Research On Vehicle Tracking Based On Intelligent Reflecting Surface In Complex Environment

Vehicle tracking is the core research content in unmanned driving,intelligent transportation,and the Internet of Things and has important theoretical significance and application prospects.Existing methods mainly utilize GNSS and roadside infrastructure to achieve vehicle tracking.Moreover,based on the roadside infrastructure,the vehicle obtains the distance parameters,such as the received signal strength,transmission delay,and direction of arrival in the roadside unit,and then applies the Kalman filter to predict the vehicle’s position.At present,domestic and foreign scholars have proposed many practical vehicle tracking algorithms based on roadside infrastructure.These algorithms usually rely on a line-of-sight link between the vehicle and infrastructure.However,in complex environments,the line-of-sight link between the vehicle and infrastructure may be absent,and the vehicle’s initial position is unknown.These factors reduce vehicle tracking performance and even fail the algorithm.In addition,in the multi-vehicle tracking system,the master vehicle is usually used to guide the slave vehicles,but the improper selection of the master vehicle further reduces the algorithm’s robustness.Therefore,overcoming the lack of line-ofsight link,acquiring the vehicle’s initial position,and selecting the master vehicle are the focus and difficulty problems of vehicle tracking in complex environments.As an emerging technology,the intelligent reflecting surface has attracted extensive attention from academia and industry due to its ability to customize wireless environments.The intelligent reflecting surface can jointly control numerous units in a software-defined manner to achieve the directional reflection of incident signals.Specifically,by optimizing the phase parameters of the intelligent reflecting surface,a virtual line-of-sight link between the roadside unit and vehicle can be established and realize the initial positioning and tracking.In the master-slave vehicle system,the intelligent reflecting surface can select the suitable master vehicle to guide the slave vehicle,thereby improving the robustness of the tracking algorithm.In summary,it is meaningful to study the vehicle tracking algorithm based on the intelligent reflecting surfaces to realize the initial positioning and tracking of the vehicle and host vehicle selection when line-of-sight links are absent.Therefore,this paper has researched vehicle tracking methods and master vehicle selection for complex environments,such as unknown vehicle’s initial position and lack of line-of-sight links.The specific research work is as follows:（1）This paper summarizes the related research on vehicle tracking and intelligent reflecting surfaces and introduces mathematical theories such as Kalman filter theory and simulated annealing algorithm for the follow-up vehicle tracking based on intelligent reflecting surfaces in complex environments.The method of research is carried out to lay the foundation.（2）Aiming at the vehicle’s initial position is unknown,and the line-of-sight link is absent in complex environments.This paper proposes a tracking algorithm based on intelligent reflecting.The algorithm first combines IRS（intelligent reflecting surface）sampling technology and simulated annealing algorithm and suggests an improved simulated annealing algorithm to realize the initial positioning of the vehicle.Based on the Kalman filter,two vehicle tracking algorithms,LOS-UKF and VLOS-UKF,are proposed,respectively.Finally,to improve the robustness of the tracking algorithm,a time delay vector comparison strategy is proposed to correct the vehicle position.Simulation experiments show that the proposed algorithm can achieve high-precision vehicle positioning and tracking performance,and can achieve effective position correction when the vehicle’s predicted trajectory deviates.（3）Aiming at the problem of multi-vehicle tracking in complex environments,this paper proposes a master-slave vehicle tracking algorithm based on intelligent reflecting surfaces.The algorithm firstly constructs the master vehicle selection strategy to select the appropriate master vehicle and guides the slave vehicle for tracking.Secondly,with the assistance of the intelligent reflecting surface,the time delay vector is constructed so that the master vehicle can obtain reliable position information.Finally,the observation vector is combined with the local inertial navigation system,and the unscented Kalman filter is used to realize the master-slave vehicle tracking.Simulation experiments show that the proposed algorithm can select the most suitable master vehicle and guide all slave vehicles to achieve high-precision tracking performance.The experimental results show that the algorithm can achieve reliable master-slave vehicle tracking.