Context-aware Nonlinear MPC for Automated Vehicles Embedding Newell's Car-Following Model

Automated Driving Systems equipping Connected and Automated Vehicles (CAVs) require a meticulous design to maximize the advantages of connectivity and automation. However, even the most widespread systems are not free from problems; for instance, the effectiveness of current Adaptive Cruise Control implementations in improving safety, traffic stability and energy consumption is questionable. Therefore, there is the need to develop solutions that avoid inadequate driving behaviours. To this end, in this work we design a Nonlinear Model Predictive Control (NMPC) strategy that resemble an ACC, embedding the Newell non-linear car-following model. The Newell model computes a reference speed profile over the prediction horizon, ensure string stability and anticipate speed oscillations. Then, using this speed as reference, the NMPC controller computes an optimal acceleration profile to drive the vehicle as fast as possible while ensuring comfort and safety. The proposed NMPC strategy is evaluated across various simulation scenarios, with several Key Performance Indicators (KPIs) related to safety, driving volatility and consumption. A conventional ACC system is employed for comparison. Through observation and comparison of the KPIs, the effectiveness of the proposed approach is demonstrated.