System Design, Motion Modelling and Planning for a Recon figurable Wheeled Mobile Robot
Access status:
Open Access
Type
ThesisThesis type
Doctor of PhilosophyAuthor/s
Reid, WilliamAbstract
Over the past ve decades the use of mobile robotic rovers to perform in-situ scienti c investigations on the surfaces of the Moon and Mars has been tremendously in uential in shaping our understanding of these extraterrestrial environments. As robotic missions have evolved there ...
See moreOver the past ve decades the use of mobile robotic rovers to perform in-situ scienti c investigations on the surfaces of the Moon and Mars has been tremendously in uential in shaping our understanding of these extraterrestrial environments. As robotic missions have evolved there has been a greater desire to explore more unstructured terrain. This has exposed mobility limitations with conventional rover designs such as getting stuck in soft soil or simply not being able to access rugged terrain. Increased mobility and terrain traversability are key requirements when considering designs for next generation planetary rovers. Coupled with these requirements is the need to autonomously navigate unstructured terrain by taking full advantage of increased mobility. To address these issues, a high degree-of-freedom recon gurable platform that is capable of energy intensive legged locomotion in obstacle-rich terrain as well as wheeled locomotion in benign terrain is proposed. The complexities of the planning task that considers the high degree-of-freedom state space of this platform are considerable. A variant of asymptotically optimal sampling-based planners that exploits the presence of dominant sub-spaces within a recon gurable mobile robot's kinematic structure is proposed to increase path quality and ensure platform safety. The contributions of this thesis include: the design and implementation of a highly mobile planetary analogue rover; motion modelling of the platform to enable novel locomotion modes, along with experimental validation of each of these capabilities; the sampling-based HBFMT* planner that hierarchically considers sub-spaces to better guide search of the complete state space; and experimental validation of the planner with the physical platform that demonstrates how the planner exploits the robot's capabilities to uidly transition between various physical geometric con gurations and wheeled/legged locomotion modes.
See less
See moreOver the past ve decades the use of mobile robotic rovers to perform in-situ scienti c investigations on the surfaces of the Moon and Mars has been tremendously in uential in shaping our understanding of these extraterrestrial environments. As robotic missions have evolved there has been a greater desire to explore more unstructured terrain. This has exposed mobility limitations with conventional rover designs such as getting stuck in soft soil or simply not being able to access rugged terrain. Increased mobility and terrain traversability are key requirements when considering designs for next generation planetary rovers. Coupled with these requirements is the need to autonomously navigate unstructured terrain by taking full advantage of increased mobility. To address these issues, a high degree-of-freedom recon gurable platform that is capable of energy intensive legged locomotion in obstacle-rich terrain as well as wheeled locomotion in benign terrain is proposed. The complexities of the planning task that considers the high degree-of-freedom state space of this platform are considerable. A variant of asymptotically optimal sampling-based planners that exploits the presence of dominant sub-spaces within a recon gurable mobile robot's kinematic structure is proposed to increase path quality and ensure platform safety. The contributions of this thesis include: the design and implementation of a highly mobile planetary analogue rover; motion modelling of the platform to enable novel locomotion modes, along with experimental validation of each of these capabilities; the sampling-based HBFMT* planner that hierarchically considers sub-spaces to better guide search of the complete state space; and experimental validation of the planner with the physical platform that demonstrates how the planner exploits the robot's capabilities to uidly transition between various physical geometric con gurations and wheeled/legged locomotion modes.
See less
Date
2017-08-02Licence
The author retains copyright of this thesis. It may only be used for the purposes of research and study. It must not be used for any other purposes and may not be transmitted or shared with others without prior permission.Faculty/School
Faculty of Engineering and Information Technologies, School of Aerospace, Mechanical and Mechatronic EngineeringAwarding institution
The University of SydneyShare