Abstract
In this paper, we present the results of a comparison of two estimators of the gross vehicle weight (GVW) and the static load of individual axles of vehicles. The estimators were used to process measurement data derived from Multi-Sensor Weigh-In-Motion systems (MS-WIM). The term estimator is understood as an algorithm according to which the dynamic axle load measurement results are processed in order to determine the static load. The result obtained is called static load estimate. As a measure of measurement uncertainty, we adopted the standard deviation of the static load estimate. The mean value and the maximum likelihood estimators were compared. Studies were conducted using simulation methods based on synthetic data and experimental data obtained from a WIM system equipped with 16 lines of polymer axle load sensors. We have shown a substantially lower uncertainty of estimates determined using the maximum likelihood estimator. The results obtained have considerable practical significance, particularly during long-term usage of multi-sensor WIM systems.
Highlights
The basic aim of the application of data fusion is to reduce the uncertainty of measurement results or to increase the effectiveness of classification, detection, and location of the object
As the results of the study presented in the following part show, the application of estimator (7) is justified even when the Multi-Sensor Weigh-In-Motion systems (MS-weigh in motion (WIM)) system is equipped with 8–10 load sensors. This paper presents both the effects of simulation tests and the effects based on the results of measurements carried out at the MS-WIM field site
The metrological properties of various load sensors used in WIM systems are presented in detail in the papers [28,29,30,31,32]
Summary
The basic aim of the application of data fusion is to reduce the uncertainty of measurement results or to increase the effectiveness of classification, detection, and location of the object. This idea involves the joint use of signals and measurement data from many sensors and information derived from other sources (e.g., a priori knowledge). Competitive fusion, where different types of sensors are used to measure the same physical quantity This may lead to information redundancy complementary fusion, where each sensor is used to measure a different property of the studied object cooperative fusion, where the correct operation of a single sensor is dependent on the results of some other sensor. The operation of the first sensor would be impossible or undesirable
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
