Experiments in multirobot coordination

Abstract

Consequent to previously published theoretical work by Marshall, Broucke, and Francis, this paper summarizes the apparatus and results of multirobot coordination experiments conducted at the University of Toronto Institute for Aerospace Studies. These experiments successfully demonstrated the practicality of cyclic pursuit as a distributed control strategy for multiple wheeled-robot systems. Moreover, the pursuit-based coordination algorithm was found to be surprisingly robust in the presence of unmodelled dynamics and delays due to sensing and information processing. The findings of this research not only bode well for continuing research on pursuit-based coordination strategies, but also for other cooperative multirobot control techniques employing similar local interactions.

A Matlab-based Remote Lab for Multi-Robot Experiments

“…Unfortunately, performing this kind of experiments is often a difficult and expensive task, particularly when dealing with algorithms designed for teams of robots. This is confirmed also by the relatively few experimental results on multi-agent algorithms that can be found in the literature (Marshall et al [2006], Ren and Sorensen [2008], Mastellone et al [2008]). Therefore, allowing students to test multi-robot algorithms on real vehicles, while being highly instructive, may become a prohibitive task, especially in case of large classes.…”

Bearing-Only Control Laws For Balanced Circular Formations of Ground Robots

“…Figure 9 shows that in our experiment the robots are stabilized to the star polygon {5/3}. Remark 5.1: When the communication graph is a fixed, directed graph with a ring topology, where agent i could only see agent (i + 1)/mod(n), then the n-agent system would behave like a team of robots in cyclic pursuit [9].…”

“…Let us characterize the set of equilibria given by (9). We represent the unit vector z i in the rotating frame by z i =…”

Teleoperation of a platoon of distributed wheeled mobile robots with predictive display

“…Numerous techniques have been proposed for the formation control of multiple distributed nonholonomic WMRs, some experimentally demonstrated with onboard sensing and estimation (e.g., Desai et al 2001;Das et al 2002;Marshall et al 2006;Mariottini et al 2009). On the other hand, many results have been reported for the teleoperation of multiple mobile robots (e.g., centralized Lee and Spong 2005;Lee 2008;Rodriguez-Seda et al 2010;Fong et al 2003, distributed Franchi et al 2012Ochiai et al 2014) and also for the teleoperation with predictive display (e.g., Kim and Bejczy 1993;Chong et al 2002;Mitra and Niemeyer 2008;Kelly et al 2011).…”