ESA GNC Conference Papers Repository

Advanced path planning algorithms for UAVs are often computationally expensive, especially planning algorithms that are based on optimization. For small UAVs, the implementation is a challenging task due to limited onboard resources with constraints in weight, computational power and size of the hardware. To this end, path planning algorithms and characteristics of onboard computers have to be considered in an integrated manner in the development phase. The approach presented in this paper is to exploit parallel computation capabilities of FPGAs together with tailored path planning algorithms. Therefore a small FPGA-based onboard computer system for atmospheric UAV application is presented and an appropriate path planning algorithm based on realtime optimization is described. For atmospheric flight vehicles, simulation results and realtime testbed results from nonlinear 6-DoF-simulations have proven the functionality of the approach. In this paper the idea is extended to planetary landing-like scenarios with vertical-take-off and landing vehicles (VTOL). Simulations with a nonlinear simulation model have proven the functionality for this application as well.