Locomotion Control System of a Humanoid Robot. A biologically inspired model
Wauquaire, Odile
Promotor(s) : Boigelot, Bernard
Date of defense : 7-Sep-2017/8-Sep-2017 • Permalink : http://hdl.handle.net/2268.2/3169
Details
Title : | Locomotion Control System of a Humanoid Robot. A biologically inspired model |
Translated title : | [fr] Système de contrôle de locomotion d'un robot humanoïde - Un modèle insprié de la biologie |
Author : | Wauquaire, Odile |
Date of defense : | 7-Sep-2017/8-Sep-2017 |
Advisor(s) : | Boigelot, Bernard |
Committee's member(s) : | Mathy, Laurent
Drion, Guillaume Bruls, Olivier |
Language : | English |
Number of pages : | 55 |
Keywords : | [en] RoboCup [en] Controller [en] Control method [en] robot [en] bipedal [en] dynamic walking [en] humanoid robot [en] biological [en] locomotion [en] simulation |
Discipline(s) : | Engineering, computing & technology > Computer science |
Institution(s) : | Université de Liège, Liège, Belgique |
Degree: | Master en ingénieur civil en informatique, à finalité spécialisée en "intelligent systems" |
Faculty: | Master thesis of the Faculté des Sciences appliquées |
Abstract
[en] The objective of this master thesis is to provide the RoboCup team of the University of Liège with a locomotion system for their humanoid robot. RoboCup is an international competition, the ultimate goal of which is to built robotic soccer players that would be able to beat the human World Cup champion team in 2050. This task is threefold. First, a locomotion control method has to be chosen. Second, the architecture and design of the method, adapted to our robot, have to be detailed. Third, tests have to be conducted in a simulator.
The approach chosen is biologically inspired and has been created by Tobias Luksch in his PhD thesis “Human-like Control of Dynamically Walking Bipedal Robot”. This method was proposing two locomotion modes: keeping the balance while standing and walking dynamically. In this work, the method is summarized and the adapted architecture is presented. Every elements of the method has been implemented for the first locomotion mode. However, the adaptation for the second locomotion mode does not require a substantial work as the skeleton would be the same as for the first one and a lot of implementation challenges have already been tackled. Then, this stable standing locomotion mode has been implemented in the Blender simulator. The implementation on a simplified model first allows to validate the method. Then, several limitations of the simulator have been encountered for the complete model of the robot. The physics engine does not always behave realistically, and the method would probably give better results in reality, out of the simulation environment.
To conclude, the implementation of the chosen control method has been completed by giving solutions to technical and design challenges. The approach chosen seems very promising after the tests performed on the simple model. Nevertheless, the choice of another simulator is necessary to complete the tests and the implementation of other locomotion modes.
File(s)
Document(s)
Annexe(s)
Cite this master thesis
The University of Liège does not guarantee the scientific quality of these students' works or the accuracy of all the information they contain.