Decentralized Multi-Cobot Navigation Under Intermittent Communication

As collaborative robots (cobots) become more prevalent in industry, there is growing need for autonomous cobots that can cooperatively navigate shared workspaces. Reliable navigation and safety become especially critical when intermittent communication failures occur, potentially due to environmental factors or network disruptions. This paper contributes to the development of a navigation scheme for a team of autonomous networked cobots under intermittent communication. In particular, the paper proposes a decentralized control approach enabling cobots to cooperatively transport an object across an industrial environment despite intermittent communication. The navigation scheme is decentralized in the sense that each cobot computes its control actions locally using only information from neighboring cobots, without relying on a central coordinator, and applies actuator commands independently based on local sensor feedback and inter-robot communication. The work presented herein provides a comprehensive framework for autonomous multi-cobot cooperative object transportation tasks, including the design of the control, navigation, and communication systems. The communication network among the cobots is modeled using directed graphs, with the graph Laplacian matrix representing the connectivity among the cobots. The proposed method is first validated using a commercial robot simulator. Its performance is then evaluated on physical cobots operating in an indoor environment with various complexities.