卫星拒止环境下危险货物运输车辆融合定位方法  

A Fusion Positioning Strategy of Vehicle for Road Transport of Dangerous Goods Under Satellite Denied Environment

作　　者：宋翔 任春晓[2,3] 杨砾 邝井国 张竟[6] SONG Xiang;REN Chunxiao;YANG Li;KUANG Jingguo;ZHANG Jing(School of Electronic Engineering,Nanjing Xiaozhuang University,Nanjing 211171,China;Research Institute of Highway,Ministry of Transport,Beijing 100088,China;School of Instrument Science and Engineering,Southeast University,Nanjing 210096,China;Beijing Prominion Publishing Co.,Ltd.,Beijing 100083,China;Hangzhou Hikvision Digital Technology Co.,Ltd.Hangzhou 310051,China;Sany Automobile Manufacturing Co.,Ltd.,Changsha 410000,China)

机构地区：[1]南京晓庄学院电子工程学院,南京211171 [2]交通运输部公路科学研究院,北京100088 [3]东南大学仪器科学与工程学院,南京210096 [4]北京卓众出版有限公司,北京100083 [5]杭州海康威视数字技术股份有限公司,杭州310051 [6]三一汽车制造有限公司,长沙410000

出　　处：《交通信息与安全》2020年第5期145-152,共8页Journal of Transport Information and Safety

基　　金：国家重点研发计划课题项目(2017YFC0804804);国家自然科学基金项目(41904024)、江苏高校“青蓝工程”项目资助。

摘　　要：为实现卫星拒止环境下危险货物运输车辆的准确可靠定位,建立一种RFID/车载低成本传感器融合定位方法。以RFID输出的信号强度信息作为信息源,采用最小二乘支持向量机方法准确估计读写器与标签之间距离,并具备不同工作环境下的高泛化能力;引入车载传感器信息,建立改进车辆运动状态模型以准确描述车辆运行状态,设计一种自适应分散化信息滤波方法实现无线射频与车载传感器信息的融合定位。为有效隔离RFID失效信息,并实现异源异步信息融合,采用分散化架构而非传统集中式滤波实现融合,同时,在滤波器中提出并融入自适应规则判断RFID信息的有效性以决定是否隔离故障信息,从而提升融合算法的准确性和鲁棒性。实车试验结果表明,融合定位精度达到了无遮挡环境下GPS的定位精度,相比于无线射频定位提升了58%,相比于航位推算提升了68%,相比于传统卡尔曼滤波融合提升了65%。特别是在存在标签失效情形下,性能的提升更为显著。Under satellite denied environment,to realize accurate and reliable positioning of vehicle for road transport of dangerous goods,a fusion vehicle positioning strategy based on integration of RFID and low-cost in-vehicle sensors is proposed.Adopting strength of the signal received as information source,range from RFID tags to the reader is estimated by implementing the LSSVM algorithm,with advantage of accuracy and generalization ability adapting to various environments.Then,in-vehicle sensors are introduced to fuse with the range from RFID.The adaptive decentral ized information filtering(ADIF)method is developed to achieve fusion.In the implementation process of ADIF,an improved vehicle motion model is established to accurately describe motion of the vehicle.To isolate the RFID failure and fuse multiple observation sources with different sample rate,instead of the centralized EKF,the decentralized ar chitecture is employed.Meanwhile,the adaptive rule is designed to judge the effectiveness of RFID information,decid ing whether to exclude RFID observations.The proposed strategy is verified through field tests.The results show that the proposed method achieves a positioning accuracy equivalent to GPS in unobstructed environments.58%accuracy improvement over RFID method,68%accuracy improvement over DR method,and 65%accuracy improvement over conventional EKF are achieved.Especially in the presence of tag failure,the performance improvement is more signifi cant.

关 键 词：智能交通 危货运输车辆 卫星拒止环境定位 自适应分散化信息滤波 无线射频技术 多传感器融合 

分 类 号：U492.22[交通运输工程—交通运输规划与管理]