Multi-drone cooperation to improve navigation integrity in low altitude urban environments

This paper uses multi-UAV cooperation to enhance the positioning performance of a swarm when GNSS measurements are affected by faults. Only pseudorange observables are accounted for and a centralized extended Kalman filter, which can deal with both cooperative and non cooperative measurements, has been developed. The filter is complemented with a fault detection and elimination strategy based on Mahalanobis distance, which removes pseudoranges that are deemed faulty. Contribution of relative sensing sources (camera or ranging devices) and formation geometry to cooperative GNSS integrity mitigation is investigated, and the concept of cooperative slope is introduced as a metric to define the best cooperative formation geometry. Results demonstrate, by simulating spoofing and multipath phenomena, that cooperation enhances the navigation performance of the UAVs in a multi vehicle formation only if the formation geometry is correctly selected with the aid of the cooperative slope metric.