A thesis submittion in partial fulfillment of the requirements for the degree of Master of Science in Information Management, School of Engineering and Technology

Thesis (M.Sc.) - Asian Institute of Technology, 2014

Due to constraints on cost, power consumption, perceptual capabilities, knowledge manage-ment, reasoning capabilities, and compute power, the adoption of advanced service robotshas thus far been limited. In order to spread the adoption of service robots, further develop-ment is crucial. This thesis is study on the use of cloud computing platforms that supportrobotic applications, specifically object recognition, expanding the compute compute andstorage capacity of the robot system. The thesis presents a system architecture, implementedprototype, and initial experimental data on a cloud infrastructure platform for recognition ofspecific 3D objects. The objective is to prove the concept of enabling robot to share knowl-edge with others via cloud services while decreasing the required cost and power consump-tion. Through testing of the speed and accuracy of recognition on the developed prototypearchitecture, I proved that the cloud platform architecture is a feasible solution for 3D objectrecognition.