Toward obstacle-specific morphology for a reconfigurable tiling robot

This paper proposes a novel method based on Fuzzy Logic Systems (FLSs) optimized to maximize the area coverage of a tiling robot by reconfiguring per an obstacle considering the “Infi” concept. The consideration of an infinite number of reconfigurable shapes by a tiling robot is defined as the “Infi” concept. The major advantage of the proposed method over state-of-the-art “Infi” methods is the ability to tune based on cost measurements without demanding an explicit set of training data. The optimization techniques, Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Simulated Annealing (SA), were used for tuning the FLSs. The FLSs tuned using these three techniques were compared with state of the art by considering hTetro through simulations. According to the statistical outcomes, the proposed method can significantly improve the area coverage compared to the conventional methods while providing performance and behavior similar to the existing “Infi” methods that demand training data. Thus, the work proposed in this paper would be of great interest for the development of a reconfigurable floor cleaning robot since the complete area coverage is a foremost feature expected from a floor cleaning robot.