Repository logo
 

TOWARDS A NOVEL RESILIENT ROBOTIC SYSTEM

dc.contributor.committeeMemberZhang, Wenjun
dc.contributor.committeeMemberGupta, Madan M
dc.contributor.committeeMemberZhang, Dan
dc.contributor.committeeMemberChen, Xiongbiao
dc.contributor.committeeMemberLanovaz, Joel
dc.contributor.committeeMemberBui, Francis
dc.contributor.committeeMemberXi, Fengfeng
dc.creatorZhang, Tan
dc.date.accessioned2017-02-07T18:21:48Z
dc.date.available2017-02-07T18:21:48Z
dc.date.created2015-01
dc.date.issued2015-02-12
dc.date.submittedJanuary 2015
dc.date.updated2017-02-07T18:21:49Z
dc.description.abstractResilient robotic systems are a kind of robotic system that is able to recover their original function after partial damage of the system. This is achieved by making changes on the partially damaged robot. In this dissertation study, a general robot, which makes sense by including active joints, passive joints, passive links, and passive adjustable links, was proposed in order to explore its resilience. Note that such a robot is also called an under-actuated robot. This dissertation presents the following studies. First, a novel architecture of robots was proposed, which is characterized as under-actuated robot. The architecture enables three types of recovery strategy, namely (1) change of the robot behavior, (2) change of the robot state, and (3) change of the robot configuration. Second, a novel docking system was developed, which allows for the realization of real-time assembly and disassembly and passive joint and adjustable passive link, and this thus enables the realization of the proposed architecture. Third, an example prototype system was built to experiment the effectiveness of the proposed architecture and to demonstrate the resilient behavior of the robot. Fourth, a novel method for robot configuration synthesis was developed, which is based on the genetic algorithm (GA), to determine the goal configuration of a partially damaged robot, at which the robot can still perform its original function. The novelty of the method lies in the integration of both discrete variables such as the number of modules, type of modules, and assembly patterns between modules and the continuous variables such as the length of modules and initial location of the robot. Fifth, a GA-based method for robot reconfiguration planning and scheduling was developed to actually change the robot from its initial configuration to the goal configuration with a minimum effort (time and energy). Two conclusions can be drawn from the above studies. First, the under-actuated robotic architecture can build a cost effective robot that can achieve the highest degree of resilience. Second, the design of the under-actuated resilient robot with the proposed docking system not only reduces the cost but also overcomes the two common actuator failures: (i) an active joint is unlocked (thus becoming a passive joint) and (ii) an active joint is locked (thus becoming an adjustable link). There are several contributions made by this dissertation to the field of robotics. The first is the finding that an under-actuated robot can be made more resilient. In the field of robotics, the concept of the under-actuated robot is available, but it has not been considered for reconfiguration (in literature, the reconfiguration is mostly about fully actuated robots). The second is the elaboration on the concept of reconfiguration planning, scheduling, and manipulation/control. In the literature of robotics, only the concept of reconfiguration planning is precisely given but not for reconfiguration scheduling. The third is the development of the model along with its algorithm for synthesis of the goal reconfiguration, reconfiguration planning, and scheduling. The application of the proposed under-actuated resilient robot lies in the operations in unknown or dangerous environments, for example, in rescue missions and space explorations. In these applications, replacement or repair of a damaged robot is impossible or cost-prohibited.
dc.format.mimetypeapplication/pdf
dc.identifier.urihttp://hdl.handle.net/10388/ETD-2015-01-1919
dc.identifier.urihttp://hdl.handle.net/10388/7736
dc.subjectResilient robot, Self-recovery, Self-reconfiguration, Architecture, Configuration Synthesis
dc.titleTOWARDS A NOVEL RESILIENT ROBOTIC SYSTEM
dc.typeThesis
dc.type.materialtext
thesis.degree.departmentBiomedical Engineering
thesis.degree.disciplineBiomedical Engineering
thesis.degree.grantorUniversity of Saskatchewan
thesis.degree.levelDoctoral
thesis.degree.nameDoctor of Philosophy (Ph.D.)

Files

Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Zhang, Tan_PhD_thesis_January_20151.pdf
Size:
5.77 MB
Format:
Adobe Portable Document Format
License bundle
Now showing 1 - 2 of 2
No Thumbnail Available
Name:
license1.txt
Size:
1002 B
Format:
Plain Text
Description:
No Thumbnail Available
Name:
license.txt
Size:
1002 B
Format:
Plain Text
Description: