Abstract/Summary

This paper describes a novel approach to controlling Active Matter by interpreting it as a form of automata. Active Matter systems exhibit many different forms of complex behaviour, some of these systems exhibit memory, with previous stimulations affecting the reaction from future stimulations. Much work has been done to find alternative computational mediums in organic and chemical systems, additionally much work has been done in the field of soft robotics exploring compliant materials and actuators. Active Matter can serve both these fields and bridge between them, allowing a single material to house both the actuator and control. With these elements so close this allows for extremely compact robotic systems, where a minuet actuator can function semi-independently. Active Matter can be used to extend the computational ability of the system as they can be used as a computational resource since their behaviour can be compared to that of an automaton, transitioning from state to state with input signals. Using EAP gels as an example of Active Matter that exhibit the necessary behaviour, a framework is introduced that adapts the behaviour to a Moore machine by converting the movement and hysteresis of the matter to a series of states, using a simplified model. The directed graph is generated, and probability distribution of the states demonstrated. This shows the optimization ability of the matter where that the matter always displays an optimized form of the generalized directed graph. By creating a framework for these systems their motion can be used to their fullest potential, utilizing them as a path planning interpreter translating a sequence of inputs to a sequence of movements. This allows a portion of the control system to be offloaded onto the Active Matter itself allowing for the creation of compact integrated robotic systems.