A Lego Mindstorms NXT-Based Test Bench for Cohesive Distributed Multi-agent Exploratory Systems: Mobility and Coordination

Abstract

A cost-effective and time-saving test bench can significantly help with the study and implementation of distributed multi-agent exploratory systems. In this paper, we introduce a computer-based test bench that controls multiple low-cost Lego Mindstorms NXT programmable robotics kits via Bluetooth. We identify mobility and coordination as the two critical components to establish a platform for exploratory systems, and describe in detail how they are implemented using the low-cost Lego Mindstorms robots. In particular, the test bench includes a variety of ground-based or airborne mobility models, such as the Random Direction, Random Way-point, Flight Plan-based, and Smooth-turn mobility models. It also includes a distributed task assignment algorithm, based upon the stochastic automaton known as the influence model. At the end of the paper, we present a search-and-rescue example to illustrate the use of the test-bench. We expect that the test bench can serve as a conduit to foster the real-world application of distributed multi-agent exploratory system concepts.