Design, development and evaluation of an experimental platform for network robotics application

This thesis presents the design and the development of low-cost networked mobile robotic platforms which are used to study various mobile robot localization problems. These scaled robots can be designed to have some of the main features found, for example in the design of future gaming robots, assistive robots, or intelligent transportation while being cost effective. Two of the features of these robots are their ability to self-localize and their inter-robot communication capabilities. This thesis presents the design and implements a low-cost miniature mobile robot with such characteristics. The robot is able to estimate its position with its inertial navigation, which is accomplished using a Kalman filter that uses a temperature biased model of the measurements from micro-electro-mechanical systems (MEMS) sensors. Cross robot communication is achieved through a wireless mesh network. We also developed and evaluated a received signal strength indicator (RSSI) based localization system. Experimental results have shown the feasibility and practicality of the proposed set-up, which can be incorporated in various research and education in network dynamical systems.