Abstract: A method for predicting mass of heavy vehicles based on networked operating data and machine learning includes: collecting operating data; extracting speed, engine output torque, satellite elevation, and gear position under a driving condition; determining a transmission ratio of the heavy vehicle based on the gear position; filtering the speed, engine output torque, and the satellite elevation using three of the plurality of filtering parameters; determining a filtered vehicle longitudinal acceleration under the driving condition using the filtered speed, and one of the plurality of filtering parameters; determining a filtered road gradient sine value under the driving condition using the filtered speed, satellite elevation, and one of the plurality of filtering parameters; and inputting the speed, engine output torque, vehicle longitudinal acceleration, road gradient sine value, and transmission ratio into a vehicle mass prediction model to obtain a predicted mass.

Abstract: A method for predicting mass of heavy vehicles based on networked operating data and machine learning includes: collecting operating data; extracting speed, engine output torque, satellite elevation, and gear position under a driving condition; determining a transmission ratio of the heavy vehicle based on the gear position; filtering the speed, engine output torque, and the satellite elevation using three of the plurality of filtering parameters; determining a filtered vehicle longitudinal acceleration under the driving condition using the filtered speed, and one of the plurality of filtering parameters; determining a filtered road gradient sine value under the driving condition using the filtered speed, satellite elevation, and one of the plurality of filtering parameters; and inputting the speed, engine output torque, vehicle longitudinal acceleration, road gradient sine value, and transmission ratio into a vehicle mass prediction model to obtain a predicted mass.