Model Predictive Control for Speed-Dependent Active Suspension System with Road Preview Information

Area of Science:

• Automotive Engineering
• Control Systems Engineering
• Mechatronics

Background:

• Active suspension systems are crucial for vehicle stability and ride comfort.
• Traditional controllers struggle with speed-dependent dynamics and time delays.
• Model predictive control (MPC) offers a framework for advanced control strategies.

Purpose of the Study:

• To develop an enhanced model predictive control (MPC) scheme for speed-dependent active suspensions.
• To improve damping control performance under varying vehicle speeds.
• To enable simplified online implementation of complex control laws.

Main Methods:

• A half-car model was augmented with speed-dependent front- and rear-wheel time delays using Padé approximation.
• Linear parameter-varying (LPV) techniques were applied to handle time-varying parameters.
• An adaptive Kalman filter was used for state estimation with sensor noise.
• Explicit control laws were derived offline using multi-parameter linear programming (mp-LP).

Main Results:

• The proposed MPC scheme significantly improved active suspension damping control.
• Performance gains were demonstrated across a range of vehicle speeds.
• The controller effectively managed speed-dependent dynamics and time delays.
• Online implementation was simplified through lookup table-based solution searching.

Conclusions:

• The LPV-based MPC offers a robust and effective solution for active suspension control.
• The method provides superior performance compared to passive control strategies.
• This approach enhances vehicle dynamics control under dynamic operating conditions.