Abstract:Human being can master rapid and continuous reaction skill for high speed targets (e.g., baseball, ping-pong ball, etc.) after a long-term training process. Taking a deep sight into this phenomenon, it is due to the fact that human being select frequently optimal movement trajectories of the arm, and then store rich“knowledge”or“experience”in brain during the training process. Inspired by this mechanism of human being, a trajectory planning method of for a 7-DOF humanoid manipulator under rapid and continuous reaction and obstacle avoidance environment is presented. Through this method, the trajectory planning problem of the humanoid manipulator for high speed targets can be transformed into a tra jectory parameterization optimum problem. Considering the physical constraint and obstacles constraint conditions of the humanoid manipulator in operation, a“target operation level”of the humanoid manipulator for high speed targets is defined to constitute the fitness function and optimization goal, then particle swarm optimization (PSO) is used to search the optimal combination of the redundant parameters of the movement tra jectory. Based on these, a knowledge database of the corresponding operation environment is constructed through the trajectory parameterization optimum method, which can make the 7-DOF humanoid manipulator achieve rapid and continuous reaction operation for the high speed target. Finally, a humanoid robot for ping-pong playing is adopted as an example, and the method is employed to solve the trajectory planning problem of humanoid manipulator for ping-pong hitting. Both numerical simulation and actual humanoid robot testing results indicate that the proposed method can not only make the operation trajectory meet the physical constraint and obstacles constraint limitation of the 7-DOF humanoid manipulator, but also make the humanoid manipulator operate with rapid and continuous reaction, which demonstrates the effectiveness of this method for the tra jectory planning problem studied.

Trajectory Planning of Robotic Manipulator in Dynamic Environment Exploiting DRL

Tra jectory Planning of 7-DOF Humanoid Manipulator under Rapid and Continuous Reaction and Obstacle Avoidance Environment

Deep Reinforcement Learning Based Trajectory Planning Under Uncertain Constraints

An Efficiently Convergent Deep Reinforcement Learning-Based Trajectory Planning Method for Manipulators in Dynamic Environments

Trajectory Planning for Teleoperated Space Manipulators Using Deep Reinforcement Learning

Path Planning of 6-DOF Free-Floating Space Robotic Manipulators Using Reinforcement Learning

Robotic Arm Motion Planning with Autonomous Obstacle Avoidance Based on Deep Reinforcement Learning

Dynamic Trajectory Planning of a 7-DOF Surgical Robot Based on HER-DDPG Algorithm

Online Trajectory Generation with Local Replanning for 7-Dof Serial Manipulator in Unforeseen Dynamic Environments

Dynamic Obstacle Avoidance Algorithm for Robot Arm Based on Deep Reinforcement Learning

Real-time Motion Planning for Robotic Teleoperation Using Dynamic-goal Deep Reinforcement Learning

A Deep Reinforcement-Learning Approach for Inverse Kinematics Solution of a High Degree of Freedom Robotic Manipulator

Integrating DeepRL with Robust Low-Level Control in Robotic Manipulators for Non-Repetitive Reaching Tasks

Combining Deep Reinforcement Learning with a Jerk-Bounded Trajectory Generator for Kinematically Constrained Motion Planning

Design and Experimental Validation of Deep Reinforcement Learning-Based Fast Trajectory Planning and Control for Mobile Robot in Unknown Environment

Dual-Arm Robot Trajectory Planning Based on Deep Reinforcement Learning under Complex Environment

Multi-Objective Optimal Trajectory Planning for Robotic Arms Using Deep Reinforcement Learning

Path Planning Method for Manipulators Based on Improved Twin Delayed Deep Deterministic Policy Gradient and RRT*

Collaborative motion planning for multi-manipulator systems through Reinforcement Learning and Dynamic Movement Primitives

Reinforcement learning for path planning of free-floating space robotic manipulator with collision avoidance and observation noise