An accurate trajectory tracking method for low-speed unmanned vehicles based on model predictive control.

Area of Science:

• Robotics and Control Systems
• Automotive Engineering
• Applied Mathematics

Background:

• Traditional model predictive control (MPC) for vehicle trajectory tracking often assumes simple road terrains, leading to reduced accuracy in real-world scenarios.
• Actual road conditions frequently involve varying curvatures, which current simplified models do not adequately address.
• Low tracking accuracy in low-speed vehicles can compromise safety and operational efficiency.

Purpose of the Study:

• To develop an advanced trajectory tracking method for low-speed vehicles that accounts for road curvature.
• To improve the adaptability and accuracy of model predictive control (MPC) in dynamic road environments.
• To enable robust trajectory tracking on roads with complex and time-varying curvatures.

Main Methods:

• Proposed a novel trajectory tracking approach utilizing model predictive control (MPC) with automatic switching between control types based on road curvature.
• Implemented linear model predictive control (LMPC) for segments with small road curvatures.
• Employed nonlinear model predictive control (NMPC), incorporating road curvature effects, for segments with large road curvatures and time-varying conditions.

Main Results:

• Simulation comparisons demonstrated significant improvements in trajectory tracking accuracy compared to basic MPC models across various road types and speeds.
• The proposed method effectively handles roads with both small and large curvatures, including time-varying ones.
• Real-time computational efficiency was maintained, ensuring practical applicability.

Conclusions:

• The developed trajectory tracking method, featuring intelligent switching between LMPC and NMPC based on road curvature, enhances accuracy on complex roads.
• This adaptive control strategy is suitable for tracking trajectories on arbitrarily complex road geometries.
• The findings pave the way for more reliable autonomous navigation and control systems in diverse driving conditions.