TY - GEN
T1 - Learning How to Walk
T2 - 2020 IEEE International Conference on Robotics and Automation
AU - Lembono, Teguh Santoso
AU - Mastalli, Carlos
AU - Fernbach, Pierre
AU - Mansard, Nicolas
AU - Calinon, Sylvain
N1 - Funding Information:
This work was supported by the European Union under the EU H2020 collaborative project MEMMO (Memory of Motion, http://www. memmo-project.eu/), Grant Agreement No. 780684.
Publisher Copyright:
© 2020 IEEE.
PY - 2020/9/15
Y1 - 2020/9/15
N2 - In this paper, we propose a framework to build a memory of motion for warm-starting an optimal control solver for the locomotion task of a humanoid robot. We use HPP Loco3D, a versatile locomotion planner, to generate offline a set of dynamically consistent whole-body trajectory to be stored as the memory of motion. The learning problem is formulated as a regression problem to predict a single-step motion given the desired contact locations, which is used as a building block for producing multi-step motions. The predicted motion is then used as a warm-start for the fast optimal control solver Crocoddyl. We have shown that the approach manages to reduce the required number of iterations to reach the convergence from ∼9.5 to only ∼3.0 iterations for the single-step motion and from ∼6.2 to ∼4.5 iterations for the multi-step motion, while maintaining the solution's quality.
AB - In this paper, we propose a framework to build a memory of motion for warm-starting an optimal control solver for the locomotion task of a humanoid robot. We use HPP Loco3D, a versatile locomotion planner, to generate offline a set of dynamically consistent whole-body trajectory to be stored as the memory of motion. The learning problem is formulated as a regression problem to predict a single-step motion given the desired contact locations, which is used as a building block for producing multi-step motions. The predicted motion is then used as a warm-start for the fast optimal control solver Crocoddyl. We have shown that the approach manages to reduce the required number of iterations to reach the convergence from ∼9.5 to only ∼3.0 iterations for the single-step motion and from ∼6.2 to ∼4.5 iterations for the multi-step motion, while maintaining the solution's quality.
UR - http://www.scopus.com/inward/record.url?scp=85092690165&partnerID=8YFLogxK
U2 - 10.1109/ICRA40945.2020.9196727
DO - 10.1109/ICRA40945.2020.9196727
M3 - Conference contribution
AN - SCOPUS:85092690165
T3 - Proceedings - IEEE International Conference on Robotics and Automation
SP - 1357
EP - 1363
BT - 2020 IEEE International Conference on Robotics and Automation, ICRA 2020
PB - IEEE
Y2 - 31 May 2020 through 31 August 2020
ER -