A Two Seismic Sensor Based Approach for Moving Vehicle Detection

摘要

A novel method for using roadside seismic (accelerometer) sensors to detect moving vehicles driven in traffic lanes is investigated in this paper. The method uses two seismic sensors installed on a road shoulder, with the recorded seismic signals split into fixed time intervals. The time delay of arrival (TDOA) is estimated by use of a generalized cross-correlation (GCC) approach. Further estimation methods can then be used to characterize various kinds of vehicle information, including vehicle speed, axle spacing, and driving direction. The accuracy of the approach is investigated using experimental results obtained from an urban street environment.