Engineering efficient and massively parallel 3D self-reconfiguration using sandboxing, scaffolding and coating

Abstract Programmable matter based on modular self-reconfigurable robots could stand as the ultimate form of display system, through which humans could not only see the virtual world in 3D, but manipulate it and interact with it through touch. These systems rely on self-reconfiguration processes to reshape themselves and update their representation, using methods that we argue, are currently too slow for such applications due to a lack of parallelism in the motion of the robotic modules. Therefore, we propose a novel approach to the problem, promising faster and more efficient self-reconfigurations in programmable matter display systems. We contend that this can be achieved by using a dedicated platform supporting self-reconfiguration named a sandbox, acting as a reserve of modules, and by engineering the representation of objects using an internal scaffolding covered by a coating. This paper introduces a complete view of our framework for realizing this approach on quasi-spherical modules arranged in a face-centered cubic lattice. After thoroughly discussing the model, motivations, and making a case for our method, we synthesize results from published research highlighting its benefits and engage in an honest and critical discussion of its current state of implementation and perspectives.