Skip to Main content Skip to Navigation
Theses

Task compatibility and feasibility maximization for whole-body control

Abstract : Producing useful behaviors on complex robots, such as humanoids, is a challenging undertaking. Model-based whole-body control alleviates some of this difficulty by allowing complex whole-body motions to be broken up into multiple atomic tasks, which are performed simultaneously on the robot. However, modeling errors and assumptions, made during task planning, often result in infeasible and/or incompatible task combinations when executed on the robot. Consequently, there is no guarantee that the prescribed tasks will be accomplished, resulting in unpredictable, and most likely, unsafe whole-body motions. The objective of this work is to better understand what makes tasks infeasible or incompatible, and develop automatic methods of improving on these two issues so that the overall whole-body motions may be accomplished as planned. We start by building a concrete analytical formalism of what it means for tasks to be feasible with the control constraints and compatible with one another. Using the model-based feasibility and compatibility metrics, we demonstrate how the tasks can be optimized using non-linear model predictive control, while also detailing the shortcomings of this model-based approach. In order to overcome these weaknesses, an optimization loop is designed, which automatically improves task feasibility and compatibility using model-free policy search in conjunction with model-based whole-body control. Through a series of simulated and real-world experiments, we demonstrate that by simply optimizing the tasks to improve both feasibility and compatibility, complex and useful whole-body motions can be realized.
Document type :
Theses
Complete list of metadatas

Cited literature [177 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-01927038
Contributor : Abes Star :  Contact
Submitted on : Monday, November 19, 2018 - 3:42:07 PM
Last modification on : Sunday, October 25, 2020 - 4:58:43 PM
Long-term archiving on: : Wednesday, February 20, 2019 - 3:01:12 PM

File

2017PA066597.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-01927038, version 1

Citation

Ryan Lober. Task compatibility and feasibility maximization for whole-body control. Robotics [cs.RO]. Université Pierre et Marie Curie - Paris VI, 2017. English. ⟨NNT : 2017PA066597⟩. ⟨tel-01927038⟩

Share

Metrics

Record views

151

Files downloads

138