The paper explores cooperative perception (CP) for unmanned aircraft systems (UAS) in the context of urban air mobility (UAM). It proposes a hybrid approach that combines local broadcast of sensor data with a central CP service to improve environmental awareness.
The authors first identify a CP data space by analyzing standards and protocols from the automotive, aviation, and drone domains. This data space includes metadata, kinematic information, mission details, detected objects, traffic guidance, and conflict elements. The authors then discuss the required information freshness and frequency for effective CP in UAM.
The proposed hybrid approach uses local broadcast for UAS to share their own data and detected objects. Ground stations collect and forward these messages to a central backend service, which aggregates the perception data and redistributes it to all UAS. This allows UAS to benefit from a wider view of the environment beyond their local sensors.
The authors evaluate this approach through simulations, comparing it to fully distributed CP and local perception without communication. The results show that the hybrid approach with a central backend significantly improves the environment awareness ratio (EAR) for UAS, reaching up to 66% compared to only 3.5% for local perception. The backend achieves a 99% EAR by aggregating data from all UAS. However, the increased communication load needs to be carefully managed to avoid channel congestion.
The authors conclude that the centralized CP service provides substantial benefits for environmental awareness in UAM, but further research is needed to optimize the communication strategies, caching, and placement of ground stations.
To Another Language
from source content
arxiv.org
Deeper Inquiries