Fuzzy manual control of multi-robot system with built-in swarm behavior

Abstract

Swarm robotics is a decentralized control architecture, where global behavior emerges as a result of local interactions between neighboring robots. The deficiency of the swarm behavior model is the stochastic nature of movement patterns, which reduces its applicability, when precise maneuvering is needed. This paper alleviates this problem by introducing fuzzy manual control of a multi-robot system utilizing the swarm behavior model. The built-in swarm behavior controls low level tasks such as formation keeping and obstacle avoidance. A fuzzy controller works as an intelligent mechanism for tuning the manual control signal received by the robots. The main advantages of the presented algorithm are: 1) deliberating the operator from low level maneuvering tasks; 2) single operator control of multi-robot group; 3) robustness, flexibility and scalability. The presented architecture was implemented and tested in a simulation environment. The introduced system can significantly improve the performance of search and rescue operations as well as exploration of dangerous environments.