Vision-Based Formation Flight Control of Aerial Robots in Communication Failure Condition

In this paper, the vision based formation control of a leader-follower aerial robotic system is discussed while communication failure between the robots is considered. Based on vision sensors and image processing computations, without any need to further communications, the follower robot obtains the line of sight angle and the corresponding distance from the leader. These information are used to calculate the leader speed and its path angle. Using model predictive control method, it is shown that even in the presence of communication failure between the robots, the flight formation can be controlled while the follower moves along the desired relative angle and keeps the corresponding distance. To this end, the control inputs are obtained in terms of normal and tangential acceleration of the follower. Simulation results show a good performance of the proposed model predictive controller in both ordinary, and communication failure conditions.