Abstract The rapid transition towards electromobility urges a concurrent focus on safety and intelligent vehicle control. In this regard, Advanced Driver Assistance Systems (ADAS) are paramount, playing a critical role in mitigating human error and enhancing road and passenger safety. However, challenges remain in the robust formulation and integration of such control systems, particularly considering the difficulty in modeling sophisticated architectures and the power synergy paths of hybrid/electric drivelines. This paper presents a comprehensive, novel methodology that utilizes a single-platform solution for model parameters tuning and online optimization of the control layers within an Adaptive Cruise Control (ACC) system for Electric Vehicles (EVs). To this aim, an intelligent Model Predictive Control (MPC) is developed, based on decentralized control modes for cruising, spacing, and braking. A unified prediction model is implemented to provide look-ahead estimation of the driving situation based on real-time measurements. The efficacy of the model prediction and control mode swapping was investigated through experimental testing of a real EV on a chassis dynamometer, with an emulated lead vehicle detected by an on-board LiDAR sensor. The single-platform, featuring updated model parameters and optimized control gains, demonstrated an ability to maintain speed-tracing and precise spacing when exposed to different disruptive scenarios. The per-mode tracking accuracy achieved 98% in cruise control, 87.8% in spacing control, and 55.0% in braking mode under coasting-only constraints. The proposed work thus offers a significant, unified solution to handle the complex challenges of driveline modeling and control system design, mitigating computational and technical difficulties.
Sharkawy et al. (Fri,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: