- Title
- Natural gait in humanoid robots
- Creator
- Khan, Uzair Ijaz
- Relation
- University of Newcastle Research Higher Degree Thesis
- Resource Type
- thesis
- Date
- 2019
- Description
- Research Doctorate - Doctor of Philosophy (PhD)
- Description
- This dissertation focusses on understanding human locomotion from an engineering point of view. The idea is to explore the phenomena underlying the human walking gait, so that it can be replicated by a biped robot. The hypothesis is that the structure determines the gait while learning advances it. This dissertation is concerned with the first half of this hypothesis. The task is to understand the locomotion, dynamics and control which emanate from the structure, so that one can see how much learning contributes to actual adult human gait. Biped locomotion is complex to analyze in general. This is often because of the use of hybrid model, which makes the kinematics of the model easier to understand. However, it makes the analysis of such a model difficult because of the presence of discontinuous terms in its differential equation. A novel mathematical model is presented here and it is checked for the presence of natural oscillations. The model ensures that the dynamics of the biped are captured in one differential equation. The natural frequency of the biped model is calculated using the analytical tools at hand. A CPG based controller is then designed to ensure entrainment to the resonant frequency between the lower body leg angles. The controller is backed up by stability analysis and verified with sufficient simulation results. The gait obtained is supposed to be optimal, as it is utilizing the phenomenon of resonance. Hence, an objective function is thus formulated and an optimal gait is presented, which is then compared with the natural gait, analytically. The comparison is backed up with analysis as well as simulations. Lastly, postural stability of the biped is ensured with a controller to track the natural trajectory for the upper body.
- Subject
- bio-inspired control; CPG; biped; limit cycle
- Identifier
- http://hdl.handle.net/1959.13/1410338
- Identifier
- uon:36168
- Rights
- Copyright 2019 Uzair Ijaz Khan
- Language
- eng
- Full Text
- Hits: 698
- Visitors: 898
- Downloads: 291
Thumbnail | File | Description | Size | Format | |||
---|---|---|---|---|---|---|---|
View Details Download | ATTACHMENT01 | Thesis | 14 MB | Adobe Acrobat PDF | View Details Download | ||
View Details Download | ATTACHMENT02 | Abstract | 4 MB | Adobe Acrobat PDF | View Details Download |