Multi-Point Trajectory Optimization for Minimum-Energy UAV Routing

Previous work on fuel-optimal trajectory planning for long endurance unmanned aircraft has focused on a point-to-point optimization task. The underlying optimization problem accounts for practical complexities, including the uncertainty in energy consumption arising from propulsion system dynamics and the influence of three-dimensional, time-varying weather conditions. Such factors often lead to optimized trajectories that deviate significantly from great-circle paths. When missions require the UAV to traverse a sequence of intermediate way points to satisfy operational objectives, a common approach is to solve a series of independent point-to-point trajectory optimization problems. However, this method may not yield the best solution with respect to energy consumption. This is because in a point-to-point routing each flight segment is optimized independently of the others. In this paper, we construct and solve a multi-point trajectory optimization problem for minimum-fuel UAV routing. We demonstrate that by optimizing over a set of multiple way points, better solutions can be obtained as compared to the sequential point-to-point approach. The key insight is that the multi-point problem formulation allows the performance over the entire mission time horizon to be considered rather than over the discrete time segments used to solve the point-to-point problems. This allows fuel consumption to be minimized by considering all previous and future actions leading to a more efficient solution.