Christian Vassallo

Email: Christian.Vassallo@laas.fr
Address: LAAS-CNRS
7, avenue du Colonel Roche
31077 Toulouse Cedex 4
FRANCE
Tel: +33 5 61 33 42 25
Fax: +33 5 61 33 69 69
Assistant: +33 5 61 33 64 69 (Delphine Fourcade)


About Me


I am a PhD Student of GEPETTO Team since October 2013 under the supervision of Philippe Souères and Olivier Stasse.

My thesis is part of the international research project “KoroiBot”, in collaboration with scientists from seven institutions in Germany, France, Israel, Italy and Netherlands.

Previously, after graduating in Computer Engineering I started my studies in the Robotics field with a Master’s in Robotics

Engineering at the University of Genoa, Italy. The collaboration with the EMARO programme gave me the possibility to be part of the ERASMUS program, completing my education at École Centrale de Nantes and the IRCCyN lab in France where I worked on my final project "Human to humanoid kinematic motion convertion based on virtual markers tracking". Here, I discovered my interest for the research.


Current Position


The thesis work consists to develop new software, inspired from human motion, to increase the walking capabilities of

the humanoid robots HRP-2 and Romeo of LAAS-CNRS. The research is based on mathematical models and methods, in particular way on optimization and learning techniques.

Basically, an investigation of human walking using motion capture system allows to extract the motion primitives of the

human in order to identify the motion principles for the generation and online control the robot gaits in various situations. The development of adequate rules for transferring human principles and movements to humanoid models and the application of optimization and learning based control approaches will allow a versatile, efficient and robust walking.


Thesis Summary


This thesis has been done within the framework of the European Project Koroibot which aims at developing ad- vanced algorithms to improve the humanoid robots locomotion. It is organized in three parts.

Part 1: How humans avoid moving obstacle crossing their way?
Study: Identification of walking strategies for avoiding a moving obstacle
Context: KoroiBot Project, Work Package 1
Collaborators: INRIA-Rennes, MimeTIC Research Team and University of Rennes 2, M2S Lab.

Summary
With the aim of steering robots in a safe and efficient manner among humans it is required to understand the rules, principles and strategies of human during locomotion and transfer them to robots. The goal of this thesis is to investigate and identify the human locomotion strategies and create algorithms that could be used to improve robot capabilities. A first contribution is the analysis on pedestrian principles which guide collision avoidance strategies. In particular, we observe how humans adapt a goal-direct locomotion task when they have to interfere with a moving obstacle crossing their way. We show differences both in the strategy set by humans to avoid a non-collaborative obstacle with respect to avoid another human, and the way humans interact with an object moving in human-like way.

Part 2: Learning Movement Primitives for the Humanoid Robot HRP-2
Study: ''Use of motion primitives to implement complex movements on humanoid robots
Context:
KoroiBot Project, Work Package 2
Partners:
University Clinic Tübingen, Department of Cognitive Neurology.''

Summary
In this part we present a work done in collaboration with computational neuroscientists from Tubingen, Germany. We propose a new approach to synthetize realistic complex humanoid robot movements with motion primitives. Human walking-to-grasp trajectories have been recorded. The whole body movements are retargeted and scaled in order to match the humanoid robot kinematics. Based on this database of movements, we extract the motion primitives. We prove that these sources signals can be expressed as stable solutions of an autonomous dynamical system, which can be regarded as a system of coupled central pattern generators (CPGs). Based on this approach, reactive walking-to-grasp strategies have been developed and successfully experimented on the humanoid robot HRP at LAAS-CNRS.

Video: https://cloud.laas.fr/index.php/s/aOk0kl5EqkVelAt

Part 3: The geometry of confocal curves for passing through a door
Study: Vision-based control to pass through a door
Context: ERC Actanthrope
Partners: Internal collaboration at LAAS-CNRS with Paolo Salaris and Jean-Paul Laumond.

Summary
In the third contribution of the thesis, we present a new approach to the problem of vision-based steering of robot subject to non-holonomic constrained to pass through a door. The door is represented by two landmarks located on its vertical supports. The planar geometry that has been built around the door consists of bundles of hyperbolae, ellipses, and orthogonal circles. We prove that this geometry can be directly measured in the camera image plane and that the proposed vision-based control strategy can also be related to human. Realistic simulation and experiments are reported to show the effectiveness of our solutions.

Video: https://cloud.laas.fr/index.php/s/Uu58q6dD6hYBKvX


Publications


Journals

International Conferences

Books and Book Chapters

Workshop