Research Of Image Guided Technology For Cable Tunnel Patrol Vehicle

Fire incidents caused by cable insulation fault offen occur in cable tunnel in recent years, which cause enormous economic losses. Traditional method for cable tunnel patrol is carried out with manual labor or by fixing monitoring instruments, to monitor the conditions of temperature and smoke. However,both fail to solve this problem thoroughly.Thus, an image guided AGV carrying monitoring instruments is proposed to be used in cable tunnel patrol in this thesis. In order to develop an experimental image guided AGV, two key techniques: path recognition and path tracking are deeply studied. The related software and hardware are designed, path tracking experiments are also carried out, the results of which verify the effectiveness of the methods proposed in this thesis.The accuracy and real-time ability are two key techniques of image guided AGV technology. In this thesis, the whole system for image acquiring and processing is designed, an iteration method is used to segment the image and mathematic morphology method is used to remove the image noise. Considering of the practical action, an improved field scanning method is used to calculate the position of the path in the image. As the experiment shows, this method could effectively suppress the noise interference, acquire the position parameters of the AGV, thus facilitates the control greatly.After the position parameters of AGV are acquired, a reasonable control strategy is required to track the path. As the AGV is a large delay and nonlinear system, it has obvious advantage to use the fuzzy control strategy. In this thesis, fuzzy control strategy is deeply studied, fuzzy controller considering our actual condition is designed. The periphery circuit is also designed. Heretofore, the whole experimental system of image guided AGV is successfully put up.Finally, path tracking under many circumstances are carried out, the host computer procedure is designed to collect the data of AGV motion. As the experiments show, the method proposed in this thesis could achieve good results, thus its effectiveness is verified.