Research On The Integrated Localization Methods For Storage AGV

With the arrival of industry 4.0 era,many countries are planning the development of intelligent factories,intelligent cities,intelligent logistics,etc.As the key to the realization of intelligent logistics,the positioning of Automated Guided Vehicle(AGV)has gradually become a research and development hotspot in this field.Therefore,how to reduce the impact of complex warehousing environment on location information has an important theoretical and practical significance.In view of the complex warehousing environment,this paper first analyzes the advantages and disadvantages of several common indoor positioning technologies.Through the analysis,it can be found that a single positioning technology will have some disadvantages,making it difficult to meet the positioning needs of AGV.In order to improve the positioning accuracy,this paper proposes an Inertial Navigation System(INS)/Ultra Wide Band(UWB)/Landmark combination positioning method,which makes full use of the advantages of the three positioning technologies to make up for the various inaccuracies of different positioning technologies.At the same time,the experimental platform of INS/UWB/Landmark combination positioning for warehousing environment is constructed,which lays the foundation for verifying the performance of the algorithm proposed in this paper.In the INS/UWB/Landmark combination positioning method,when there is no landmark information,the data fusion filter is used to fuse the positioning information of INS and UWB to obtain the optimal position estimate at the current time;when there is landmark information,the data fusion filter is used to fuse the INS and the landmark data to obtain the optimal position estimate at the current time,which effectively improves the positioning accuracy.Based on INS/UWB/Landmark combined positioning,in order to overcome the influence of the Line-of-Sight(LOS)/Non-Line-of-Sight(NLOS)mixed environment on the combined positioning system,this paper proposes based on the loose and tight combination algorithm of the Interacting Multiple Model(IMM)algorithm.For the loose combination positioning algorithm,when there is no landmark information,the difference between the AGV position measured by INS and UWB is used as the observation vector,and the data fusion is performed through the Kalman filter(KF),and the AGV position measured by INS is corrected;when there is landmark information,the observation vector is the difference between the AGV position measured by the INS and the landmark,data fusion is performed by the KF,and the AGV position measured by the INS is corrected.For the tight combination positioning algorithm,when there is no landmark information,the distance difference between AGV and reference node measured by INS and UWB is taken as observation vector,and the data is fused by extended Kalman filter(EKF),and the AGV position measured by INS is corrected;when there is landmark information,the observation vector is the difference of AGV position measured by INS and landmark,and the data is fused by KF,and the AGV position measured by INS is also corrected.Experimental results shows that the IMM algorithm can effectively reduce the LOS/ NLOS mixed environmental error.The loose combination positioning strategy has the advantages of high fault tolerance and easy implementation,and the positioning accuracy is higher than the single positioning technology.The tight combination uses the original data of each sensor,which has the advantage of higher positioning accuracy,so the positioning accuracy is higher than the loose combination.