Adaptive path tracking control of unmanned agricultural machinery with fixed-time super-twisting sliding mode based...

Area of Science:

• Robotics and Automation
• Control Systems Engineering
• Agricultural Engineering

Background:

• Intelligent operation of unmanned agricultural machinery relies on accurate path tracking.
• Traditional methods struggle with oscillations and external disturbances, limiting performance.

Purpose of the Study:

• To propose a fixed-time super-twisting sliding mode adaptive path-tracking control for unmanned agricultural vehicles.
• To enhance robustness and adaptability in agricultural machine path tracking.

Main Methods:

• Utilized Regularized Least Squares Extreme Learning Machine (RLS-ELM) for improved adaptability.
• Designed a generalized terminal sliding mode surface incorporating lateral and heading deviations.
• Developed a Super-Twisting Sliding Mode control law with RLS-ELM for disturbance compensation.

Main Results:

• The proposed Extreme Learning Machine-Adaptive Fixed-Time Generalized Super-Twisting (ELM-AFGST) method reduced lateral mean absolute errors by 24.5% and 27.4%.
• Heading mean absolute errors were decreased by 5.4% and 30.8% compared to existing methods.
• Field experiments validated the superior performance of the ELM-AFGST method.

Conclusions:

• The developed path-tracking control offers a robust solution for unmanned agricultural vehicles.
• The method provides a solid theoretical framework for high-precision path tracking in agriculture.
• This advancement supports the intelligent operation and efficiency of autonomous agricultural machinery.