Cooperative driving, enabled by communication between automated vehicle systems, promises significant benefits to fuel efficiency, road capacity, and safety over single-vehicle driver assistance systems such as adaptive cruise control (ACC). However, the responsible development and implementation of these algorithms poses substantial challenges due to the need for extensive real-world testing. We address this issue and introduce OpenConvoy, an open and extensible framework designed for the implementation and assessment of cooperative driving policies on physical connected and autonomous vehicles (CAVs). We demonstrate the capabilities of OpenConvoy through a series of experiments on a convoy of multi-scale vehicles controlled by Platooning to show the stability of our system across vehicle configurations and its ability to effectively measure convoy cohesion across driving scenarios including varying degrees of communication loss.

翻译：协同驾驶通过自动化车辆系统间的通信实现，相较于自适应巡航控制等单车驾驶辅助系统，有望在燃油效率、道路通行能力和安全性方面带来显著提升。然而，由于需要进行大量的真实世界测试，这些算法的负责任开发与实施带来了重大挑战。针对此问题，我们提出了OpenConvoy，这是一个开放且可扩展的框架，专为在物理连接的自动驾驶车辆上实现和评估协同驾驶策略而设计。我们通过一系列实验展示了OpenConvoy的能力：实验使用由编队行驶控制的多尺度车辆组成的车队，证明了我们系统在不同车辆配置下的稳定性，以及其在包括不同程度通信中断在内的多种驾驶场景中有效测量车队凝聚力的能力。