System-Level Optimization Model for a Snake-Inspired Robot Based on a Rectilinear Gait

Abstract

Physical parameters of modules and gait parameters affect the overall snake-inspired robot performance. Hence the system-level optimization model has to concurrently optimize the module parameters and the gait. The equations of motion associated with the rectilinear gait are quite complex due to the changing topology of the rectilinear gait. Embedding these equations in the system-level optimization model leads to a computationally challenging formulation. This paper presents a system-level optimization model that utilizes a hierarchical optimization approach and meta-models of the pre-computed optimal gaits to reduce the complexity of the optimization model. This approach enabled us to use an experimentally validated physics-based model of the rectilinear gait and yet at the same time enabled us to create a system-level optimization model with a manageable complexity. A detailed case study is presented to show the importance of concurrently optimizing the module parameters and the gait using our model to obtain the optimal performance for a given mission.

A survey of snake-inspired robot designs

“…In much the same way, future snake-inspired robots will need to be able to switch between multiple preprogrammed gaits types while in operation to adapt to unexpected changes in the terrain. In addition, gaits will need to be optimized during the operations [48]. Future snake-inspired robot designs will also need to be able to achieve much higher forward and turning velocities than what is currently possible.…”