Navigation of a mobile robot using stereo vision

Computer vision system is one of the main aplications in the mobile robot perception systems. Mobile robots may be able to navigate in an unknown environment acquiring visual information of their surroundings with the aim of estimating the position and orientation of an obstacle which is in front of it. Moreover, the pose of the robot has to be estimated as accurate as possible during navigation. The main purpose of this thesis is to develope stereo vision depth perception and obstacle detection systems for collision free navigation of a real hydraulically actuated experimental crawling mobile robot in an unknown indoor environment. Within the scope of this thesis, a robust stereo depth perception and obstacle detection algorithms are developed. To express our approach easily, a Stereo Vision Graphical User Interface is also prepared. In addition in this study for safely navigation of the mobile robot within its environments and reliably move from start point to destination point, a navigation algorithm, obstacle avoidance strategy and motion planning algorithms are developed. In order to implement the prepared perception algorithm and navigation algorithms, an experimental crawling vehicle was constructed and modified into a mobile robot and architecture details such as robot hardware and software were described in this thesis. To accomplish position control of mobile robot in two dimensional plane, kinematics of the mobile robot was analyzed and to control the robot locomotion system, a motion control algorithm was developed by the use of the kinematic data of the robot. In order to control the crawling robot easily and to combine all developed algorithms, Manual and Autonomous Navigation Graphical User Interfaces were prepared and implemented on the robot. After optimization and combining all algorithms and codes, behavior of robot was observed. With the helps of developed algorithms, the robot was able to reach destination point without collision with any obstacles.