordinate of the i-th point on the curve L2.

[0077] Third, the embodiment of the present application adopts a piecewise seventh-order polynomial planning strategy to optimize the robot joint space, ensure a smooth transition of the acceleration curve, and effectively suppress the motor resonance phenomenon.

[0078] Furthermore, we set the i-th joint at the starting point t j and end time The locations are and The speed parameter is and The acceleration is and The jerk is and Based on the above conditions, this application comprehensively defines the boundary constraints of the robot joint spatial motion to achieve precise control:

[0079]

[0080] Given that the above equation contains eight constraints, to ensure that there is a valid solution, the polynomial requires at least eight independent coefficients. Therefore, a seventh-order polynomial is selected to describe the joint displacement in the interval:

[0081] q i (t) = a0 + a1t + a2t 2 +a3t 3 +a4t 4 +a5t 5 +a6t 6 +a7t 7

[0082] In order to solve the eight coefficients, the boundary conditions are converted into matrix equations. By setting reasonable time matrices and boundary conditions, the polynomial parameters can be directly solved.

[0083] Fourth, the starting point and end point P of the robot's motion in the known space start (X start ,Y start ,Z start ), P end (X end ,Y end ,Z end ), if we choose a suitable sample point P insert (X insert ,Y insert ,Z insert ), then the three-point equation of plane θ is obtained:

[0084]

[0085] In order to ensure that the path connecting three points in the plane is smooth and continuous, it is a basic requirement to use at least a cubic polynomial. start , P insert , P end The exact positions p of the three points, and further clarified the slopes at these three points The cubic polynomial interpolation method can be used to accurately construct a smooth curve of the entire path, such as Figure 3 shown.

[0086] Fifth, after obtaining the joint trajectory and motion path, the torque of each joint can be obtained by the following formula:

[0087]

[0088] Where f is the implicit function of Lagrangian dynamics recursion, g is the function of converting q i (t) The other four items are t all ,q i (t) is the transformed function; thus the energy consumption of the i-th joint can be described as:

[0089]

[0090] Among them, combining the planning of Cartesian space path and joint space trajectory, q i (t) can be obtained by P insert , R max , α, Therefore, the solution model is finally transformed into:

[0091]

[0092] Where Q is the robot's workspace area, d is a fixed distance coefficient, and ε is a real number.

[0093] From the above model, it can be inferred that if a suitable t is selected during joint space trajectory planning all , then the whole solution process is to optimize the parameters of the path curve in Cartesian space.

[0094] Step 5. After obtaining the motion energy consumption model through step 4, in order to pursue the ideal path of optimizing energy consumption, it is essentially to explore the optimal solution in a complex high-dimensional nonlinear space.

[0095] This embodiment uses the PSO algorithm as a solution strategy in order to accurately locate the path with the lowest energy consumption. The PSO algorithm is a population X=(x1,...x i ,...x M ), the position of the i-th particle is X i =(x i1 ,...x iT ), the velocity of the i-th particle is V i =(v i1 ,...v iT ), the global extreme value of the population is B ω =(b ω1 ,...b ωT ), the individual extreme value of the particle is B i =(b i1 ,...b iT ). Particle x i Update the velocity and position as follows:

[0096]

[0097] In the formula, i = 1, ..., N, k = 1, ..., T, i is the population size, k is the number of evolutions, r1 and r2 are random numbers between 0 and 1, s1 and s2 are the individual and group learning factors, respectively, and ω is the inertia weight value.

[0098] When it comes to solving high-dimensional nonlinear problems, the two major drawbacks of the traditional PSO algorithm are the inefficiency caused by the long search time and the low accuracy caused by being trapped in the local optimal solution.

[0099] (1) The key to the problem of inefficiency caused by the long algorithm search time is to find a balance between the breadth of the search direction and the precision of the step size. The embodiment of the present application innovatively introduces the sigmoid activation function in the neural network to dynamically define ω, aiming to achieve adaptive adjustment of the inertia weight of each particle, retain the advantage of fast convergence, and enhance the flexibility of the search process, thereby effectively shortening the execution time of the algorithm. The sigmoid is shown in the following formula:

[0100]

[0101] This strategy encourages the global exploration capability to be enhanced by increasing the ω value when the f(k-1)-f(k) value is large, that is, when the current solution is far from reaching the local optimal area; on the contrary, when the f(k-1)-f(k) value tends to be smaller, it means that it has approached a local optimal domain. At this time, the rapid decay characteristics of the sigmoid function are used to reduce the ω value in time to make fine adjustments and enhance the local search accuracy.

