Speed Advice for Connected Vehicles at an Isolated Signalized Intersection in a Mixed Traffic Flow Considering Stochasticity of Human Driven Vehicles

This paper aims to design speed advisory profile for connected vehicles (CVs) at an isolated signalized intersection in a mixed traffic flow of CVs and human driven vehicles (HDVs). To consider the uncertainty of HDVs, we propose the concept of
$\alpha $
-percentile trajectory of HDVs. We calculate the advisory speed of a CV that follows a HDV, assuming that the HDV moves along its
$\alpha $
-percentile trajectory. We use dynamic programming to solve the optimization problem. Simulation experiments show that, roughly speaking, the benefits of fuel consumption and travel time would increase with the increase of inflow rate and market penetration rate (MPR) of CVs. The maximum benefits would be achieved at the smallest value of
$\alpha $
when the traffic flow is undersaturated. However, when the traffic flow is oversaturated, the maximum benefits would be achieved at intermediate value of
$\alpha $
provided the MPR of CVs is not large. With the further increase of the MPR of CVs, the benefits would become not so sensitive to
$\alpha $
in the range
$\alpha \le 40$
%, but significantly decrease with the increase of
$\alpha $
when
$\alpha >40$
%. Finally, the impact of driver heterogeneity, inflow heterogeneity, tracking error and delay has been investigated.