Using 2D ORCA restrictions in MPC to avoid downwash disturbance in multi-UAV path-following

Abstract

This work presents an adaptation of the MPC-ORCA (Model Predictive Control - Optimal Reciprocal Collision Avoidance) navigation technique for multi-UAVs, implementing 2D ORCA constraints in the MPC problem to address the problem of disturbance caused by downwash in small UAVs. As a proof of concept, we showed a downwash disturbance scenario with two micro quadrotors in the real world, motivating our choice of 2D ORCA restrictions. Then, to evaluate the applicability of the proposed approach, we implemented the algorithm in two simulation groups, a simpler numerical simulation using only the equations of motion and a more complete simulation, covering the flow of air particles. The results of the simulations show that the proposed method effectively prevented the overlap of the vehicles, while attempting to maintain the agents' performance of the path-following task within a shared curve.