Cooperative Maneuver Planning for Mixed Traffic at Unsignalized Intersections Using Probabilistic Predictions

Abstract

Intersections are among the scenarios that are most crucial for efficiency and traffic flow on roads. Several approaches to traffic control at intersections exist, each with its own advantages and drawbacks. These days, wireless connections between road users, automated vehicles, and intelligent infrastructure enable new ways of coordinating traffic. However, the gradual deployment of those advanced technologies leads to a heterogeneous mixture of partially automated, connected, and legacy vehicles. Planning and coordinating maneuvers for this mixed traffic is a challenge and subject to current research, as it can achieve significant efficiency improvements in those scenarios. In this paper, we propose a new maneuver planning system for cooperative connected vehicles in mixed traffic at unsignalized intersections, which often occur in urban areas.Our system consists of a probabilistic multi-modal prediction based on a driver model and an efficient optimization algorithm to find the best maneuvers. We present the functionality of our approach and evaluate the impact on traffic efficiency using simulations of two different intersection layouts at various rates of cooperative vehicle penetration.

Automatic Intersection Management in Mixed Traffic Using Reinforcement Learning and Graph Neural Networks

“…Another metric of interest is the delay, which is defined in accordance with [13] for vehicle ν i as…”

“…The state of the art in AIM consists of a broad variety of approaches, many of which can be allocated to reservationbased systems [4]- [7], optimization algorithms [8]- [11], or tree-based methods [12], [13]. In the following, we discuss a selection of works in commonly used paradigms and refer the reader to surveys like [14] for a more extensive overview.…”

“…The authors demonstrate their model's ability to generate viable trajectories in an overtaking maneuver for varying levels of cooperation. In the context of AIM, decision trees enhanced by probabilistic predictions of scene evolvement can be used to find the crossing order, which yields the highest expected efficiency [13] in mixed traffic scenarios.…”

Cooperative Behavior Planning for Automated Driving Using Graph Neural Networks

“…A prevalent issue with optimization-based approaches is the unfavorable scaling of computational demand with increasing traffic density. In [6], a novel AIM scheme is presented, that is also capable of handling mixed traffic, i.e., simultaneous usage by automated and human-driven vehicles.…”

“…The authors are with the Institute of Measurement, Control and Microtechnology, Ulm University, D-89081 Ulm, Germany. 2 The authors are with the Robert Bosch GmbH, Corporate Research, D-71272 Renningen, Germany.…”