Abstract
As urban landscape expands all over the world, there is an increasing need for an efficient pavement cleaning method to maintain the urban hygiene. Currently pavement cleaning robots are fixed in shape and faces limitation in cleaning efficiency as it is unable to access tighter areas. The outcome of the thesis is as follows. 1) Design and kinematic analysis of a reconfigurable sweeping robot. 2) Development of vision based approach for detection and classification of pavement width. 3) Development of a full scale autonomous reconfigurable sweeping robot using vision feedback control. 4) Experimental design and validation of the proposed research approach.