[0102] (2) In order to solve the problem of low accuracy caused by the algorithm being easily trapped in the local optimal area, the step size needs to be increased to encourage particles to escape from the local optimal trap. This goes against the principle of fine search. For this reason, the taboo area in the TS (Tabu search) algorithm is introduced to solve this problem by exchanging the algorithm space complexity for time complexity.

[0103] The concept of taboo area is defined in this application as: the local optimal solution X i =(x i1 ,...x iT ) is the geometric center o of the space. The average value of the n-step search step length η in the search process is the radius r. The feasible solution space that satisfies the following formula is the taboo area:

[0104] (xx i1 ) 2 +...+(xx iT ) 2 ≤r 2

[0105] like Figure 4As shown in the figure, after simulation verification, when n is set to 10, the algorithm achieves a good balance between convergence speed and solution accuracy. In summary, the algorithm flow chart is Figure 5 shown.

[0106] Based on the same inventive concept as the optimal motion planning method for the inchworm robot in the aforementioned embodiment, the present application provides an optimal motion planning system for the inchworm robot based on an improved PSO algorithm, comprising:

[0107] Kinematic model building module, using DH method to build the kinematic model of the inchworm robot;

[0108] An obstacle analysis module, which analyzes the obstacle crossing capability of the inchworm robot based on the kinematic model;

[0109] A dynamic model building module is used to construct a dynamic model of the inchworm robot by using a Lagrangian method based on DH parameter representation on the basis of the kinematic model to obtain its dynamic expression;

[0110] The motion planning module includes optimizing the robot's motion energy consumption; constructing an error tracking framework based on the planning of cubic polynomials in Cartesian space and septad polynomials in joint space; planning the robot's trajectory based on the continuous jerk curve in joint space to obtain its displacement, velocity, acceleration, and jerk trajectory; planning a continuous motion curve path in Cartesian space to generate a feasible trajectory for the robot's efficient operation; and constructing a comprehensive model for optimizing the robot's energy consumption.

[0111] The minimum energy consumption path solving module uses an improved PSO algorithm to solve the minimum energy consumption path; the improved PSO algorithm includes an improvement one for the inefficiency caused by too long search time, and the improvement one uses an adaptive adjustment strategy to dynamically control the inertia weight value;

[0112] as well as,

[0113] A second improvement is made to address the problem of low precision caused by being easily trapped in a local optimal solution. The taboo area in the TS algorithm is introduced to exchange the algorithm space complexity for time complexity.

[0114] In another embodiment, a computer device is provided, which may be a server, and includes a processor, a memory, and a network interface connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal via a network connection. The computer program is executed by the processor to implement the optimal motion planning method for the inchworm robot.

[0115] Furthermore, to verify the effectiveness of the proposed method, three valid paths close to and far from the optimal path were selected for comparative testing, denoted as path 1, 2, and 3, respectively. Figure 6 That is, it is a schematic diagram of climbing different paths, and the energy consumption and maximum torque values ​​in the same time are shown in the following table, among which the path energy consumption obtained by the planning method of the present invention is the lowest.

[0116]

[0117] In summary, this application has the following technical effects:

[0118] 1. The DH method is used to carry out kinematic modeling of the inchworm robot. In order to echo the kinematic modeling, the Lagrangian method based on DH parameter representation is used for the robot dynamic modeling to more accurately derive the specific torque information of each joint of the robot.

[0119] 2. Aiming at the two major drawbacks of the traditional PSO algorithm: low efficiency caused by long search time and low accuracy caused by easy to fall into local optimal solution, the sigmoid function is introduced and the taboo area is defined to solve the above defects and improve the accuracy of motion planning.

[0120] 3. After defining the taboo area, the optimal number of search steps is obtained by evaluating the relationship curve between the number of search steps and the convergence speed and solution accuracy in the algorithm, thereby improving the feasibility of motion planning;

[0121] 4. By comparing the optimal path obtained with other feasible paths within the same motion time, it is proved that the energy consumption of the optimal path obtained is the lowest, thereby proving the feasibility of the motion planning method of the present application.

