Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

GAMEOPT+: Improving Fuel Efficiency in Unregulated Heterogeneous Traffic Intersections via Optimal Multi-agent Cooperative Control

Nilesh Suriyarachchi, Rohan Chandra, Arya Anantula, John S. Baras, Dinesh Manocha

TL;DR

This work proposes GameOpt+, a novel hybrid approach for cooperative intersection control in dynamic, multi-lane, unsignalized intersections that combines an auction mechanism and an optimization-based trajectory planner that operates 100 times faster while ensuring fairness, safety, and efficiency.

Abstract

Better fuel efficiency leads to better financial security as well as a cleaner environment. We propose a novel approach for improving fuel efficiency in unstructured and unregulated traffic environments. Existing intelligent transportation solutions for improving fuel efficiency, however, apply only to traffic intersections with sparse traffic or traffic where drivers obey the regulations, or both. We propose GameOpt+, a novel hybrid approach for cooperative intersection control in dynamic, multi-lane, unsignalized intersections. GameOpt+ is a hybrid solution that combines an auction mechanism and an optimization-based trajectory planner. It generates a priority entrance sequence for each agent and computes velocity controls in real-time, taking less than 10 milliseconds even in high-density traffic with over 10,000 vehicles per hour. Compared to fully optimization-based methods, it operates 100 times faster while ensuring fairness, safety, and efficiency. Tested on the SUMO simulator, our algorithm improves throughput by at least 25%, reduces the time to reach the goal by at least 70%, and decreases fuel consumption by 50% compared to auction-based and signaled approaches using traffic lights and stop signs. GameOpt+ is also unaffected by unbalanced traffic inflows, whereas some of the other baselines encountered a decrease in performance in unbalanced traffic inflow environments.

GAMEOPT+: Improving Fuel Efficiency in Unregulated Heterogeneous Traffic Intersections via Optimal Multi-agent Cooperative Control

TL;DR

This work proposes GameOpt+, a novel hybrid approach for cooperative intersection control in dynamic, multi-lane, unsignalized intersections that combines an auction mechanism and an optimization-based trajectory planner that operates 100 times faster while ensuring fairness, safety, and efficiency.

Abstract

Better fuel efficiency leads to better financial security as well as a cleaner environment. We propose a novel approach for improving fuel efficiency in unstructured and unregulated traffic environments. Existing intelligent transportation solutions for improving fuel efficiency, however, apply only to traffic intersections with sparse traffic or traffic where drivers obey the regulations, or both. We propose GameOpt+, a novel hybrid approach for cooperative intersection control in dynamic, multi-lane, unsignalized intersections. GameOpt+ is a hybrid solution that combines an auction mechanism and an optimization-based trajectory planner. It generates a priority entrance sequence for each agent and computes velocity controls in real-time, taking less than 10 milliseconds even in high-density traffic with over 10,000 vehicles per hour. Compared to fully optimization-based methods, it operates 100 times faster while ensuring fairness, safety, and efficiency. Tested on the SUMO simulator, our algorithm improves throughput by at least 25%, reduces the time to reach the goal by at least 70%, and decreases fuel consumption by 50% compared to auction-based and signaled approaches using traffic lights and stop signs. GameOpt+ is also unaffected by unbalanced traffic inflows, whereas some of the other baselines encountered a decrease in performance in unbalanced traffic inflow environments.
Paper Structure (20 sections, 3 theorems, 23 equations, 10 figures)

This paper contains 20 sections, 3 theorems, 23 equations, 10 figures.

Table of Contents

  1. Introduction
  2. Related Work
  3. Solutions for Improving Fuel Efficiency
  4. Towards Navigating Unregulated Traffic Intersections
  5. Main Contributions:
  6. Intersection Control Formulation
  7. Modeling the Physical Intersection
  8. Vehicle Dynamics and modeling
  9. Dynamic Traffic Intersection Planning & Control Problem
  10. Sponsored Search Auctions
  11. GameOpt+
  12. Game: Selecting a Priority Order, $q$
  13. Multi-lane Intersection Planning and Control
  14. Opt: Computing Optimal Velocities
  15. Experiments and Results
  16. ...and 5 more sections

Key Result

Theorem III.1

Incentive compatibility for dynamic intersections: For each agent $i$ for $i=1,2,\ldots,n$ and $n = n_1 + n_2 + n_3 + n_4$ where $n_j$ represents the number of vehicles on the $j^\textrm{th}$ arm, bidding $\bm{b}_i = \bm{\zeta}_i$ is the dominant strategy.

Figures (10)

  • Figure 1: GameOpt+: We present a new approach for improving fuel efficiency via optimal real-time planning and control in dynamic, heterogeneous, multi-agent unsignalized intersections. We show that at identical input traffic flow levels, our approach results in the most fuel efficient traffic flow.
  • Figure 2: Regions of interest: We highlight the control zone in which vehicles communicate with the control tower and the conflict zone in which vehicles cross over to their desired target road.
  • Figure 3: GameOpt+ overview: Our approach begins by reading in the positions and velocities of all agents in the control zone (blue). Our approach is hybrid; in the planning phase, GameOpt+ collect the bids from every agent and generate an optimal priority sequence (Section \ref{['subsec: priority_order']}). In the controls phase, we use an optimization-based trajectory planner to compute the optimal velocity for each agent that satisfies the priority order while simultaneously guaranteeing safety and real-time performance (Section \ref{['sec:optimization']}).
  • Figure 4: Conflict handling at multi-lane intersections. We propose a novel strategy for conflict handling using vehicle grouping based on non-colliding trajectories.
  • Figure 5: Performance in terms of throughput and time-to-goal in comparison to baselines, with increasing vehicle input flow rate.
  • ...and 5 more figures

Theorems & Definitions (6)

  • Theorem III.1
  • proof
  • Theorem III.2
  • proof
  • Theorem III.3
  • proof