Article Preview
TopIntroduction
Distributed control of multiple mobile robots has received a great deal of attention in recent years. This growing area of research finds its inspiration from different systems that exist in nature. There are many examples of such systems as a flock of birds or a swarm of insects. A salient feature of such systems is that the individuals in the system can share information with their neighbors locally and through which the global behaviors of the overall system may be achieved. Numerous applications exist in the use of multiple mobile robots. For instance, this can be found in a variety of military missions such as surveillance and reconnaissance, or search and rescue, and in civilian applications such as environmental sensing and monitoring, and cooperative transportation (Qu, 2009; Red & Beard, 2008).
The design of distributed control for multiple robots is challenging because interactions among robots are often local, directional and intermittent due to limited sensing/communication capabilities of individual robots. Thorough study has been done addressing this challenge by assuming simple linear models for robots (Ren & Beard, 2008; Qu, 2009; Bullo et al., 2009; Saber et al., 2007). For instance, formation control of multi-robots was studied in (Desai et al., 1998; Leonard & Fiorelli, 2001) under a fixed sensing and communication structure among robots. For time varying sensing and communication, the neighboring control rule was proposed in (Vicsek et al., 1995) and rigorously proved in (Jadbabaie et al., 2003). It was shown that all systems in the group converge to the same value if the underlying undirected sensing communication topologies among systems are connected. More complicated sensing and communication topologies were studied in (Ren & Beard, 2005; Lin et al., 2004; Saber et al., 2007; Qu et al., 2008; Wang et al., 2006). By explicitly considering robot dynamics, a discontinuous control was proposed in (Dimarogonal & Kyriakopoulos, 2007) and stability was analyzed using nonsmooth Lyapunov theory. Time-varying controls were designed and analyzed using average theory in (Lin et al., 2005). A number of experimental results have been reported recently, which deal with multi-robot coordination (Marshall et al., 2006), leader-follower flocking (Gu & Wang, 2009), formation control (Antonelli et al., 2009; Reyes & Tanner, 2015), and containment control for multiple vehicles (Cao et al., 2011).