Planning of robot trajectories with genetic algorithms

Abstract

Uses genetic algorithms (GA) for planning the stages of the trajectory of a robot arm called Jeca III. First, the GAs are used for planning the trajectory in the cartesian plane with obstacle avoidance, and some new operations, like crossover, are shown. Second, planning in the joint spaces are implemented using the classical GA with some modifications. This stage is divided into two parts: initial positioning and incremental positioning. The initial positioning has the purpose of locating the end effector of the robot arm in the first point of the trajectory, and the incremental positioning of moving the end effector to the next point of the trajectory. The result is a complete trajectory planning with the GAs, demonstrating the flexibility of this technique of artificial intelligence.