Automated Guided Vehicle System Based On The Multi-sensor Combination Of Research And Applications

Automatic Guided Vehicle (AGV, Automatic Guided Vehicle) has a variety of application prospects and huge values in manufacturing fields like flexible manufacturing, logistics and automated storage, etc. It has become an important and indispensable part of modern enterprises in automation equipments. With the continuous development of AGV technology, there have been machine-vision based guidance technologies, which quickly became a hot topic of research. Compared with the traditional guidance technologies, vision-guided technology has an advantage of lower cost, more adaptable and more flexible installation. However, the existing vision-guided technology is not mature enough yet, which mainly reflected in the deficiency of stability and robustness.This paper presented a multi-sensor based control system for AGV guidance, which using CCD camera, RFID, encoder, gyro, sonar and other sensors. We achieved in AGV control by using an approach of information fusion and combined guidance. The paper introduces the following aspects:First of all, a multi-sensor information fusion technology has been analyzed and discussed to solve the pose estimation problem in the process of AGV guidance. Kalman filtering has been used to fuse the information of encoders and electronic gyroscopes, and then been verified by experiments.Then, a multi-sensor based guidance technology has been introduced, which is a combination approach of machine vision, RFID, encoders, gyroscopes and other sensors. An expert-PID control method and a fuzzy control method are also introduced, which take into account different driving conditions.Thirdly, in order to run the whole system on a limited resources platform, a well-designed software architecture that is based on data flow techniques has been implemented.Finally, the multi-sensor based AGV guidance system verified by experiments has been proved to meet all project requirements.