Two-layer structure algorithm for estimation of commercial vehicle mass

Abstract

Aiming at the problem of mass estimation for commercial vehicle, a two-layer structure mass estimation algorithm was proposed. The first layer was the grade estimation algorithm based on recursive least squares method and the second layer was a mass estimation algorithm using the extended Kalman filter. The estimated grade was introduced as the observation quantity of the second layer. The influence of the suspension deformation on grade estimation was considered in the first layer algorithm, which was corrected in real time according to the mass and road grade estimated by the second layer algorithm. The proposed estimation algorithm was validated via a co-simulation platform involving TruckSim and MATLAB/Simulink. Finally, a road test was carried out, and the evaluation method using the root mean square error was proposed. According to the test, the average value of the root mean square error reduces from 871.65 to 772.52, grade estimation is more accurate, and the convergence speed of mass estimation is faster, compared with estimation results of the extended Kalman filter method.

Truck mass estimation method based on the on-board sensor

“…However, the compression amount of the rear suspension will be affected by the road gradient, that is to say, the output voltage value of the angle sensor will be affected by the road gradient. To solve this problem, the mass gain coefficient is introduced to decompose the complex relationship in equation (24) into equations (25) and (26).…”

“…Equation (26) shows the relationship between the vehicle mass and the output voltage of the angle sensor under the condition that the road gradient is 0. Then, on the road whose road gradient is not 0, the actual mass of the automobile should be equal to the mass of the automobile calculated on the road whose road gradient is 0 multiplied by the mass gain coefficient, as shown in equation (27)…”

“…Based on the LS method, Rhode and Gauterin 25 used a total least squares (TLS) with increments to estimate the vehicle mass. Cai et al 26 proposed a two-layer structure algorithm combining RLS and EKF: decoupling the road gradient from the car mass, using the first layer to estimate the road gradient, and the second layer algorithm to calculate the car mass by the road gradient. Li et al 27 developed a two-layer structure based adaptive estimation strategy to estimate vehicle mass and road gradient: this two-layer structure mitigates the coupling effect between fast changed parameters (slip ratio, friction coefficient, etc.)…”

A Fusion Algorithm for Estimating Time-Independent/-Dependent Parameters and States

“…A Kalman filter (KF) is also commonly used [14][15][16][17][18] as a special case of the recursive Bayesian estimation algorithm with assumptions of linear state-space representation and Gaussian probability distribution. However, both RLS and the KF are limited for linear systems, and hence, KF variants, such as extended KF [19][20][21][22][23] or unscented KF [24][25][26], are more suitable for nonlinear estimation. The extended KF usually provides "first-order" approximations to the optimal terms.…”

Intelligent Two-Step Estimation Approach for Vehicle Mass and Road Grade

“…Chu et al [25] proposed an estimator based on a combined kinematic and dynamic model to eliminate the influence of different frequency noise. Furthermore, two-layer estimation algorithms were proposed to alleviate the coupling effect between the two parameters and improve the computational efficiency [26]- [29]. In the first layer, the mass or grade was estimated, and it was taken as a known parameter in the second layer to estimate the other parameter.…”