摘要：AbstractThe goal of this work is to explore ways of generating state trajectories for dynamical systems subject to computational constraints, obstacles and priority assignment. The algorithms are developed for a miniature unmanned aerial vehicle (UAV) in a modular fashion and include (1) a genetic algorithm (GA) for solving the travelling salesman problem (TSP) with respect to priorities and obstacle avoidance, (2) a projective algorithm (PA) for finding the shortest paths around obstacles, (3) a quadratic program (QP) for minimum-snap polynomial trajectory generation subject to equality constraints to guarantee avoidance of static obstacles. Combined, the algorithms enable simple and computationally efficient motion planning with support in both R2and R3exemplified in a real-time implementation.
