摘要：AbstractThis paper presents an algorithm to smoothly connect points in unknown environments with obstacles, using quartic splines. Rapidly Exploring Random Trees (RRTs) are used to find a path, with post-processing towards favorable geometry for a multirotor. The RRT-generated waypoint sequence is fed into the spline generation algorithm, which robustly provides a smooth dynamically feasible path. The procedure is set up towards respecting vehicle constraints, optimizing for reaching the goal point as fast as possible. In order to obtain a fast and reliable algorithm, we settle for suboptimal solutions. Simulation results show favorable performance even for challenging obstacle environments.
关键词：KeywordsTrajectoryPath PlanningGuidancenavigationcontrol of vehiclesAutonomous systems
