Conceptual

Additive Feedback Control of Automated Vehicles for Smoothing Mixed Traffic Flow

A class of additive feedback controllers lets a minority of automated vehicles damp backward-propagating stop-and-go waves in mixed human/automated traffic using only locally sensed inter-vehicle spacing and relative speed, with no assumption of equilibrium traffic speed and no vehicle-to-vehicle communication. The controlled vehicle tracks a smoothed virtual version of its predecessor's speed with finite-time convergence and car-following safety, and an iterative optimization procedure selects the control parameters that best reduce traffic oscillations and fuel consumption.