الفهرس | يوجد فقط 14 صفحة متاحة للعرض العام |
المستخلص The aim of this thesis is to develop a robust controller of a mobile robot regardless of its morphology or malfunction. Many researches implemented a controller for a mobile robot with different techniques, some of them use a fuzzy system for the implemen- tation but they use a fixed rule base dependent on the mobile robot’s morphology causing a failure in the controller when a malfunction oc- curs. In this thesis, we overcome this problem by implementing this controller using (Evolvable Hardware) EHW. The EHW is designed to adapt, like the chameleon changing its color to blend with the environment to changes in task requirements or changes in the environment, through its ability to reconfigure its own hardware structure dynamically and autonomously. This capacity for adaptation is achieved by employing efficient search algorithms known as genetic algorithm. This adaptation has great potential for the de- velopment of innovative industrial applications. The Mobile robot controller used in this thesis is a fuzzy-genetic controller. The controller is implemented by hardware that can change the rule base according to the result generated from the genetic algo- rithm to be suitable with the robot’s morphology. By this method, the controller adjusts itself autonomously. Finally, we implement a new obstacle avoidance technique to over- come obstacle collision. The controller was implemented using Xilinx FPGA. SIM-MECH and Virtual Reality MATLAB’s toolboxes are used for simulating the mobile robots to evaluate its performance. Different mobile robot’s morphology (two wheel mobile robot, like car mobile robot and omniwheel mobile robot) were studied to check the proposed method. Simulations have been performed with or with- out obstacles and with or without malfunction in the robot. |