中文摘要：

针对基于浮动车辆数据（floating car data，FCD）的城市道路交通信息采集系统存在的问题，提出一种基于最小二乘支持向量机（LS-SVM）和证据理论的数据融合方法，通过融合地感线圈采集的交通流量信息，提高FCD系统交通速度信息采集的准确性．利用LS-SVM回归得到速度一流量关系曲线的临界速度参数，再根据历史数据库用统计方法计算出流量一速度关联规则的可信度矩阵，在得到这些经验知识的基础上，定义了两种证据源的基本概率分配函数．最后，通过D-S证据理论对两种证据源进行数据融合，获得融合后的速度信息．实地跑车实验结果论证了融合算法的有效性和可靠性．

英文摘要：

A traffic data fusion method was proposed using least squared support vector machine （LS-SVM） and evidence theory. Two types of original traffic data were collected. One was the traffic speed calculated from FCD （floating cars data） system. The other was the traffic volume collected by loop detectors. Firstly, the relationship curve between traffic speed and traffic volume was obtained by LS-SVM regression, and the critical speed was obtained from this curve. According to the above critical speed and the association matrix which converted the loop detector＇s volume into traffic speed, the basic probabilistic functions of the two sources were defined. Then the D-S theory was used to integrate these two sources of traffic information thus obtaining new traffic speed information. The driving-car experiment results show the effectivity and reliability of the proposed method.