Evolving Motion of Robots with Muscles

Abstract

Abstract. The objective of this work is to investigate how effective smart materials are for generating the motion of a robot. Because of the unique method of locomotion, an evolutionary algorithm is used to evolve the best combination of smart wire activations to move most efficiently. For this purpose, a robot snake was built that uses Nitinol wire as muscles in order to move. The most successful method of locomotion that was evolved, closely resembled the undulating motion of the cobra snake. During experimentation, one of the four Nitinol wires snapped, and the algorithm then enabled adaptive behaviour by the robot by evolving another sequence of muscle activations that more closely resembled the undulations exhibited by the earthworm.

An Evolutionary Approach to Damage Recovery of Robot Motion with Muscles

“…However, no one has used smart materials and evolutionary algorithms together in a robot before. This work uses nitinol wires as muscles within a robot snake, with the muscle activation sequences evolved using genetic algorithms and finite state machines [5].…”

“…The SAS used in the experiments uses twelve NiTi wires (diameter=0.176mm, activation (Austenite) temperature=70 o C, recommended current 200mA). The body of the robot snake is split into four segments (this is an extension from the first SAS prototype described in [5] which used only a single segment). These segments are readily detachable from each other and the whole structure can be expanded or segments replaced.…”

“…It has 20 pins and with the structure of the circuit board is capable of activating up to 12 pins in parallel (PTBx and PTDx excluding PTB0 and PTB3). [4,5] …”

“…The only wires leaving the snake are the power supply and RS232 (both designed to minimise weight and resistance to movement). Another important factor that was realised in previous experiments [4,5] and taken into consideration is that the resistance of the wires remains more or less constant no matter how far the SAS moves.…”

Learned mutation strategies in genetic programming for evolution and adaptation of simulated snakebot

“…The automated mechanisms for prompt generation of nearoptimal solutions to such complex, ill-posed problems are usually based on various models of learning (ontogenesis) or evolution (phylogenesis) of species in the Nature [7,9,16]. The proposed approach of employing genetic programming (GP) implies that the code, which controls the locomotion of the Snakebot is automatically designed by a computer system via simulated evolution through selection and survival of the fittest in a way similar to the natural evolution of species [8].…”