Mass estimation of ground vehicles based on longitudinal dynamics using IMU and CAN-bus data

2013 ◽

Cited By ~ 3

Author(s):

Mostafa Salama ◽

Vladimir V. Vantsevich

Keyword(s):

Rolling Resistance ◽

Computational Time ◽

Unmanned Ground Vehicle ◽

Ground Vehicles ◽

Power Losses ◽

Ground Vehicle ◽

Longitudinal Dynamics ◽

Straight Line ◽

On Line ◽

Terrain Characteristics

Studies of the tire-terrain interaction have mostly been completed on vehicles with steered wheels, but not much work has been done regarding skid-steered Unmanned Ground Vehicles (UGV). This paper introduces a mathematical model of normal and longitudinal dynamics of a UGV with four skid-steered pneumatic tire wheels. Unlike the common approach, in which two wheels at each side are treated as one wheel (i.e., having the same rotational speeds), all four wheels in this study are independently driven. Thus the interaction of each tire with deformable terrain is introduced as holonomic constraints. The stress-strain characteristics for tire-soil interaction are analyzed based on modern Terramechanics methods and then further used to determine the circumferential wheel forces of the four tires. Contributions of three components of each tire circumferential force to tire slippages are modeled and analyzed when the tire normal loads vary during vehicle straight-line motion. The considered tire-soil characteristics are mathematically reduced to a form that allows condensing the computational time for on-line computing tire-terrain characteristics. Additionally, rolling resistance of the tires is analyzed and incorporated in the UGV dynamic equations. Moreover, the paper describes the physics of slip power losses in the tire-soil interaction of the four tires and applies it to small skid-steered UGV. This study also formulates an optimization problem of the minimization of the power losses in the tire-soil interactions due to the tire slippage.

Download Full-text

The Method of Mass Estimation Considering System Error in Vehicle Longitudinal Dynamics

Energies ◽

10.3390/en12010052 ◽

2018 ◽

Vol 12 (1) ◽

pp. 52 ◽

Cited By ~ 8

Author(s):

Nan Lin ◽

Changfu Zong ◽

Shuming Shi

Keyword(s):

Colored Noise ◽

Estimation Method ◽

Recursive Least Squares ◽

Cruise Control ◽

Unknown Parameters ◽

System Error ◽

Mass Estimation ◽

Longitudinal Dynamics ◽

Mass Error ◽

Vehicle Mass

Vehicle mass is a critical parameter for economic cruise control. With the development of active control, vehicle mass estimation in real-time situations is becoming notably important. Normal state estimators regard system error as white noise, but many sources of error, such as the accuracy of measured parameters, environment and vehicle motion state, cause system error to become colored noise. This paper presents a mass estimation method that considers system error as colored noise. The system error is considered an unknown parameter that must be estimated. The recursive least squares algorithm with two unknown parameters is used to estimate both vehicle mass and system error. The system error of longitudinal dynamics is analyzed in both qualitative and quantitative aspects. The road tests indicate that the percentage of mass error is 16%, and, if the system error is considered, the percentage of mass error is 7.2%. The precision of mass estimation improves by 8.8%. The accuracy and stability of mass estimation obviously improves with the consideration of system error. The proposed model can offer online mass estimation for intelligent vehicle, especially for heavy-duty vehicle (HDV).

Download Full-text