Research On AGV Hybrid Navigation And Positioning System Based On Inertia And Vision

Due to the continuous development of information technology and the reform of manufacturing industry,the intelligent transformation of production line has become an inevitable trend of modern industrial production in order to reduce human misoperation,reduce production cost of enterprises and improve production efficiency and product quality.With its high degree of automation,safety and flexibility,automatic guided vehicle plays a very key role in the whole production and transportation,therefore,it is significant to study AGV technology for this industrial automation change.At present,the main navigation methods of AGV are electromagnetic navigation,laser navigation,visual navigation,inertial navigation and hybrid navigation.Considering the advantages and disadvantages of various navigation methods,the AGV system with mixed inertial and visual navigation and positioning is proposed,and the work of this paper is as follows.(1)Design and completion of the hybrid inertial and vision-based navigation and positioning system,including electrical design,component selection,software design,the preparation of the upper computer interface and the construction of communication between each module.(2)To address the problem that QR codes are contaminated and cannot be recognized in the current navigation and positioning system using QR codes.In this regard,a series of artificial beacons are designed and a method based on convolutional neural network to identify artificial beacons is proposed.The results show that when the beacons are contaminated,the AGV can still recognize the information represented by them better,solving the problem that the information cannot be read when the beacons are contaminated.(3)Kalman filtering is performed on the output data of inertial navigation elements.In order to further improve the positioning accuracy of AGV navigation,a beacon positioning algorithm is proposed for different beacon patterns,so as to calculate the deviation angle and deviation distance of AGV,and the results show that this positioning algorithm can further correct the deviation of inertial navigation.Through tests and experiments,the results show that the inertial and vision-based AGV hybrid navigation and positioning system has high stability and good deviation correction capability,which has practical application value and can be further improved and put into practical use.