Research On Indoor And Outdoor Path Planning Technology For Heavy-duty AGV Intelligent Drivin

With the progress of society and the high requirements in the industrial production process,AGV(Automated Guided Vehicle)has been widely used with the advantages of reliability,safety,high efficiency and low cost.This paper uses heavy-duty AGV as a research platform.In order to improve the efficiency of transportation and the safety of operation,the path planning algorithm and navigation and positioning are studied.1.The classification and general solution steps of AGV path planning algorithms are briefly introduced,and the principles of three common path planning algorithms are expounded.Combined with the indoor and outdoor operation scenarios of heavy-duty AGVs,the indoor and outdoor hybrid navigation and positioning scheme is determined.2.The indoor magnetic navigation based on magnetic stripe and the outdoor navigation based on visual lane line detection of heavy-duty AGV are studied.The improved incremental PID control algorithm through limiting,rounding and variable speed integration improves the deviation correction efficiency,and adopts the improved method based on adaptive threshold to extract edge points,which improves the detection and recognition rate of lane lines.At the same time,the INS/GNSS fusion positioning method is studied,and the results show that the cumulative error of INS can be effectively corrected by GPS,and the real-time position information of the target can be accurately obtained.3.Based on the actual operation scenario of overloaded AGV,the A* algorithm has been improved,adding dynamic weights and inflection point cost functions,and reducing the quadrant strategy of search nodes.The results show that compared with the traditional A* algorithm,the search time is shorter and less The inflection point is more suitable for the overloaded AGV of this article.The LPA* algorithm is studied and analyzed,and it is shown that the re-search of the A* algorithm has fewer search nodes and less search time in a dynamic environment.4.Finally,carry out the sports car test through the heavy-load AGV experimental platform,verify and analyze the magnetic navigation,the visual navigation of the lane line detection,the PID control algorithm and the INS/GNSS fusion positioning,and realize the upper computer control software through the QT platform,and verify the improvement The effectiveness of the path planning algorithm.