Pro-social control of connected automated vehicles in mixed-autonomy multi-lane highway traffic

We propose pro-social control strategies for connected automated vehicles (CAVs) to mitigate jamming waves in mixed-autonomy multi-lane traffic, resulting from car-following dynamics of human-driven (HDVs). Different existing studies, which focus mostly on ego vehicle objectives CAVs an individualistic manner, we devise a algorithm. The latter takes into account the (i.e., driving comfort and traffic efficiency) both surrounding HDVs improve smoothness entire observable traffic. Under model predictive (MPC) framework that uses acceleration lane change sequences as optimization variables, problem individualistic, altruistic, is formulated non-convex mixed-integer nonlinear program (MINLP) relaxed convex quadratic through converting piece-wise-linear constraints due optimal velocity with relative (OVRV) linear by introducing slack variables. Low-fidelity simulations using OVRV high-fidelity PTV VISSIM simulator show altruistic can provide significant performance gains over terms efficiency single- roads.