Use this URL to cite or link to this record in EThOS:
Title: A layered control architecture for mobile robot navigation
Author: Qiu, Jiancheng
Awarding Body: University of Luton
Current Institution: University of Bedfordshire
Date of Award: 1998
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
This thesis addresses the problem of how to control an autonomous mobile robot navigation in indoor environments, in the face of sensor noise, imprecise information, uncertainty and limited response time. The thesis argues that the effective control of autonomous mobile robots can be achieved by organising low level and higher level control activities into a layered architecture. The low level reactive control allows the robot to respond to contingencies quickly. The higher level control allows the robot to make longer term decisions and arranges appropriate sequences for a task execution. The thesis describes the design and implementation of a two layer control architecture, a task template based sequencing layer and a fuzzy behaviour based low level control layer. The sequencing layer works at the pace of the higher level of abstraction, interprets a task plan, mediates and monitors the controlling activities. While the low level performs fast computation in response to dynamic changes in the real world and carries out robust control under uncertainty. The organisation and fusion of fuzzy behaviours are described extensively for the construction of a low level control system. A learning methodology is also developed to systematically learn fuzzy behaviours and the behaviour selection network and therefore solve the difficulties in configuring the low level control layer. A two layer control system has been implemented and used to control a simulated mobile robot performing two tasks in simulated indoor environments. The effectiveness of the layered control and learning methodology is demonstrated through the traces of controlling activities at the two different levels. The results also show a general design methodology that the high level should be used to guide the robot's actions while the low level takes care of detailed control in the face of sensor noise and environment uncertainty in real time.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: H671 Robotics ; robotics ; robot navigation ; layered control architecture