Actions, Observations, and Decision-Making: Biologically Inspired Strategies for Autonomous Aerial Vehicles

摘要

This paper details the development and demonstration of an autonomous aerial vehicle embodying search and find mission planning and execution strategies inspired by foraging behaviors found in biology. It begins by describing key characteristics required by an aerial explorer to support science and planetary exploration goals, and illustrates these through a hypothetical mission profile. It next outlines a conceptual bioinspired search and find autonomy architecture that implements observations, decisions, and actions through an 'ecology' of producer, consumer, and decomposer agents. Moving from concepts to development activities, it then presents the results of mission representative UAV aerial surveys at a Mars analog site. It next describes hardware and software enhancements made to a commercial small fixed- wing UAV system, which include a new development architecture that also provides hardware in the loop simulation capability. After presenting the results of simulated and actual flights of bioinspired flight algorithms, it concludes with a discussion of future development to include an expansion of system capabilities and field science support.