A key challenge facing modern airborne delivery systems, such as parafoils, is the ability to accurately and consistently deliver supplies into difficult, complex terrain. The presence of terrain obstacles combined with a high degree of environmental wind uncertainty and under-actuated parafoil dynamics requires an efficient form of both prediction and risk assessment. Our work has focused on the development of a new sampling-based motion planner known as chance-constrained RRT (CC-RRT) which is capable of generating robust, and dynamically feasible trajectories in real-time for linear Gaussian systems subject to process noise, localization error, and uncertain environmental constraints.
Unlike traditional RRT motion planners, the CC-RRT algorithm we have developed guarantees a lower bound on planning feasibility by producing a tree of state distributions. Using a novel wind model and classification scheme, the effects of future wind variation are translated into uncertainty distributions around the expected mean state and can be efficiently checked for constraint violation at each time step of the trajectory. Simulation results have demonstrated that the CC-RRT algorithm can enable a large, autonomous parafoil to robustly execute collision avoidance and precision landing on mapped terrain, while reducing the worst-case impact of wind disturbances relative to state-of-the-art approaches.