Improved Area Covering in Dynamic Environments by Nonlinear Model Predictive Path Following Control

The problem of area coverage using unmanned aerial vehicles is becoming increasingly important, e.g. for search-and-rescue missions, exploration, and environmental monitoring. The area coverage control problem spans multiple levels, from path planning to tracking and flight control, which are normally time scale separated. Area coverage in highly dynamic environments requires tight integration of these levels to achieve maximum performance and to avoid collision. We propose to fuse the path following level and the mobile platform control level to achieve better performance. By doing so, dynamic constraints, static and dynamic obstacle avoidance and energy efficiency are taken into account. Specifically, we shift the assignment of the reference path velocity from the planner to the controller which leads to improved performance while satisfying all system constraints and therefore increases safety and reliability. Simulation results for a quadcopter illustrate the performance of the proposed algorithm.