Trajectory planning in workspaces with obstacles taking into account the dynamic robot behaviour (Q2489516)

The path planning algorithms for robotic systems try to obtain a sequence of robot configurations that fulfils some conditions, mainly collision avoidance. The algorithms devoted to determine the optimal trajectory for robotic systems try to obtain a temporal history of the evolution of robot joint coordinates in order to minimize the required time or energy consumption. The present paper considers the second area of problems, namely the obstacle avoidance minimizing the robot energy consumption. The proposed trajectory planner is arranged in two stages: 1. Obtaining a discrete configuration space; 2. Obtaining an optimal and feasible trajectory. At the first stage, unlike other planners, the problem is set out in Cartesian space, and the motion between adjacent configurations is characterized by the travel time. At the second stage, a trajectory in the joint space is generated that minimizes the total required time and is dynamically compatible with robot features.