Encounter Aware Flight Planning in the Unmanned Airspace

Ensuring safe and efficient operations for unmanned aircraft is vital to their integration into the civil airspace. This paper proposes a set of encounter-based metrics that can be used to estimate the complexity and thus the capacity of an unmanned airspace. We show that these metrics are able to provide a more detailed view of safety failure characteristics in the airspace than traditional metrics. The encounter-based metrics are used to formulate a novel path planning algorithm that combines 4D trajectory optimization with a flight scheduling heuristic in order to reduce the number of difficult to solve encounters in the airspace. The proposed algorithm is distributed in nature and can be applied to an airspace managed by multiple service suppliers that are capable of generating flight plans. Through simulation, we show that this approach is able to outperform other distributed flight planning methods and can perform nearly as well as a centralized approach that jointly optimizes flight plans simultaneously. We also demonstrate that this approach can scale to high operational volumes and is able to efficiently handle spikes in airspace usage.