Paths for Peer Behavior Monitoring Among Unmanned Autonomous Systems
Rick Dove · 2009
Aberrant behavior in unmanned weapon systems can be caused by design flaws, system malfunction, malevolent control penetration, and human error. As systems manifest more autonomous operation, unpredictable emergent behaviors add to this list. The near-term expected future calls for swarms of heterogeneous unmanned autonomous systems (UAS) to be employed on a mission. The emergent group behaviors will add new dimensions to testing, posing potentially explosive centralized monitoring and evaluation tasks with large groups. The impossibility of specifying and testing all potential situational conditions is recognized, and the safety of the testing environment itself is of concern. Lessons from social animal life show that peer behavior monitoring has evolved to detect and mitigate aberrant behavior among members, and mitigating that behavior if it is evaluated as intolerable. This article explores a foundation for peer behavior monitoring among UAS under both test and mission conditions.