Abstract
This paper considers the motion coordination problem of autonomous vehicles in an intersection of a traffic network. The featured challenge is the design of an intersection traffic manager, in the form of a supervisory control algorithm, that regulates the motion of the autonomous vehicles in the intersection. We cast the multivehicle coordination task as an optimization problem, with a one-dimensional search-space. A model- and optimization-based heuristic method is employed to compute the control policy that results in the collision-free motion of the vehicles at the intersection and, at the same time, minimizes their delay. Our approach depends on a computation framework that makes the need for complex analytical derivations obsolete. A complete account of the computational complexity of the algorithm, parameterized by the configuration parameters of the problem, is provided. Extensive numerical simulations validate the applicability and performance of the proposed autonomous intersection traffic manager.
Highlights
In the last decade, significant progress has been made in integrating wireless communications, advanced sensing technologies, and computationally powerful embedded systems on-board vehicles and transportation systems. ese augmenting technologies are projected to mitigate existing problems of urban traffic such as congestion, safety, reliability, and sustainability [1]. e most notable challenge in road traffic management systems is the control of traffic in junction nodes of the transportation network where roads cross or merge such as intersections, roundabouts, and onramps [2, 3]
E emergent integrated networked systems of autonomous, computing, and sensing nodes are brought into play to automate the traffic management of intersections. e overarching objective of this trending research field is to improve the safety and efficiency performance of intersections, compared to traditional means of traffic control Journal of Robotics. e longterm vision is to have autonomous vehicles transporting in accident-free traffic systems. e featured module of automated intersections is the control algorithm, referred to as Autonomous Intersection Manager (AIM) [7, 8], that is responsible for coordinating the collision-free motion of the vehicles within the intersection and optimizes a performance metric of the traffic system
To facilitate the optimization search, we introduce a preparatory procedure that bounds the domain of the optimization problem and outputs a set of candidate solutions that correspond to trajectories with unreserved paths in the intersection region (IR). ese candidate solutions, which serve as a pool for initial conditions to the optimization procedure, improve the performance of the latter by initiating the search with partially feasible trajectories—in terms of safety in the IR—that may lead the search faster to an optimal solution
Summary
Significant progress has been made in integrating wireless communications, advanced sensing technologies, and computationally powerful embedded systems on-board vehicles and transportation systems. ese augmenting technologies are projected to mitigate existing problems of urban traffic such as congestion, safety, reliability, and sustainability [1]. e most notable challenge in road traffic management systems is the control of traffic in junction nodes of the transportation network where roads cross or merge such as intersections, roundabouts, and onramps [2, 3]. Traditional nonlinear systems theory is required for the design of the low-level feedback control laws that steer the vehicles based on their kinematic (or dynamic) configuration; elements from computational optimization and dynamic programming are needed to seek for an optimal control policy that minimizes the performance metrics of the optimization problem; and heuristic algorithms are requisite to efficiently search for an acceptable solution in the high-dimensional problem space. All these computational and analytical components need to be amalgamated into a single framework that will be computationally tractable and applicable to a computing system.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
