Fast Trajectory Tracking Control Algorithm for Autonomous Vehicles Based on the Alternating Direction Multiplier...

Area of Science:

• Robotics and Control Systems
• Automotive Engineering
• Computational Optimization

Background:

• Real-time trajectory tracking is crucial for autonomous vehicle safety and efficiency.
• Model Predictive Control (MPC) is a common control strategy, but its computational demands can limit real-time performance.
• Existing methods like Active Set Method (ASM) and Interior Point Method (IPM) face challenges with computational speed.

Purpose of the Study:

• To improve the real-time performance of autonomous vehicle trajectory tracking.
• To enhance the computational speed of Model Predictive Control (MPC) algorithms.
• To evaluate the effectiveness of the Alternating Direction Multiplier Method (ADMM) in optimizing MPC for vehicle control.

Main Methods:

• Applied the Alternating Direction Multiplier Method (ADMM) to the receding optimization of Model Predictive Control (MPC).
• Formulated the vehicle dynamics model and output equation, introducing auxiliary and dual variables.
• Transformed the quadratic programming problem and vehicle dynamics constraints into an ADMM- solvable form, utilizing a decreasing penalty factor.

Main Results:

• The proposed ADMM-based MPC algorithm demonstrated improved trajectory tracking accuracy.
• The method achieved superior real-time performance compared to Active Set Method (ASM) and Interior Point Method (IPM).
• Computational time remained stable even with increased prediction time domains, confirming ADMM's effectiveness.

Conclusions:

• The ADMM approach effectively enhances the real-time computational speed of MPC for autonomous vehicle trajectory tracking.
• The proposed method offers a viable solution for achieving both high accuracy and real-time performance in autonomous driving systems.
• ADMM integration presents a significant advancement in optimizing control algorithms for dynamic vehicle applications.