Research On Obstacle-avoidance Path Planning For Dual-arm Robot Based On Time Sequence A~* Algorithm

In the background of industry 4.0 and in Chinese manufacturing 2025,robotics are widely used in industrial production,medical treatment,military activities,and home services.With the continuous improvement of people's requirements for industrial,medical and service levels,single-arm robots can no longer meet people's needs.Therefore,the research on the technology related to the dual-arm robot is of great significance.Compared with single-arm robots,double-arm robots have higher adaptability to complex tasks and higher utilization of working space.Therefore,the research on coordinated motion of double-arm robots is a main direction of robot development today.Among them,the motion path planning of the robot arm is the key technology of the robot arm motion control,and its planning effect directly affects the use cost and efficiency of the robot arm.Manipulators are generally used in complex obstacle environments,and effective obstacle avoidance planning is required during the movement.Therefore,this paper takes the dual-arm robot as the research object,and studies the kinematics and obstacle avoidance path planning of the robotic arm.This paper mainly discusses from the following aspects:Firstly,the kinematics model of the six DOF Manipulator is established by using D-H theory.Equations were derived based on the forward and inverse kinematics of the manipulator.Aiming at the collision detection problem in the planning process of the robotic arm,this thesis analyzes the cylindrical envelope method and the bounding sphere envelope method,and proposes a capsule bounding box,which simplifies the robotic arm model.The problem of collision detection of a robotic arm is transformed into a problem of calculating space line-to-line distance,and,distance between space point and line.Secondly,the principle and implementation process of A * algorithm are introduced,and the numerical characteristics of A * algorithm nodes are analyzed.It is found that some nodes in the close table are not helpful for finding the optimal path.So,a new evaluation method is proposed to avoid analysis of these nodes.By adjusting the data storage method of the open table,the time consumption of data retrieval and data update in the open table is further reduced,and the problem that the A~* algorithm takes too long to calculate in high dimensions is alleviated.After that,a time sequence A~*(TSA~*)algorithm was proposed in the path planning of the dual-arm robotic.The motion planning for the main arm is conducted in a static environment.And then,the motion planning for the slave arm is conducted,while the position of the main arm is treated as a known dynamic obstacle.In the planning process,time parameters are assigned to each path node,and an adaptive motion waiting strategy and smoothing process are introduced.Finally,in order to verify the above-mentioned theoretical work,a virtual prototype is established for simulation experiments,and the verification is performed on a dual robotic arm experimental platform.The analyzed robot arm path was converted into device programming code,and the assembly operation of the equipment is guided.Experimental results show that the TSA * algorithm proposed in this thesis can quickly and effectively implement obstacle avoidance path planning for dual-arm robots,and achieve the expected research goals of the thesis.