In general my research focuses on how advanced control schemes and machine learning can be integrated in order solve problems that would be otherwise intractable using simply one or the other.

One example involves the management and optimization of multiple tasks for highly redundant robots (humanoids specifically). Often, when a robot is directed to perform multiple tasks simultaneously, as in whole-body control, one or more tasks may interfere and prevent the robot from accomplishing the ensemble. By applying various artificial intelligence and machine learning tools on top of a whole-body controller, these conflicts can be managed and in many cases eliminated.


IROS 2015

author = {Lober, R and Padois, V and Sigaud, O},
title = {Variance Modulated Task Prioritization in Whole-Body Control},
booktitle = {IEEE/RSJ International Conference on Intelligent Robots and Systems},
year = {2015},
pages = {1--6}}


Variance Modulated Task Prioritization in Whole-Body Control

Ryan Lober, Olivier Sigaud, Vincent Padois

2015 IEEE/RSJ International Conference on Intelligent Robots and Systems


Whole-Body Control methods offer the potential to execute several tasks on highly redundant robots, such as humanoids. Unfortunately, task combinations often result in incompatibilities which generate undesirable behaviors. Prioritization techniques can prevent tasks from perturbing one another, but often to the detriment of the lower priority tasks. For many tasks, static prioritization is not necessary or even appropriate because the task is variable in nature, as in reaching. In this paper, we show that such task variability can be used to modulate task priorities during their execution, to temporarily deviate certain tasks as needed, in the presence of incompatibilities. We first present a method for mapping from task variance to task priority and then provide an approach for generating task variance when none is provided. Through three common conflict scenarios, we demonstrate that mapping from task variance to priorities reactively solves a number of task incompatibilities.

Humanoids 2014

author = {Lober, R and Padois, V and Sigaud, O},
title = {Multiple Task Optimization using Dynamical Movement Primitives for Whole-Body Reactive Control},
booktitle = {IEEE International Conference on Humanoid Robots},
year = {2014},
pages = {1--6}}



Whole-body controllers provide the tools to execute multiple simultaneous tasks on humanoid robots, but given the robot's internal and external constraints, interferences are often generated which impede task completion. Priorities can be assigned to each task to manage these interferences, unfortunately, this is often done at the detriment of one or more tasks. In this paper we present a novel framework for defining and optimizing multiple tasks in order to resolve potential interferences prior to task execution and remove the need for prioritization. Our framework parameterizes tasks with Dynamical Movement Primitives, simulates and evaluates their execution, and optimizes their parameters based on a general compatibility principle, which is independent of the robot's topology, tasks or environment. Two test cases on a simulation of a humanoid robot are used to demonstrate the successful optimization of initially interfering tasks using this framework.


Mingxing 2015

author = {Liu, M and Lober, R and Padois, V},
title = {Whole-Body Hierarchical Motion and Force Control for Humanoid Robots},
journal = {Autonomous Robots},
volume = {Special issue on Whole-body control for Humanoid Robots},
year = {2015},
pages = {1--13}}


Whole-Body Hierarchical Motion and Force Control for Humanoid Robots

Mingxing Liu, Ryan Lober, Vincent Padois

2015 Autonomous Robots Special issue on Whole-body control for Humanoid Robots


Robots acting in human environments usually need to perform multiple motion and force tasks while respecting a set of constraints. When a physical contact with the environment is established, the newly activated force task or contact constraint may interfere with other tasks. The objective of this paper is to provide a control framework that can achieve realtime control of humanoid robots performing both strict and non strict prioritized motion and force tasks. It is a torque-based quasi-static control framework, which handles a dynamically changing task hierarchy with simultaneous priority transitions as well as activation or deactivation of tasks. A quadratic programming problem is solved to maintain desired task hierarchies, subject to constraints. A generalized projector is used to quantitatively regulate how much a task can influence or be influenced by other tasks through the modulation of a priority matrix. By the smooth variations of the priority matrix, sudden hierarchy rearrangements can be avoided to reduce the risk of instability. The effectiveness of this approach is demonstrated on both a simulated and a real humanoid robot.


ISIR-WBC Introduction

In this video I give a more in depth introduction to the ISIRWholeBodyController and show some of the preliminary functionalities we have been working on over the past few weeks. The video is meant to be a sort of crash course in using the controller (for the courageous) so it is a bit long.

Right now we are only in the fixed base case, but we are hard at work on debugging the controller in simulation for the floating base case. In the immediate future we will be working on a trajectory tracking layer for smoother movement.


Coming soon...