Effects of safety protocols on unmanned vehicle ground operations
Recent advances in unmanned and autonomous vehicle technology are accelerating the push to integrate these vehicles into human-centered environments such as commercial aviation and public roads. Much of the current research into autonomous systems examines improving the performance of individual unmanned vehicles or improving the safety of their interactions with individual humans; very little examines the behavior of the broader system. For large-scale transportation systems, real-world field trials involving unmanned vehicles are difficult to execute due to concerns of cost, feasibility of construction, and the maturity of the technologies. This paper describes the use of an agent-based model of unmanned vehicle behavior in human-centered environments to explore the effects of their implementation in these domains. In particular, this work explores how safety protocols governing the integration of manned and unmanned vehicles affect performance in an aircraft carrier ground control environment. Three different types of futuristic unmanned vehicle control architectures are considered in conjunction with four different types of safety protocols: dynamic, area, temporal, and combined area-plus-temporal separation. Results demonstrate that measures of safety vary widely across these systems, demonstrating distinct tradeoffs of safety and mission performance, as well as across different safety measures.