Research On Obstacle Avoidance Motion Planning Of 6-DOF Manipulator

The robot intelligence mainly refers to the robot can complete the task autonomously,that is,to achieve the robot autonomous obstacle avoidance movement planning.Under the background of intelligent manufacturing and "Made in China 2025",efficient and flexible autonomous obstacle avoidance motion planning ability has become an urgent and practical requirement for robots in current production and life.In this paper,the 6-dof manipulator is taken as the research object,and the forward and inverse kinematics,forward and inverse dynamics,obstacle avoidance path planning and multi-objective optimal trajectory planning of the manipulator are studied around the autonomous obstacle avoidance motion planning of the manipulator.The main research contents are as follows:(1)Kinematics and dynamics modeling of manipulator: Based on the DH method,the forward and inverse kinematics model of the manipulator is established.A closed inverse kinematics solution algorithm is obtained through the transformation matrix and the analysis of singular points.The dynamic model of the manipulator is established based on the Newton-Eulerian method.The dynamics simulation of the manipulator is carried out based on Simscape,and the correctness of the dynamic solution function is verified.The working space of the manipulator is obtained by the Monte Carlo method.The effectiveness of the inverse kinematics solution algorithm is verified by the simulation of double-arm handling.(2)Obstacle avoidance path planning for manipulator: Considering that RRT algorithm and RRTConnect algorithm need to repeat excessive search in complex environment,which leads to the defect that nodes are too close to each other and edge crossing occurs,a fast path planning algorithm for mechanical arm based on sparse deadpoint preservation strategy is proposed.Several algorithms are compared and verified in2 D and 3D environments.The results show that the SDPS-RRTConnect algorithm greatly reduces the number of collision detection and improves the planning speed,which proves the excellent effect of the improved algorithm in complex environment.(3)Multi-objective optimal trajectory planning of the manipulator: First,the basic trajectory planning is introduced,and the linear interpolation,circular interpolation and attitude interpolation are planned.On this basis,the time optimal trajectory planning is carried out,and the required objective function and constraints are given.Although the time-optimal trajectory planning can reduce the time for the manipulator to complete the expected trajectory,it also leads to an increase in energy consumption and impact.Therefore,the multi-objective optimization of the trajectory based on the NSGA-II algorithm is carried out,and the optimization objective functions of time,energy and impact are given.The desired solution can be obtained from the Pareto front-end by normalizing the objective weight function.(4)Simulation and experiment: Taking the self-developed robot in the laboratory as the research object,a robot simulation environment is built in MATLAB.Obstacles and the posture of the manipulator are set,and the path planning of the manipulator is carried out based on the SDPS-RRTConnect algorithm,and the obstacle-free path is successfully obtained.Experiments were carried out with a real robot,and the robotic arm successfully reached the target point without collision.For the optimal trajectory planning,the genetic algorithm is used to achieve the optimal effect of time,and the NSGA-II algorithm is used to achieve the optimal effect of time,energy,and impact multi-objective,and the displacement,velocity,acceleration,torque,etc.in the movement process are all within the constraints.