On the Analogy in the Emergent Properties of Evolved Locomotion Gaits of Simulated Snakebot

Wheelless, limbless snake-like robots (Snakebots) feature potential robustness characteristics beyond the capabilities of most wheeled and legged vehicles – ability to traverse terrain that would pose problems for traditional wheeled or legged robots, and insignificant performance degradation when partial damage is inflicted. Moreover, due to their modular design, Snakebots may be cheaper to build, maintain and repair. Some useful features of Snakebots include smaller size of the cross-sectional areas, stability, ability to operate in difficult terrain, good traction, high redundancy, and complete sealing of the internal mechanisms (Dowling, 1997), (Worst, 1998). Robots with these properties open up several critical applications in exploration, reconnaissance, medicine and inspection. However, compared to the wheeled and legged vehicles, Snakebots feature (i) smaller payload, (ii) more difficult thermal control, (iii) more difficult control of locomotion gaits and (iv) inferior speed characteristics. Considering the first two drawbacks as beyond the scope of our work, and focusing on the issues of control and speed, we intend to address the following challenge: how to develop control sequences of Snakebot’s actuators, which allow for the fastest possible speed of locomotion achievable with Snakebot morphology. For many tasks and robot morphologies, it might be seen as a natural approach to handcraft the locomotion control code by applying various theoretical approaches (Burdick et al, 1993), (Hirose, 1993), (Zhang et al, 22). However, handcrafting might not be feasible for developing the control code of a real Snakebot due to its morphological complexity and the anticipated need of prompt adaptation under degraded mechanical abilities and/or unanticipated environmental conditions. Moreover, while a fast locomotion gait might emerge from relatively simply defined motion patterns of morphological segments of Snakebot, the natural implication of the phenomenon of emergence in complex systems is that neither the degree of optimality of the developed code nor the way to incrementally improve the code is evident to the human designer (Morowitz, 2002). Thus, an automated, holistic approach for evaluation and incremental optimization of the intermediate solution(s) is needed (e.g. based on various models of learning or evolution in Nature) (Kamio et al, 2003), (Mahdavi &