Optimal Trajectory Design for Conflict Resolution and Collision Avoidance of Flying Robots using Radau-Pseudo Spectral Approach

Abstract

This paper proposed a new methodology for optimal decentralized trajectory planning of multiple flying robots, in an obstacle-laden environment by incorporating obstacle avoidance and conflict resolution into one unified optimal control trajectory generation framework. In other words, an optimal trajectory that; firstly, passes through all obstacles, secondly maintains the safe flying zone for each of flying robot could be determined. In development of this path planner a Radau-pseudo spectral optimization method has been employed. The nonlinear point mass equations of motion of the flying vehicle in three dimensional space are utilized and realistic operational constraints influencing on flying vehicle state and control variables are considered via simplified mathematical models. The effectiveness of the proposed approach for solving centralized trajectory optimization problems, has been demonstrated via presenting a case study including two robots forced to fly in a hazardous area with fixed obstacles.