An efficient time-varying gesture-determined dynamical (TV-GDD) scheme is proposed for motion planning of redundant dual-arms manipulation. Motion planning for such tasks on humanoid robots with a high number of degrees-of-freedom (DOF) requires computationally efficient approaches to generate the expected joint configuration when given the end-effector tasks. To do so, we investigate a time-varying joint-limits constrained quadratic-programming (QP) approach and an efficient numerical computing method. This strategy provides feasible solutions at a low computation cost within physical limits. In addition, the joint configuration can be adjusted dynamically according to the expected gestures and tasks. Comparative simulations and experimental results on a humanoid robot demonstrate the effectiveness and feasibility of the scheme.
CITATION STYLE
Zhang, Z., Kong, L., & Niu, Y. (2018). A time-varying-constrained motion generation scheme for humanoid robot arms. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10878 LNCS, pp. 757–767). Springer Verlag. https://doi.org/10.1007/978-3-319-92537-0_86
Mendeley helps you to discover research relevant for your work.