Abstract
Modular self-reconfigurable robots (SRRs) have redundant degrees of freedom and various configurations. There are two hard problems imposed by SRR features: locomotion planning and the discovery of multiple locomotion patterns. Most of the current research focuses on solving the first problem, using evolutionary algorithms based on the philosophy of searching-for-the-best. The main prob‐ lem is that the search can fall into a local optimum in the case of a complex non-linear problem. Another drawback is that the searched result lacks diversity in the behaviour space, which is inappropriate in addressing the problem of discovering multiple locomotion patterns. In this paper, we present a new strategy that evolves an SRR’s controller by searching for behavioural diversity. Instead of converging on a single optimal solution, this strategy discovers a vast variety of different ways to realize robot locomotion. Optimal motion is sparse in the behaviour space, and this method can find it as a by-product through a diversitykeeping mechanism. A revised particle swarm optimiza‐ tion (PSO) algorithm, driven by behaviour sparseness, is implemented to evolve locomotion for a variety of config‐ urations whose efficiency and flexibility is validated. The results show that this method can not only obtain an optimized robot controller, but also find various locomo‐ tion patterns.
Highlights
The major features of self-reconfigurable robots (SRRs) are adaptability and robustness, owing to their modularity [20]
This methodology has been applied to a maze robot and a simple humanoid planning simula‐ tion and it exhibited better performance compared with objective-based algorithms in trap problems and complex non-linear problems [9]
The results show that the behaviour sparseness-based (BSB)-particle swarm optimiza‐ tion (PSO) algorithm can be applied to evolve the locomotion for any configuration
Summary
The major features of self-reconfigurable robots (SRRs) are adaptability and robustness, owing to their modularity [20]. The mapping between the gene space and the behaviour space is non-linear, complex and difficult to find the law for the locomotion planning problem [15] These algorithms search for the best solutions directly and are likely to fall into a local optimum when dealing with complicated problems. This paper presents a new search strategy for addressing UBot SRR [19] locomotion problems: namely, motion planning and multiple patterns This strategy discards the philosophy of searching-for-the-best, i.e., taking objectivefunction value as the fitness value. The present paper is structured as follows: Section 2 introduces the search strategy based on behavioural sparseness and the evolutionary locomotion framework that is used; Section 3 describes the UBot SRR module and the 3D dynamics simulator, UBotSim; Section 4 evolves the locomotion for a worm-like configuration and some other typical configurations; Section 5 describes the algorithm’s suitability to discover multiple locomotion patterns for SRRs; Section 6 discusses the differences between two
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
