Adaptive Model Predictive Control for Achieving Lane Tracking

This paper introduces a solution for the steering control system, which plays a crucial role in constructing autonomous vehicles. This study investigates the application of Adaptive Model Predictive Control (AMPC) for lane tracking in autonomous driving. It utilizes a 3 Degree-of-Freedom bicycle model and incorporates dynamic adjustments through sensor fusion. The research emphasizes the systematic tuning of control weights and real-time adjustments to showcase the effectiveness of AMPC in various driving scenarios. The simulation results validate the accuracy, safety, and reliability of AMPC, and ongoing objectives involve further development and the evaluation of sensor fusion, particularly with depth cameras. The use of the Unreal Engine 3D simulation enhances the practicality of the approach, highlighting the potential for real-time implementation of AMPC in autonomous driving and lane following. Continuous advancements in control algorithms and sensor technologies are expected to contribute to the ongoing improvement in the performance of AMPC.