Integrating Coordinated Path Following Algorithms to Mitigate the Loss of Communication Among Multiple UAVs.

摘要

The thesis addresses the problem of mid-air collision avoidance among multiple Autonomous Unmanned Aerial Vehicles (UAVs) capable of communicating their flight states across a time-varying communication network. The UAVs capabilities to (a) follow a given path and to (b) exchange and coordinate their relative position while on the path are considered the key factors enabling the time-critical coordination that in turn guarantees the safety of flight. The thesis is based on the key results of the recently developed concept of Coordinated Path Following (CPF) for multiple autonomous agents. While the path-following methodology is adapted without modification, the information exchange over the time-varying communication network and its impact on the performance of coordination was analyzed in a comparative study. The impact of the time-varying information flow is represented by the loss of link ratio, which is the ratio of time without information exchange to the nominal timeframe of communication in a given bidirectional network. The particular coordination metrics utilized are the coordination error (difference between the relative positions of UAVs on the paths) and the Euclidian distance between the UAVs (space separation). On the other hand, the control effort necessary to achieve the desired coordination is represented by the level and variation of the commanded velocity profile. The particular goal of the numerical study was to understand the amount of control effort required to achieve the desired separation of UAVs capable of exchanging a minimum number of parameters over a degrading communication network.