In this work, a distributed model predictive control (MPC) scheme based on consensus theory is proposed for the formation control of a group of quadcopters. The MPC scheme provides velocities for the quadcopters, which are considered as holonomic agents modeled at kinematic level. We propose soft and hard constraints for the MPC problem to address collision and obstacle avoidance as well as to maintain the connectivity of the communication topology during the motion of the agents to reach the desired formation. The contributions of this work are the following: First, we propose an integrated solution for the three tasks, including connectivity maintenance, which is uncommon in existing approaches, in addition to dynamic formation control and collision/obstacle avoidance. Second, we show that using both soft and hard constraints in the MPC problem gives better performance than using only one of the two. Third, unlike most MPC-based schemes from the literature, the effectiveness of our approach is validated through real experiments for a group of quadcopters.
