作者: Ruijie He , Abraham Bachrach , Michael Achtelik , Alborz Geramifard , Daniel Gurdan
关键词:
摘要: The MAV â08 competition focused on the problem of using air and ground vehicles to locate rescue hostages being held in a remote building. To execute this mission, number technical challenges were addressed, including designing micro vehicle (MAV), geo-locate targets, planning motion reach hostage location without detection. In paper, we describe complete system designed for competition, present our solutions three that addressed within system. First, summarize design MAV, focusing navigation sensing payload. Second, vision state estimation algorithms used track features, stationary obstacles moving adversaries, from sequence images collected by MAV. Third, algorithm generate plans approach building undetected adversaries; these adversaries are tracked air. We examine different variants search their performance under conditions. Finally, provide results systemâs during mission execution.