Path planning for autonomous parafoils using particle chance constrained rapidly-exploring random trees in a computationally constrained environment

Abstract

Particle chance constrained rapidly-exploring random trees (PCC-RRT) is a sampling-based path-planning algorithm which uses particles to approximate an uncertainty distribution. In this thesis, we study the use of PCC-RRT on an autonomous parafoil. Specifically, we explore the behavior of PCC-RRT in a computationally constrained environment by studying the tradeoff between the number of samples and number of particles per sample and its effect on miss distance in single-threaded coded with a time constraint. We compare the results generated with the PCC-RRT planner to the equivalent data from a nominal planner using rapidly-exploring random trees (RRT) to determine the effect of robustness.