Non-Contact Vehicle Weighing Method Based On Tire Mechanics And Computer Vision

With the rapid development of the transportation industry,vehicle overloading has become more and more serious.Overloading vehicle is one of the important reasons for the damage and failure of traffic infrastructures,and also endangers the safety of human life.Vehicle weight identification is the prerequisite for overload management.It is also important for the service performance and life assessment of transportation infrastructures.Existing vehicle weighing methods,including static weighing,pavement weigh-in-motion,and bridge weigh-in-motion,have some disadvantages,such as high cost and poor durability.Due to various problems existing in current vehicle weighing methods,the objective of this study is to establish a novel non-contact vehicle weighing method based on tire mechanics and computer vision.The main contents of this study are as follo ws:(1)The tire force-deformation equation is established.The geometric structure of the tire is analyzed by the mechanical model during the deformation process.The tire contact length,contact width,and contact area are represented by the tire structure parameters.Based on the principle that the tire force is the product of inflation pressure and contact area,a theoretical calculation equation of tire force-deformation is established.(2)The established tire force-deformation equation is verified and modified through finite element analysis,and a new tire footprint calculation model is proposed.The tire finite element models of different specifications are established and analyzed under different working cases.Simulation results show that the tire footprint approximates different shapes at different deformation stages,which depend on the relationship between the contact width and the tread width.Accordingly,a new tire footprint calculation model is proposed and the established tire force-deformation equation is modified.(3)Combining the established tire force equation with computer vision technology,a novel non-contact vehicle weight identification method is proposed.The tire image is collected through the vision system.Next,the tire vertical deflection,the specification parameters,and the inflation pressure value are extracted from the tire image by image processing algorithms.Finally,the established tire force-deformation equation is used to calculate the tire force,vehicle axle weight,and gross weight.(4)The feasibility of the proposed non-contact vehicle weight identification method and the accuracy of the established tire force equation are verified in the field test.Different passenger cars and trucks equipped with tires of different specifications are selected for investigation.The calculation results of tire force equations under different footprint assumptions are compared.The results show that none of the tire force equations established based on the traditional footprint assumption can accurately calculate the axle load and gross weight of the vehicle.In contrast,the improved equation shows better results and the errors are all within 10%.