Cooperative Autonomous Collision Avoidance System for Unmanned Aerial Vehicle

Autonomous collision avoidance system (ACAS) was defined and investigated in this paper to support UAVs integration to the national airspace system. This includes not only UAVs on-board system, but also the definition of requirements, collision avoidance structure, and the avoidance rules. This paper focuses on the cooperative avoidance, where UAVs (or any aircraft) involved avoid each other using rules previously agreed by involved parties. A novel algorithm of avoidance was developed, named as Selective Velocity Obstacle (SVO) method. Several simulations were conducted and show satisfying result on how well the algorithm work to avoid separation violations. In the end of the paper, using Monte Carlo simulation, violation probabilities were derived for three setups. These simulations shows the performance of the developed algorithm for cooperative ACAS, and suggesting the need to derive a new parameter, i.e., the minimum required turning rate of avoidance.