[0122] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. The optimal motion planning method of the inchworm robot based on the improved PSO algorithm is characterized by: The steps include: Step 1: Use the DH method to construct the kinematic model of the inchworm robot; Step 2: Based on the kinematic model, analyzing the obstacle-crossing capability of the inchworm robot; Step 3: Based on the kinematic model, a Lagrangian method based on DH parameter representation is used to perform dynamic modeling on the inchworm robot to obtain its dynamic expression; Step 4: Perform motion planning for the inchworm robot, specifically: Step 4-1: Optimize robot motion energy consumption; Step 4-2: Construct an error tracking framework based on the programming of cubic polynomials in Cartesian space and septenary polynomials in joint space; Step 4-3: Perform continuous trajectory planning for the robot based on the joint space acceleration curve to obtain the trajectory of its displacement, velocity, acceleration, and jerk; Step 4-4: In Cartesian space, plan a path with continuous motion curves to generate a feasible trajectory for efficient operation of the robot; Step 4-5: Construct a comprehensive model for optimizing the robot's energy consumption; Step 5: Use the improved PSO algorithm to solve the path with the lowest energy consumption; The improved PSO algorithm includes an improvement one for the inefficiency caused by the long search time, wherein the improvement one dynamically controls the inertia weight value by an adaptive adjustment strategy; as well as, A second improvement is made to address the problem of low precision caused by being easily trapped in a local optimal solution. The taboo area in the TS algorithm is introduced to exchange the algorithm space complexity for time complexity.

2. The optimal motion planning method for an inchworm robot based on an improved PSO algorithm according to claim 1, characterized in that: In step 1, velocity kinematics and position kinematics analysis are integrated to achieve a comprehensive and detailed kinematic description of the inchworm robot.

3. The optimal motion planning method for an inchworm robot based on an improved PSO algorithm according to claim 1 or 2, characterized in that: In step 2, the intrinsic relationship between the height and width of the obstacle to be crossed is determined by considering the longitudinal radius of the gripper jaw, the height from the gripper jaw to the adjacent joint, and the length of the connecting rod between the joints.

4. The optimal motion planning method for an inchworm robot based on an improved PSO algorithm according to claim 1, characterized in that: The motion energy consumption in step 4-1 includes the energy consumption when the motor joint torque is consistent with the robot's motion direction, the energy consumption when the direction is opposite, and the energy consumption of motor heating.

5. The optimal motion planning method for an inchworm robot based on an improved PSO algorithm according to claim 1 or 4, characterized in that: In step 4-2, for two curves composed of multiple discrete points, their similarity is characterized by calculating the sum of the squares of the Euclidean distances between these discrete points, thereby accurately measuring the spatial proximity between the curves and obtaining the error tracking framework.

6. The optimal motion planning method for an inchworm robot based on an improved PSO algorithm according to claim 5, characterized in that: In step 4-3, piecewise seventh-order polynomial planning is used to ensure smooth transition of the acceleration curve and suppress motor resonance.

7. The optimal motion planning method for an inchworm robot based on an improved PSO algorithm according to claim 6, characterized in that: The adaptive adjustment strategy specifically introduces the sigmoid activation function in the neural network to dynamically define the inertia weight value.

8. The optimal motion planning system of the inchworm robot based on the improved PSO algorithm is characterized by: include: Kinematic model building module, using DH method to build the kinematic model of the inchworm robot; An obstacle analysis module, which analyzes the obstacle crossing capability of the inchworm robot based on the kinematic model; A dynamic model building module is used to construct a dynamic model of the inchworm robot by using a Lagrangian method based on DH parameter representation on the basis of the kinematic model to obtain its dynamic expression; Motion planning module, including optimizing the energy consumption of robot motion; building an error tracking framework based on the programming of cubic polynomials in Cartesian space and septad polynomials in joint space; The robot is planned based on the continuous jerk curve in the joint space to obtain the trajectory of its displacement, velocity, acceleration, and jerk; in Cartesian space, a continuous path of motion curve is planned to generate a feasible trajectory for the efficient operation of the robot; a comprehensive model for optimizing the robot's energy consumption is constructed; The minimum energy consumption path solving module uses an improved PSO algorithm to solve the minimum energy consumption path; the improved PSO algorithm includes an improvement one for the inefficiency caused by too long search time, and the improvement one uses an adaptive adjustment strategy to dynamically control the inertia weight value; as well as, A second improvement is made to address the problem of low precision caused by being easily trapped in a local optimal solution. The taboo area in the TS algorithm is introduced to exchange the algorithm space complexity for time complexity.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of optimal motion planning of the inchworm robot based on the improved PSO algorithm described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of optimal motion planning of an inchworm robot based on an improved PSO algorithm as described in any one of claims 1 to 7 are implemented.

Images