Adhesion Coefficient Identification of Wheeled Mobile Robot under Unstructured Pavement

Area of Science:

• Automotive Engineering
• Road Surface Analysis
• Control Systems

Background:

• Unstructured pavement poses significant challenges for accurately determining road adhesion coefficients due to uneven slopes.
• Existing methods struggle with the complexities of variable road surfaces, impacting vehicle dynamics and safety.

Purpose of the Study:

• To develop an improved estimation method for the peak adhesion coefficient of unstructured pavement.
• To enhance the accuracy and reliability of adhesion coefficient identification under challenging road conditions.

Main Methods:

• An extended Kalman filter (EKF) approach is employed for real-time estimation.
• An equivalent suspension model is introduced to optimize vertical wheel load calculations.
• Vehicle acceleration and posture data are modified and integrated for improved accuracy.

Main Results:

• The proposed method significantly improves the identification accuracy of road adhesion coefficients.
• Simulation experiments demonstrate at least a 3.6% improvement in estimation accuracy.
• The algorithm's precision and effectiveness were validated through multi-condition simulations using Carsim.

Conclusions:

• The developed extended Kalman filter-based method effectively estimates the peak adhesion coefficient of unstructured pavement.
• The integration of an equivalent suspension model and modified vehicle data enhances estimation precision.
• This algorithm offers a robust solution for improving vehicle control and safety on complex road surfaces.