Section: New Results

Control Architecture for Adaptive and Cooperative Car-Following

Participants : Carlos Flores, Fawzi Nashashibi.

The general scope of this work deals with three open challenges in the state-of-the-art of cooperative car-following systems:

1) Deal with the impact of not only communication links delays, but also heterogeneity between vehicles' dynamics in the same string. This should be targeted ensuring the gap-regulation robustness without degrading the expected performance to keep car-following benefits (individual and string stability). In particular, when a heterogeneous string is formed, the differences between vehicles dynamics introduce disturbances in the closed loop system affecting the string stability. In [22] we presented an online Cooperative Adaptive Cruise Control (CACC) feedforward adaptation with a fractional-order feedback controller for stable heterogeneous strings of vehicles. Simulations demonstrate the advantages over conventional homogeneous structures as well as system’s capability to both enhance stability and guarantee string stability regardless the vehicles distribution.

2) Design a modular architecture that permits to introduce cooperative string driving in urban environments, where interaction with vulnerable road users is highly probable. In this context, a cooperative car-following/emergency braking system with prediction-based pedestrian avoidance capabilities using vehicle-to-vehicle and vehicle-to-pedestrian communication links has been proposed in [14] and validated with RITS platforms.

3) Further extend the benefits of Adaptive Cruise Control (ACC) and Cooperative Adaptive Cruise Control (CACC) applications on traffic flow and safety, having strict ${ℒ}_2$ string stability as a hard constraint, employing different calculus techniques for the control design task. A fractional-order-based control algorithm is employed to enhance the car-following and string stability performance for both ACC and CACC vehicle strings, including communication temporal delay effects has been presented in [15]. Simulation and real experiments have been conducted for validating the approach.

The aforementioned contributions have been developed in the framework of the VALET project ANR-15-CE22-0013. They have been also implemented in the vehicle platforms of RITS team, for the sake of validation and further demonstration of the final VALET system.

This scientific work can be found as well in the thesis manuscript of Carlos Flores entitled "Control Architecture for Adaptive and Cooperative Car-Following" (cf. [8]).