Decentralized Vehicle Coordination: The Berkeley DeepDrive Drone Dataset

2022-09-19

arXiv.org

AI Summary

In unstructured roads lacking explicit right-of-way rules—common in densely populated developing countries—autonomous driving faces challenges in decentralized, cooperative motion planning. To address this, we introduce the first open-source aerial video dataset comprising 20 representative traffic scenarios, and propose a consensus-based priority emergence modeling framework that uncovers fundamental principles governing dynamic priority formation in human social driving without explicit signaling. Our framework integrates vehicle detection, trajectory estimation, and multi-agent behavioral modeling to achieve interpretable, data-driven simulation of yielding behaviors in real-world traffic flows. Experiments demonstrate that our approach significantly outperforms conventional rule-based planners in both collision avoidance rate and traffic throughput. This work establishes the first reproducible, scalable empirical foundation and modeling paradigm for socially aware autonomous driving.

Abstract

A significant portion of roads, particularly in densely populated developing countries, lacks explicitly defined right-of-way rules. These understructured roads pose substantial challenges for autonomous vehicle motion planning, where efficient and safe navigation relies on understanding decentralized human coordination for collision avoidance. This coordination, often termed"social driving etiquette,"remains underexplored due to limited open-source empirical data and suitable modeling frameworks. In this paper, we present a novel dataset and modeling framework designed to study motion planning in these understructured environments. The dataset includes 20 aerial videos of representative scenarios, an image dataset for training vehicle detection models, and a development kit for vehicle trajectory estimation. We demonstrate that a consensus-based modeling approach can effectively explain the emergence of priority orders observed in our dataset, and is therefore a viable framework for decentralized collision avoidance planning.

Problem

• Lack of right-of-way rules in understructured roads
• Challenges in autonomous vehicle motion planning
• Underexplored social driving etiquette for collision avoidance

Innovation

• Consensus-based modeling for decentralized coordination
• Aerial video dataset for social driving analysis
• Development kit for vehicle trajectory estimation