A novel approach for generalising walking gaits across embodiments and behaviours

Our goal is to introduce a more appropriate method of generalising walking gaits across different subjects and behaviours. Walking gaits are a result of complex factors that include variations resulting from embodiments and tasks, making techniques that use average template frameworks suboptimal for systematic analysis. The proposed work aims to devise methodologies for being able to represent gaits and gait transitions such that optimal policies may be recovered. The problem is formalised using a walking phase model, and the nullspace learning method is used to generalise a consistent policy. This policy can serve as reference guideline to quantify and identify pathological gaits. We have demonstrated robustness of our method with motion-capture data with induced gait abnormality. Future work will extend this to kinetic features and higher dimensional features.