构建可扩展的自动驾驶车队测试平台，验证协同驾驶在复杂环境下的表现。
ConvoyNext: A Scalable Testbed Platform for Cooperative Autonomous Vehicle Systems
- 设计模块化可扩展平台，支持异构车辆协同测试
- 在多种通信条件下验证车队行为，包括故意丢包场景
- 开源平台助力交通效率与安全研究，适合智能网联车团队使用
协同自动驾驶系统的发展依赖于多智能体间的有效协作，旨在提升交通效率、燃油经济性和道路安全性。尽管前景广阔，真实世界中对这类系统的测试仍面临重大挑战，且对控制策略、轨迹建模方法及通信鲁棒性的验证至关重要。为此，我们提出ConvoyNext——一个可扩展、模块化且可扩展的平台，专为真实世界中协同驾驶行为的评估而设计。通过一系列涉及自动驾驶车队在复杂轨迹中行驶的实验，我们展示了ConvoyNext在异构车辆配置下的稳健性，以及在不同通信条件（包括故意丢包）下评估车队行为的有效性。结果表明，ConvoyNext是一个全面的开放获取测试平台，有助于推动协同自动驾驶系统的研究。
原文摘要 · Abstract (English)
The advancement of cooperative autonomous vehicle systems depends heavily on effective coordination between multiple agents, aiming to enhance traffic efficiency, fuel economy, and road safety. Despite these potential benefits, real-world testing of such systems remains a major challenge and is essential for validating control strategies, trajectory modeling methods, and communication robustness across diverse environments. To address this need, we introduce ConvoyNext, a scalable, modular, and extensible platform tailored for the real-world evaluation of cooperative driving behaviors. We demonstrate the capabilities of ConvoyNext through a series of experiments involving convoys of autonomous vehicles navigating complex trajectories. These tests highlight the platform's robustness across heterogeneous vehicle configurations and its effectiveness in assessing convoy behavior under varying communication conditions, including intentional packet loss. Our results validate ConvoyNext as a comprehensive, open-access testbed for advancing research in cooperative autonomous vehicle systems.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。