A Comparative Study of H∞ and Model Predictive Control Approach to Eco-Driving. Issue 20 (2021)

• Record Type: Journal Article
• Authors: Chen, Di; Kim, Youngki
• Is Part Of: IFAC-PapersOnLine. Volume 54:Issue 20(2021)
• Page Start: 889
• Page End: 894
• Publication Date: 2021
• Subjects: H∞ control -- adaptive cruise control -- eco-driving; Automatic control -- Periodicals; 629.805
• DOI: 10.1016/j.ifacol.2021.11.284
• Languages: English
• ISSNs: 2405-8963

Abstract

Abstract: This paper presents a simple but effective approach to designing an ecological adaptive cruise controller based on the H∞ technique. The controller is designed to minimize the H∞ norm of the transfer function from the internal states and the lead vehicle's velocity (disturbance) to the ego vehicle's acceleration to prompt driving smoothness. It implicitly handles position constraints via the minimization of H∞ norm of the transfer function from the lead vehicle's velocity as a disturbance to the time-headway tracking (performance). The proposed controller is beneficial as it is prediction and optimization-free and hence is lightweight for online implementation. The performance of the proposed controller is compared with a model predictive control (MPC)-based car-following approach, specifically, acceleration minimization, in consideration of electrical and hybrid powertrain operations. Simulation results show that despite its simple linear feedback structure, the H∞ controller can achieve good performance compared to the MPC controller in the presence of prediction inaccuracy for different types of electrified powertrains.