Locomotion of both walking robots and biomechanical systems show redundancies in the joint angle trajectory required to move the leg to a certain position and biomechanical systems additionally in muscle recruition. This paper shows the use of efficient dynamics modeling, dynamics algorithms and optimal control techniques to solve for optimal solutions to (goal oriented or measured reference) motions that as well solve the redundancy problems. Forward dynamics calculations are performed to solve both problems synchronously. A new approach proposed by the authors to the forward dynamics simulation and optimization problem outperforms commonly used methods by two orders of magnitude in numerical efficiency.
展开▼
