Environment Modelling Of Electromagnetic Echo And Localization Of Moving Platform Based On SLAM Method

In recent years,the emergence and rapid development of consumer UAV(unmanned aerial vehicle)have bring a big convenience to many different fields.However,due to the lack of supervision and technical support,the UAV has also caused some trouble in public safety and brought panic to the public.Researches in detection and tracking of low,small,slow(LSS)target,especially small UAV as the representative,have become an important concern in the target detection field.In this paper,to solve the problem of LSS target detection,we proposed a methodology to construct the map of the environment through an electromagnetic SLAM method.With the sensor system to locate the platform and continuous wave radar to detect the environment,the electromagnetic SLAM achieve the goal of real-time simultaneous localization and mapping,as well as obtaining the electromagnetic features of target.The main work is divided into three parts,radar echo simulation of strong scatters in complex environment,localization with wide-beam multi-antenna radar based on coherent angle measurement and real-time localization and mapping based on radar echo.The main work and contributions accomplished in this paper are as follows:1)Based on radar equation and ray tracing method,the echo simulation is transformed into calculations of time delay and radar's peak power.We have designed three different methods to simulation the radar echo,corresponding to distributed point target,continuous plate target and complex shaped target respectively.We obtained the bistatic RCS of every single point target through CST simulation while taking the multipath effect into account.For the continuous plate target,we divided it into two part,ends point and line,dealing ends points with the way above and using backscattering coefficient for the line.In this method,the real-time simulation demand is fulfilled and the efficiency is improved.2)Based on the coherent angle measurement method,in the case where the spacing of two antenna arrays is far bigger than half the wavelength,the DOA estimation of the radar echo signals is realized with the classical MUSIC algorithm,the directional antenna orientation information and sliding window accumulation methods.The distance between the target and the detection platform is calculated by matched filtering method,and the azimuth information of the target is obtained accurately in real time.3)Based on the classical SLAM algorithm,a PF-SLAM method is proposed,which utilizes the particle filter method to estimate the position of the platform.The algorithm can realize the simultaneous platform localization and map construction in the non-linear motion model and non-Gaussian environment.Weighted average processing of historical landmark location data and feedback correction of platform location are used to ensure the estimation of distributed point landmarks more accurate and convergent.4)Real-time environment map construction is realized by using the continuous target azimuth information obtained by echo simulation and interpretation.With window fitting,line segment clustering and weighted averaging method,the discrete point target map is transformed into line segment target map.The contour information of buildings in the environment is obtained accurately with the errors of distance and angle estimations less than 5%.