Click on the titles below to expand the information about each abstract.
Viewing 1 results ...
Barati, K, Shen, X, Li, N and Carmichael, D G (2022) Automatic Mass Estimation of Construction Vehicles by Modeling Operational and Engine Data. Journal of Construction Engineering and Management, 148(03).
- Type: Journal Article
- Keywords: Earthmoving construction; Operational parameters; Engine load; Mass; On-road equipment; Artificial neural networks (ANN); Experimental studies;
- ISBN/ISSN: 0733-9364
- URL: https://doi.org/10.1061/(ASCE)CO.1943-7862.0002225
Earthmoving operations employ heavy-duty vehicles, including trucks and scrapers, to transport soil and rock on and off construction sites. Such construction activities are commonly scheduled and paid for based on the amount of earth moved. A number of metric and volumetric tools and techniques, including weighbridges, load–volume scanners (LVS), and strain gauges, have been developed to measure the payload of vehicles. These methods are costly, time-consuming, and labor-intensive, and may affect the production rate and cost of construction projects. This study develops an automatic mass estimation technique for on-road construction vehicles considering both operational and engine data. Acceleration rate, speed, and road slope are investigated as the operational variables, while engine load is considered as an engine attribute to estimate vehicle mass. A global positioning system-aided inertial navigation system (GPS-INS) and an engine data logger are integrated to collect the field data. Experiments are conducted on several construction vehicles to collect a wide range of data under various operational conditions. After assuring the quality of field data obtained in this study, artificial neural networks (ANNs) were developed to model the mass of construction equipment based on operational and engine parameters. The model was validated by comparing the estimated mass data with the actual values measured by a weighbridge in the experiment. The results show that the proposed model achieves greater than 90% accuracy in predicting the mass of on-road construction vehicles.