Vehicle’s Weight Estimation Using Smartphone’s Acceleration Data to Control Overloading

摘要

We propose an overloading control system with a novel method to estimate vehicle weight using the sensor data from a smartphone mounted on the vehicle. The conventional method based on fixed weigh stations has limited coverage, and is expensive to install and maintain. Our proposed system overcomes these limitations by using smartphones, which are portable and cheaper. A multiple linear regression model is created using vertical acceleration statistical features and loading status classification as explanatory variables to estimate the vehicle’s weight. A pilot experiment estimating a trolley’s weight was followed by an experiment estimating an actual vehicle’s weight to verify the feasibility of using our method. We achieved average error of 593 kg, which accounted for 5.89% of the true average vehicle’s weight.