Research On Motion Planning Of Six-Axis Industrial Robotic Arm Based On RRT*

In recent years,there have been continuous innovations and advancements in robot technology,resulting in a wide range of applications for mechanical arms in various fields.These advancements have led to the increasing use of mechanical arms in automated assembly,detection,processing,and other industries,making them an indispensable tool in many industries.Due to the variety and intricacy of application scenarios,conventional mechanical arm motion planning methods based on manual teaching are becoming more and more insufficient in meeting the requirements of different industries.As a result,a multitude of approaches have been suggested by scholars and experts from both national and international origins to accomplish independent motion planning of the manipulator without relying on manual teaching.Despite these advancements,due to the continued complexity of application scenarios,there is still a need for further improvement and optimization of existing motion planning methods.This paper centers on the movement planning technique of a six-DOF manipulator.It explores three aspects: path planning,trajectory optimization,and trajectory tracking.Firstly,the paper discusses the path planning and trajectory optimization of the robot arm based on kinematics.Then,it delves into trajectory tracking based on dynamics to achieve an overall motion planning method.The specific tasks are as follows:(1)We conduct research on the kinematic and dynamic models of a six-axis industrial robot.Firstly,we use the MDH parameter method to establish the forward kinematic model of the robot arm.Based on this model,we implement the inverse kinematic analysis of the robot arm using the analytical method as the basis for collision-free path planning of the robot arm.Afterwards,we utilize the Lagrangian approach to dynamically model the robot arm,serving as the foundation for achieving precise trajectory tracking of the robot arm.(2)To address the problem of strong randomness and low success rates in motion planning path searches for a 6R robotic arm in non-standard environments,we studied an improved version of the Rapidly-exploring Random Tree*(RRT*)algorithm.We used an adaptive target bias strategy and a sliding sampling pool strategy based on Sobol sequences to improve the algorithm’s goal orientation,a local node rejection strategy to improve the algorithm’s running speed,and a re-search strategy to improve the algorithm’s success rate,thereby improving the overall efficiency of the algorithm.We used Matlab as a tool to conduct path planning experiments in a virtual environment and compared the algorithm with various path planning algorithms using planning time,planning success rate,and path cost as indicators at different step sizes.The empirical findings indicated that the suggested algorithm attained the optimal efficiency concerning route expense and planning success ratio.At various step sizes,contrasted with the conventional RRT* algorithm,the proposed method resulted in an average increase of 32% in planning success rate,path cost decreased by an average of 2%,and planning time decreased by an average of 90.36%.Compared to various RRT algorithms,the planning time of this algorithm is longer,but it performs better in terms of success rate,path cost,and path curvature,which is important for subsequent trajectory optimization.(3)Due to the fact that the path obtained from path planning algorithms consists of line segments connected by discrete points on each joint,it lacks higher-order continuity and direct application may result in vibration,posing a safety hazard.Thus,additional optimization of the planned trajectory is required to enhance the fluidity of the manipulator’s movement and decrease its operational duration..To achieve this goal,B-spline curves are used for optimization,ensuring higher-order continuity of the robot arm’s motion trajectory.The kinematic constraints of each joint can be simplified by constraining the corresponding control points.Time nodes are encoded and the NSGAII algorithm is used to shorten the running time while ensuring the kinematic constraints.The population expansion strategy is employed to ensure synchronous starting and stopping of all joints in the manipulator,as well as to accelerate convergence.This trajectory optimization method is applied to the path nodes generated by different path planning algorithms in the previous section and simulated using Matlab.The experimental results show that the optimized robot arm trajectory has higher-order continuity while meeting the kinematic constraints,and the running time is reduced from 7 seconds planned by traditional genetic algorithms to 4.75 seconds.Moreover,the improved RRT* algorithm in this paper has certain advantages in the smoothness of the path.Ultimately,the validity of the algorithm is confirmed through real-world experiments in the presence of obstacles,utilizing the Zhongkeshengu 6-DOF robotic arm.(4)In order to enable the robotic arm to move more accurately along the planned trajectory and alleviate the difficulty of parameter tuning in traditional control methods,we improved the traditional PD-type iterative learning control and proposed a variable gain iterative learning control method to track the robotic arm’s motion trajectory.By using gain transformation rules related to tracking error,the gain coefficients are adaptively updated in real-time,resulting in better tracking accuracy and control effectiveness.This simplifies the process of parameter tuning,improves its fault tolerance to control gain parameters,and accelerates the error iteration convergence process.Finally,a simulation model was built in Simulink for experimentation.The experimental results show that this method has significant advantages over traditional PD-type iterative learning control in terms of parameter tuning,and the required number of iterative learning cycles is further reduced,achieving fast,stable,and precise following of the planned path.Figure [65] Table [4] Reference [114]...