.png){kind=icon}
.png){kind=icon}
Return
A mass estimator based on hybrid model-data driven method for heavy-duty vehicles
.png){kind=icon}
PRE
AI
L
J
D
DOI：10.1177/09544070251337208.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
As a basic parameter, vehicle mass is closely related to economic and safety performance, etc. However, mass estimation faces challenges such as model accuracy and generalization ability. To solve these problems, a novel hybrid model-data driven vehicle mass estimation method is proposed. The model-driven method is used for calculating mass, and the data-driven method constructs a random forest-based data classifier to determine whether the data can be used; Then, by finding the relationship between data and classification results, the data structure is determined, based on the principle of increasing recall while ensuring a high precision; Finally, the proposed method is validated by monitoring data, the classification results have a high precision of more than 85%, and the error range of estimated mass is within +/- 5%. The excellent data classification and mass estimation capabilities enable mass monitoring for heavy-duty vehicles, which is important for global energy management, traffic safety management, etc.
Keywords:
Hybrid model-data driven method
precision
random forest-based classifier
vehicle mass estimation
Journal
P
IF:
1.5
Papers:
442
Citations:
0
Organization
Cited Papers