A Variable-Sampling Time Model Predictive Control Algorithm for Improving Path-Tracking Performance of a Vehicle.

Area of Science:

• Robotics and Control Systems
• Autonomous Vehicle Navigation
• Automotive Engineering

Background:

• Model Predictive Control (MPC) is widely used for autonomous vehicle path tracking.
• Existing MPC algorithms can suffer from path tracking errors, especially in complex driving scenarios.
• Optimizing control input and sampling time is crucial for enhancing MPC performance.

Purpose of the Study:

• To propose a novel MPC algorithm that improves path tracking performance.
• To reduce path tracking errors by adaptively updating the sampling time based on control inputs.
• To compare the performance of the proposed MPC algorithm against existing methods.

Main Methods:

• Developed a novel MPC algorithm that adjusts sampling time based on lateral velocity and front steering angle.
• Constructed simulation scenarios involving straight and curved driving paths using MATLAB's Automated Driving Toolbox.
• Calculated optimal control inputs for each step of the MPC algorithm.
• Compared the path-following performance and computation times of the proposed and existing MPC algorithms via simulation.

Main Results:

• The proposed MPC algorithm demonstrated superior path-following performance compared to the existing MPC algorithm.
• Simulations confirmed reduced path tracking errors with the novel MPC approach.
• The algorithm's effectiveness was validated across varied driving path scenarios.

Conclusions:

• The novel MPC algorithm effectively enhances path tracking performance in autonomous driving systems.
• Adaptive sampling time adjustment based on control inputs is a key factor in reducing tracking errors.
• The proposed method offers a promising advancement for more accurate and reliable autonomous vehicle control.