Data Games: A Game-Theoretic Approach to Swarm Robotic Data Collection

TL;DR

This paper introduces a game-theoretic cooperative data collection strategy for autonomous vehicle fleets to efficiently gather diverse training data, improving ML model robustness while reducing bandwidth and labeling costs.

Contribution

It formulates the data collection as a multi-player game, providing a strategy that converges to an optimal policy with minimal information and demonstrating theoretical and empirical performance gains.

Findings

Outperforms standard benchmarks by up to 21.9% on perception datasets.

Theoretical guarantees align with experimental results.

Effective in adverse weather conditions for autonomous driving.

Abstract

Fleets of networked autonomous vehicles (AVs) collect terabytes of sensory data, which is often transmitted to central servers (the ''cloud'') for training machine learning (ML) models. Ideally, these fleets should upload all their data, especially from rare operating contexts, in order to train robust ML models. However, this is infeasible due to prohibitive network bandwidth and data labeling costs. Instead, we propose a cooperative data sampling strategy where geo-distributed AVs collaborate to collect a diverse ML training dataset in the cloud. Since the AVs have a shared objective but minimal information about each other's local data distribution and perception model, we can naturally cast cooperative data collection as an $N$-player mathematical game. We show that our cooperative sampling strategy uses minimal information to converge to a centralized oracle policy with complete information about all AVs. Moreover, we theoretically characterize the performance benefits of our game-theoretic strategy compared to greedy sampling. Finally, we experimentally demonstrate that our method outperforms standard benchmarks by up to $21.9\%$ on 4 perception datasets, including for autonomous driving in adverse weather conditions. Crucially, our experimental results on real-world datasets closely align with our theoretical guarantees.