Research On Unmanned Vehicle Positioning Technology For Urban Delivery

The existing urban unmanned delivery vehicle positioning systems often need to be equipped with expensive sensors to solve the "indoor dilemma" of unmanned delivery.Considering that one of the original intentions of developing unmanned vehicles is to reduce the cost of distribution,this paper studies a low-cost urban unmanned distribution vehicle positioning technology that can achieve seamless indoor and outdoor positioning and meet the needs of "everything to home".The TCOFDM(Time&Code Division-Orthogonal Frequency Division Multiplexing)positioning system based on the fusion of communication and navigation signals.Combining it with the Inertial Navigation System(Inertial Navigation System,INS)based on the idea of vector tracking,that is,integrating INS on the TC-OFDM vector receiver.Performing deep fusion in this way can fully combining the advantages of the two and can enable unmanned vehicles to achieve indoor and outdoor positioning functions in the complex environment of urban distribution.The main research contents of this paper are as follows:1.Unmanned vehicles have serious short-term multipath effects in urban distribution environments,especially when they enter indoors.Aiming at this problem,a vector tracking baseband algorithm based on an improved Strobe phase detector is designed.The Strobe phase detector is improved by using ten correlators to suppress the multipath error in each channel of the receiver,and combined with the mutual assistance between each channel of the vector tracking structure,the positioning accuracy of the unmanned vehicle in the multipath environment is improved.2.Aiming at the problem that the traditional integrated navigation algorithm based on the minimum mean square error criterion cannot solve the problem of the decline of the filtering accuracy in the non-Gaussian noise environment of the unmanned vehicle integrated navigation system,an adaptive maximum correlation entropy unscented Kalman filter(AMUKF)algorithm is proposed.Combining the maximum correlation entropy criterion(MCC)with the unscented Kalman filtering algorithm(UKF)and improve the adaptability of the maximum correlation entropy criterion.The experimental results show that the proposed scheme has higher estimation accuracy than the unscented Kalman filtering algorithm based on the minimum mean square error criterion,and the filtering accuracy is improved by 33.55%.3.The experimental platform and data acquisition system based on software receiver are built.The TC-OFDM/INS data were fused by the software receiver according to the established deep combined structure and integrated navigation algorithm,and the sports car experiment was carried out in the environment of simulated urban distribution.The experimental results show that the average positioning error of the deep combined positioning scheme is about 0.6m,which can meet the indoor and outdoor positioning requirements of the unmanned vehicles in urban delivery.