The Lateral Tracking Control for the Intelligent Vehicle Based on Adaptive PID Neural Network

Area of Science:

• Robotics and Control Systems
• Automotive Engineering
• Artificial Intelligence

Background:

• Intelligent vehicles represent complex nonlinear systems.
• Path tracking control is crucial for autonomous navigation.
• Existing control models require enhanced real-time performance and robustness.

Purpose of the Study:

• To design a lateral control dynamic model for intelligent vehicles.
• To develop a robust path tracking controller for autonomous navigation.
• To provide a theoretical foundation for integrated vehicle control systems.

Main Methods:

• Established a vehicle dynamics model (transfer function) using vehicle parameters.
• Developed a second-order control system model based on the CARMA model.
• Employed forgetting factor recursive least square estimation (FFRLS) for system parameter identification.
• Designed a neural network PID controller for lateral path tracking.

Main Results:

• The proposed lateral control model demonstrated high real-time performance.
• The neural network PID controller exhibited significant robustness in path tracking simulations.
• The system successfully achieved accurate lateral path following.

Conclusions:

• The developed controller and model offer a strong theoretical basis for intelligent vehicle autonomous navigation.
• This research lays groundwork for future advancements in coupled vertical and lateral vehicle control.
• The findings highlight the potential of AI-driven control strategies in enhancing vehicle autonomy.