Constrained Model-based Predictive Control for Obstacle Avoidance

This article describes the use of predictive control for the control of an unmanned aerial vehicle in an unfamiliar three-dimensional environment. The vehicle must reach fixed targets at known positions while avoiding ellipsoid static obstacles that are detected on the way. For obstacle avoidance, constraints are implemented such that the predicted future UAV path does not intersect with the obstacles. The implementation of this approach is first tested with a hardware-in-the-loop system. Real flights with a 2.41 m wingspan airplane successfully confirm the applicability of the method. KeywordsModel predictive control; Obstacle avoidance