AGV Wireless Positioning And Tracking Technology Based On UWB

In order to ensure that Automatic Guided Vehicles(AGVs)can face the warehousing and logistics of complex environmental changes effectively,the most important thing is to obtain real-time and reliable location information.Ultra-Wideband(UWB)indoor positioning technology has gradually developed into the research front of high-precision indoor positioning technology is widely used in AGV positioning and navigation systems in large-scale work scenes due to its strong anti-interference,high precision,and low power consumption.However,the complex environments,such as factories and warehouses,complex physical characteristics,the presence of walls,and obstacles causing serious multipath effects and non-line-of-sight(NLOS)propagation,raise special challenges for reliable UWB positioning.In order to ensure that the AGV deliver the goods to the target location safely,quickly and accurately,this thesis proposes a high-precision indoor positioning and tracking algorithm with low cost,low power consumption and strong anti-interference:(1)For static target positioning,we give a Chan-Taylor positioning algorithm based on swarm intelligence algorithm.First,establish an iterative optimization model based on Bat Algorithm(BA),transform the problem of solving the target position into the minimum value of the difference between the estimated position and the actual position.We propose a random perturbation factor to improve the search ability of the individual bat.In order to solve the problem that the bat algorithm is easy to fall into the local optimal,we integrate the simulation annealing algorithm(SA)to improve the global search ability of the algorithm to obtain the estimated position of the target.Finally,the final position of the target is determined by the Chan-Taylor algorithm.When the base station is blocked to varying degrees,the research results show that,the proposed algorithm,compared with the existing NLOS elimination algorithm and positioning algorithm,improve the positioning accuracy in 11 ? 23 cm.Therefore,the proposed algorithm can improve the positioning accuracy of the positioning algorithm under the LOS/NLOS environment effectively.(2)For low-speed moving target localization,we present an object tracking algorithm based on Elman neural network(ENN)optimization and feedback learning algorithm.We propose an online training based on filter difference,measurement difference,and gain matrix of Interactive Multi-Model-Unscented Kalman Filter(IMM-UKF)to achieve feedback and optimize estimates simultaneously.In the feedback strategy,train the scale factor to adjust the error covariance matrix of the dynamic model adaptively;And in the optimization strategy,train the correction component to optimize the state vector to improve the final state estimation.ENN is a gradient descent method,which will appear that the training speed is slow and easy to fall into the local optimum,add Particle Swarm Optimization algorithm(PSO)to optimize the ENN threshold and weight to improve target tracking accuracy of the algorithm.Then combine the static target positioning algorithm toimprove precision.According to the simulation results,compared with the existing interactive multi-model filtering algorithms,the average estimated error of the algorithm in this thesis reduce by 32.19%,and convergence speed capability improve by 7% ? 18.2%.In summary,the UWB-based AGV static target positioning and dynamic target tracking algorithms in this thesis provide a new idea for real-time and reliable AGV positioning in the LOS/NLOS environment.