LSwarm: Efficient Collision Avoidance for Large Swarms with Coverage Constraints in Complex Urban Scenes

In this paper, we address the problem of collision avoidance for a swarm of\nUAVs used for continuous surveillance of an urban environment. Our method,\nLSwarm, efficiently avoids collisions with static obstacles, dynamic obstacles\nand other agents in 3-D urban environments while considering coverage\nconstraints. LSwarm computes collision avoiding velocities that (i) maximize\nthe conformity of an agent to an optimal path given by a global coverage\nstrategy and (ii) ensure sufficient resolution of the coverage data collected\nby each agent. Our algorithm is formulated based on ORCA (Optimal Reciprocal\nCollision Avoidance) and is scalable with respect to the size of the swarm. We\nevaluate the coverage performance of LSwarm in realistic simulations of a swarm\nof quadrotors in complex urban models. In practice, our approach can compute\ncollision avoiding velocities for a swarm composed of tens to hundreds of\nagents in a few milliseconds on dense urban scenes consisting of tens of\nbuildings.\n