Analysis and generation of highly dynamic motions of anthropomorphic systems : application to parkour

Abstract : This thesis proposes an original and interdisciplinary approach to the treatment of whole-body human movements through the synergistic utilization of biomechanics, motor control and robotics. Robust methods of biomechanics are used to record, process and analyze whole-body human motions. The Uncontrolled Manifold approach (UCM) of motor control is extended to study highly dynamic movements processed in the biomechanical study, in order to determine if hypothesized dynamic tasks are being controlled stably by the central nervous system. This extension permits also to infer a hierarchical organization of the controlled dynamic tasks. The task space formalism of motion generation in robotics is used to generate whole-body motion by taking into account the hierarchy of tasks extracted in the motor control study. This approach permits to better understand the organization of human dynamic motions and provide a new methodology to generate whole-body human motions with anthropomorphic systems. A case study of highly dynamic and complex movements of Parkour, including jumps and landings, is utilized to illustrate the proposed approach.
Document type :
Theses
Liste complète des métadonnées

https://tel.archives-ouvertes.fr/tel-02014771
Contributor : Abes Star <>
Submitted on : Monday, February 11, 2019 - 5:12:06 PM
Last modification on : Friday, April 12, 2019 - 4:22:20 PM

File

2017TOU30375b.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : tel-02014771, version 1

Citation

Galo Xavier Maldonado Toro. Analysis and generation of highly dynamic motions of anthropomorphic systems : application to parkour. Robotics [cs.RO]. Université Paul Sabatier - Toulouse III, 2017. English. ⟨NNT : 2017TOU30375⟩. ⟨tel-02014771v1⟩

Share

Metrics

Record views

50

Files downloads

10