In disaster monitoring, Unmanned Aerial Vehicles (UAVs) are becoming more suitable than earth observing satellites and manned helicopters, especially in emergent situations. However, decisions must be made as to which attributes the UAV must possess when developing new models. Combined with NOLH experiment design and multi-agent simulation (MAS) technology, a new architecture evaluation method, termed Rough Set Based Fuzzy Neural Network (RSBFNN), is proposed to simulate and analyze both the UAV attributes and desired effectiveness. Experimental results show that this method is very dependable when comparing predicted and actual responses and yields better performance than other models. This method best fits the design of the disaster monitoring UAV, and can be used to perform a series of dynamic trade studies, in which various architecture alternatives are examined and compared.
Digital Object Identifier (DOI)
Li, Zhifei; Zhu, Yifan; Yang, Feng; and Wang, Wenguang
"Study on the Evaluation of UAV Disaster Monitoring System Architecture based on the RSBFNN Algorithmic Method,"
Applied Mathematics & Information Sciences: Vol. 09
, Article 40.
Available at: https://dc.naturalspublishing.com/amis/vol09/iss3/40