Development of distributed control architecture for multi-robot systems

Abstract

The execution of complex tasks by teams of robots has been widely investigated in the last decades, since many operations are too risky or difficult to be performed by humans or by a single robot. The complexity and variety of applications of mobile robotics make the coordination of teams a big problem, as several topologies of control systems, from simple single processes to large networks with distributed elements that are capable of switching function, may be necessary. Although simple solutions exist, more efficient approaches use distributed communication architectures and components abstraction layers. Available proposals provide many components and interfaces, complicating their understanding and operation. This paper presents a generic control architecture that provides the developer with a small amount of elements implemented safely and on high-performance libraries. The simplicity and modularity of the proposal allow implementation of features such as control of heterogeneous robots, data source and command destination transparency and platform and language independence. The ability to support with reliability, transparency and ease the development of various scenarios of autonomous mobile robotics make the proposed architecture a powerful and valuable tool in the design and operation of these systems.