| Automatic guided vehicle is a carriage, traction robot which has been widely used in automated logistics, warehousing, manufacturing assembly line, food and other special industry. In recent years, with the rapid development and progress of science and technology, as well as a significant increase in the cost of human resources, the application of AGV has been more and more common in various fields. So far, we can find that the largest proportion of two kinds of AGV guidance form which operated in factory workshop are electromagnetic guided AGV and band guided AGV, these two guided modes belong to fixed path guidance means, their operation path can not be easily changed, with poor flexibility and high environmental requirement in the ground of operating region. But in actual production application, it usually need a lot of free path operation modes of AGV.According to the needs of navigation system when AGV running in the outdoor environment, this topic designed a kind of AGV positioning system, which is based on RTK-GPS and dead reckoning multiple-sensor integrated navigation and whose operation path is freedom. The main contents of this thesis are as follows:(1) The single point positioning principle of GPS navigation and the main influence factors of its measurement errors are studied. In order to meet the precision of AGV positioning when using GPS module as the absolute positioning method, differential GPS technology is used to process the measurement data, here three difference method is chosen to process the ranging code to realize Real Time Kinematic and to get centimeter level of navigation and positioning accuracy, which has been analyzed whether it can satisfy the feasibility of AGV positioning accuracy or not. Then the technology principle of Dead reckoning, electronic map and map matching between electronic maps and GPS navigation are introduced.(2) On the basis of studying the function of each module and basic requirement of the outdoor AGV positioning system, a integrated navigation system based on GPS/DR/MM is designed. Then the function of each sensor and each function module has been analyzed and the connection mode between are described.(3) The purpose and method of GPS/DR data fusion in integrated navigation structure are discussed, and Kalman filtering is used for data fusion of each sensor. The mathematical model of Kalman filtering theory is summed up and the combination structure of GPS/DR is selected, and the author set up a mathematical model of data fusion for this system by using Kalman filter algorithm, and further optimized the algorithm.(4) The electronic map software for the system is established using ArcMap software under the ArcGIS components, the Dijkstra algorithm is used as the shortest path planning algorithm of AGV traveling. Finally, running experiments is adopted to verify the feasibility of system. We can see that by using the way of compound guidance, it can improve the accuracy, fault tolerance and navigation stationarity of the system significantly, as well as made up the existing inherent defects of single guidance mode.The first part of this thesis mainly present the design principles of positioning system, and the effect and functions of sensors in this system, though these to ensure the feasibility of using this method for AGV navigation theoretically. Then made a more clear description of the mathematical model, algorithm, and software using in system, made them formed a certain system. Through the study of this positioning system, enriched the application of the outdoor AGV positioning system, gives us a new design sample or reference in the future about using GPS, DR and MM composite guidance for navigation of AGV positioning system. |
PDF Full Text Request