Trajectory Deformation Based on Energy Optimization and Obstacle Avoidance

Abstract

In this paper, on the basis of energy optimization and obstacle avoidance, the trajectory generation of the robotic arms of two joints is carried out by human-robot interaction considering the physical limit of joints. Different from the related work, we not only consider the energy consumption of the manipulator in the process of trajectory generation, but also consider the feasibility of the trajectory, and increase the inequality constraint on the original equation constraint to avoid the robotic arm exceeding the joint limit. After transforming the original optimization problem into a quadratic programming problem with equality and inequality constraints, we use neural network to solve the optimal trajectory. Finally, the simulation results show that the increased joint constraint can ensure the actual trajectory of the generated trajectory and improve the safety of the actual operation.