To accomplish many useful tasks, robots must be capable of navigating complex, dynamic environments. For this reason, robots often take inspiration from the adaptive responsiveness inherent to living organisms. Our existing model of rat neuromechanics [1] has provided insight into neuromuscular control structures that can generate quadruped locomotion. However, that model is limited by only using a single antagonistic muscle pair per hind limb joint. This work demonstrates a neuromuscular rat hind leg model of a complete muscle set with biologically derived attachment points as reported in the literature. The redundant nature of the muscle architecture provides new challenges for coordinating walking, including the need to account for synergistic muscle contractions. A linear decomposition of generalized muscle activation is applied to functional muscle groups responsible for actuating hind limb actuation. Functional group activation signals are then combined to recreate the generalized inputs previously determined to coordinate motion indicating a possible method of representing muscle synergy in the expanded model. This expanded hind limb model is fundamental to the analysis of three-dimensional motion as well as the impact of more sophisticated muscle synergy decompositions.
CITATION STYLE
Young, F., Hunt, A. J., & Quinn, R. D. (2018). A neuromechanical rat model with a complete set of hind limb muscles. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10928 LNAI, pp. 527–537). Springer Verlag. https://doi.org/10.1007/978-3-319-95972-6_57
Mendeley helps you to discover research relevant for your work.