Communications and Localisation for Cooperating Autonomous Mobile Robots

摘要: Heterogeneous teams of cooperating robots are ideal candidates for applications where the presence humans is impossible or should be avoided, not only ensuring safety people they replace, but also allowing execution tasks otherwise impossible. Moreover, if team heterogeneous, i.e. have different sensors/actuators, it possible to optimise use such components reducing associated costs while maintaining full functionality. However, one most attractive reasons using possibility maximising utility whole system; e.g. increasing effectiveness surveillance by performing cooperative sensing, improving rate coverage in search and rescue missions, motion coordination transport large parts. There several key factors that enable cooperation; this work we will focus on two them: a)exchanging information; b)tracking relative positions. For factor a), propose a new solution communicating data amongst heterogeneous robots, which due mobility requirements dynamic composition, with nodes joining leaving on-line, done through wireless medium. Thus, developed communication protocol support group membership, based fully decentralised proximity (ad-hoc) communications. improve efficiency channel bandwidth uses controlled transmissions, synchronized circular temporal framework Time Division Multiple Access (TDMA) type. One main innovations achieving distributed synchronization without global clock tolerating uncontrolled traffic external team. Resorting consensus technique show method, reception instants messages exchanged among members, assures vast range operational scenarios. supports multi-hop forwarding algorithm reduces end delivery delay and, importantly, drastically variability delay, isolating each node throughout network, significantly its behaviour analysability. b) above, methods derive positions from local RF In particular, both signal strength time flight paradigms them an innovative hybrid capable performance any alone. These ranging then used produce pairwise distance measurements positions, other. Finally, introduce another innovation allows us determining confidence determined method as function distances between robots.