Simultaneous Exploration and Exploitation of a Wind Field for a Small Gliding UAV
Nicholas Lawrance, Salah Sukkarieh · 2010
ying vehicle exploiting favourable wind conditions to extend ight duration without the use of internally carried energy. The current research presents an approach for simultaneously mapping and utilising a wind eld for energy-gain ight. However, an adequate estimate of the wind eld is required in order to generate energy gain paths. Conversely, the exploration required to generate a useful map estimate requires energy. This work aims to address these problems simultaneously by attempting to maintain a wind map based on observations collected during the ight and use the currently available map to generate energy gain paths. The proposed method utilises estimates of the energy cost and potential reward for taking dierent actions. Wind estimation is performed using Gaussian process (GP) regression. A path planner generates and selects paths based on energy eciency and eld exploration. The scheme is tested in simulation with wind elds consisting of single and multiple stationary thermal bubbles.