Research On Path Planning And Positioning Of Vision AGV

With the continuous development of intelligent logistics,the research of Automated Guided Vehicle(AGV)has became the focus of many domestic and foreign manufacturers.However,the current key technologies of AGV cannot meet the urgent needs of the market.Therefore,in order to meet the needs of the market,a low-cost AGV system on the basis of Simultaneous Localization and Mapping(SLAM)technology is studied and designed in this paper.The main work of the thesis is as follows:First of all,the A~* algorithm of path planning is researched and optimized.After studying the traditional A~* algorithm,a new design method of heuristic function is proposed.At the same time,the least binary fork sorting is used to optimize the algorithm operation time.Finally,the Bezier curve is used to optimize the smoothness of the overall path generated by the algorithm.Among them,for the problem of encountering obstacles after optimization using Bezier curve alone,an algorithm of combining 2nd and 3rd Bezier curve is proposed.Through simulation analysis,the feasibility of this improved method is verified,which is more suitable for AGV path planning.Secondly,the positioning of SLAM algorithm is researched and optimized,and a positioning optimization algorithm combining EKF algorithm and QR code is proposed.A QR code technology which is simple to post,easy to identify,and easy to handle after recognition is proposed to eliminate the cumulative errors based on the theory of multi-sensor information fusion.At the same time,the location,quantity and processing methods of QR codes is studied.The following experimental results show that the fusion algorithm improves the accuracy of visual positioning and has high robustness.Finally,the whole design and construction of the visual AGV system is completed,and the optimized algorithm is verified by experimenting.Among them,the AGV platform design includes the overall system frame design and the layered design of the visual AGV navigation system.The path planning experiment and obstacle avoidance experiment is carried out on the visual AGV platform at the same time.The experimental results show that the visual AGV designed in this paper meets the requirements of flexibility and reliability,and has high applicability and robustness.