Sliding Mode Observer-based Nonlinear Predictive Control for Ecological Cruise Control System

This paper presents a perturbation observer-based controller to optimize the fuel consumption of an ecological cruise control (eco-cruise) system. In the proposed approach, system perturbations are treated as augmented state variables. A sliding mode observer, which utilizes real-time measurements of fuel consumption and longitudinal velocity, is then designed to estimate these states and refine the initial vehicle model. The observer's stability is guaranteed using the direct Lyapunov method. Its performance is experimentally validated through tests involving various dynamic maneuvers. Subsequently, a nonlinear model predictive controller (NMPC) is designed based on this enriched model. The NMPC tracks a desired velocity while optimizing fuel consumption and adhering to motor torque constraints. Co-simulation using CarSim demonstrates the efficacy of the proposed method. The results confirm the controller's high performance in the presence of model perturbations and control input limitations.

کلیدواژه ها:

Ecological cruise control ، Fuel consumption ، Sliding mode observer ، Nonlinear model predictive control