A measurement of real travel speed for road segment with magnetic sensors

Abstract

Purpose -The purpose of this paper is to introduce a weighted average method to process speed measurements from multiple magnetic sensors, which are installed on road segments. Speeds are weighted-averaged in a fix duration time (5 minutes) for each sensor across location index of the sensor where it was installed. The proposed method is evaluated with numeric and simulation results. Design/methodology/approach -Unlike traditional vehicle average speed measurements, the authors propose a weighted-average speed measurement method of road segment, using wireless magnetic sensor nodes, which are installed on the measured road segment. Magnetic sensors offer a non-contact vehicle detection method, and small sensors with relatively low power consumption. Using magnetic sensors, the local changes in the Earth's magnetic field caused by the presence of a moving vehicle can be measured and the vehicle's speed obtained. Next, using adaptive weighted average algorithm and space weighted algorithm in a fixed period, the weighted average travel speed of road segment can be obtained. Findings -In current literature, there are many methods to measure vehicles' speed on road, such as image-based, radar-based, GPS-based, double-loop-based or magnetic sensor-based, but most of them only provide individual vehicle speed. Using probe vehicles, mean travel speed of road segment can be obtained, but it is costly on hardware and measurement, because many probe-vehicles need to be used on roads and many measurements need to be done everyday. GPS data can be used to provide valuable travel speed data for Intelligent Transportation System (ITS). However, not every vehicle is equipped with GPS and to access ID numbers for personal cars would entail privacy problems. Mean travel speed of road segment is obtained based on statistical average speed. Generally, statistical average speed is used, which is based upon Gaussian distribution is not true in traffic systems. Originality/value -By using wireless magnetic sensor nodes, vehicle instantaneous speeds are obtained in a fixed time when vehicles are passing over sensor nodes and then the adaptive weighted average speed on each sensor node location is computed based on the monitoring data from each sensor node in the fixed time. Considering different weights of each lane and road space (in the middle of the road segment or near the intersection), the proposed scheme can obtain the weighted-average speed of the road segment.