Navigation of an Automated Guided Vehicle in Indoor Environment

碩士 === 國立臺灣大學 === 農業機械工程學系 === 84 === The thesis studied the guidance of an automated guided vehicleAGV) in indoor environment. In order to illustrate theeasibility of the derived approach, an automated guided vehicleas built. The entire structure of the...

Full description

Bibliographic Details
Main Authors: Kuo,Lun-Chia, 郭倫嘉
Other Authors: J.J.Chou
Format: Others
Language:zh-TW
Published: 1996
Online Access:http://ndltd.ncl.edu.tw/handle/36056491147654382096
Description
Summary:碩士 === 國立臺灣大學 === 農業機械工程學系 === 84 === The thesis studied the guidance of an automated guided vehicleAGV) in indoor environment. In order to illustrate theeasibility of the derived approach, an automated guided vehicleas built. The entire structure of the AGV could be divided intoour modules:positioning module,navigation module, commuciationodule, and vehicle module. Concerning the positioning module,aCD camera was applied to acquire environment information. Throughhe process of calibration and mapping, the feature mark on theGV was recognized via Hough transformation. To generate the seedoint for the image processing, neuro network approach was usedo predicted the movement of the vehicle. In the navigationodule, environment model was constructed by cell decompositionethod, and then the optimal path was obtained through routingpproach. If necessary, B-spline curves could be used to generatesmooth path. Regarding the communication module, a navigationnd monitor/control computer communicated with the computer onhe vehicle by radio modems. As for the vehicle module, PWM wasdopted to drive motor and error was compensated through PDontroller. As a result, it was proved that a safe and optimal path coulde generated in the model-base environment. We also planned amooth path composed by B-spline curves, and the smooth path wasven better then the original path in terms of distance.egarding the performance of AGV, the deviation was around 2.7cmn a straight path with length 51.2cm. Yet there were turningoints in the 107.6cm planned path, the maximum error was up to.8cm or so.