Collision avoidance based on line-of-sight angle

Abstract

This paper addresses the problem of collision avoidance for fixed-wing unmanned aerial vehicles. The angular velocity of the aircraft is adjusted in order to avert a possible collision with cooperative or uncooperative obstacles. The novelty of this work is that the control law uses only the line-of-sight angle as feedback, which can be obtained from an inertial measurement unit and a gimbaled camera mounted onboard the vehicle. This work aims at providing a solution to the collision avoidance problem for small lowcost unmanned aerial platforms, which are not equipped with sensors capable of measuring data such as position and velocity of the obstacle. This problem of practical relevance is faced from a theoretical standpoint. A Lyapunov based analysis is outlined, which provides safety guarantees under a given set of assumptions that the obstacle must satisfy. Simulation results are presented to validate the theoretical findings.

Using Natural Language to Enable Mission Managers to Control Multiple Heterogeneous UAVs

“…Once the vehicles takeoff, they must have the ability to replan their trajectories to avoid obstacles. Example algorithms that guarantee avoidance along with satisfaction of mission constraints and vehicle dynamic constraints are presented in [33][34][35] (Fig. 8).…”

“…Furthermore, these algorithms are able to avert a possible collision with cooperative or uncooperative obstacles without foreknowledge of the trajectory but with only an online, inaccurate prediction of the obstacle's trajectory. In fact, collision avoidance can be achieved with only the knowledge of the line-of-sight angle only [33]. Intervehicle Communication.…”