Global Localization and Local Path Planning of a Mobile Robot for Indoor Environments Using Stereo Vision and Ultrasonic Information

• Main Authors: Sheau-Huey Chen, 陳筱慧
• Other Authors: Chin-Shyurng Fahn
• Format: Others
• Language: en_US
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/30848732697210377324

碩士 === 國立臺灣科技大學 === 資訊工程系 === 96 === In recent years, there have been more and more researchers taking interest in the development of autonomous mobile robots. These robots can help people in daily life, such as office robot, entertainment robot, and security robot. There are many issues; for example, perception of environments, localization, path planning, map building, and execution action, about the robot navigation in indoor environments. To move accurately and safely in an indoor environment is one of the important functionality of robot navigation. The aim of this thesis is to make the robot can know its location and the position of the obstacles surrounding it, and guide the robot to avoid the obstacles without collision. On our experimental robot, the stereo vision system plays an important role during the navigation, and the ultrasonic sensors system acts as an assistant to help the robot walking safely. We adopt stereo vision techniques to localize robot's location using the known landmarks when the robot navigates in an indoor environment. In obstacle detection, we combine the edge and color features to detect obstacles; in the experiments, we use the specific colored cones as obstacles. After obtaining the environmental information from stereo vision and ultrasonic sensors, the robot can plan the local path. The planed path can keep the robot moving safely and correctly. Then the robot will stop the navigation if it reaches the destination where we set the place as the goal. The experiments reveal that the relative error range of the global localization results is about 0°~10.4° for the orientation, 0~0.21 m for the lateral location, and 0.05~0.12 m for the depth location. These global localization results are sufficient to provide for the local path planning which can guide the robot to move safely and to keep it on the expected trajectory in an indoor environment.