Delay-compensating cooperative adaptive cruise control of vehicular platoons utilizing multiple-predecessor-following communication topology

Cooperative Adaptive Cruise Control (CACC) and its evolution in recent research constitute one of the primary reasons that intelligent transportation systems have seen such great improvements over the last few years. Particularly, this control methodology aims to improve safety and efficiency while simultaneously guaranteeing string stability of the platoon of vehicles. Information exchange between vehicles that use CACC creates a state-of-the-art control strategy that outperforms classical Adaptive Cruise Control (ACC) methods. As a result, traffic oscillations are eliminated, and disturbance propagation in the platoon vanishes. On the other hand, the practical implementation of such controllers also comes with some difficulties. One of these includes actuation delays, where the control signal is not imposed instantly, while others stem from the heterogeneity of the platoons. These challenges can significantly reduce performance and even degrade individual vehicle stability and string stability. Thus, a predictor-based CACC law is introduced for heterogeneous platoons under constant delays. The main goal is to achieve both individual vehicle stability and string stability under these conditions. This work considers a platoon model where each vehicle is inter-connected utilizing the Multiple-Predecessor-Following (MPF) communication topology. The proposed controller is based on a nominal delay-free CACC design. Continuing, we build upon this framework and incorporate a predictor-based mechanism to compensate for the presence of actuation delays. This controller provides feedback on the spacing, speed, and acceleration errors in order to achieve disturbance attenuation and improved performance. At the same time, an analysis regarding the control parameters is provided in order to highlight the importance of the design parameters, such as the time headway and the number of predecessor vehicles communication. A comprehensive analysis of stability and string stability is presented. This analysis is based on transfer function representations with respect to the speed errors. Upon this framework, strong emphasis on heterogeneity and delay-compensation is given, while also sufficient conditions for string stability are also generated. The results show that the use of a predictor-based CACC law effectively eliminates input delays while simultaneously significantly reducing time headway. Finally, numerical simulations are presented, with the goal to validate the theoretical analysis. The experiments show how delays, heterogeneity and information topology affect the stability of the system. They also show how the predictor-based CACC methodology compensates for the delays while improving performance.