A New Trajectory Planning Method of Redundant Manipulator Based on Adaptive Simulated Annealing Genetic Algorithm (ASAGA)

In this paper, the adaptive simulated annealing genetic algorithm (ASAGA) is presented by integrating the advantages of adaptive mechanics, simulated annealing algorithm and simple genetic algorithm (SGA). More successful result is got in manipulator trajectory planning using ASAGA compared with simple genetic algorithm. The experiment results show that the method can be used in manipulator trajectory planning effectively and will shed new light on how to bring the redundancy into full play to improve the movement of manipulator

1 Introduction

The motion of robot manipulator has been an active area of research. A critical problem is the position and posture control–––determine the position and posture of executor according to tasks. For given task, the joints' value of the manipulator should be given from one point to another point. The proposed solutions to this problem, often called path planning, can be broadly categorized into three classes of method: algebraic, geometrical and numerical. Algebraic or geometrical method does not well in getting the solutions in usual form. Numerical method is computational expensive. So there are drawbacks respectively in all the above methods. Reference [7] proposed a method on genetic, which can realize the open-loop control and program off-line, thus the proposed algorithm makes senses to the robot design and control.