Cloud robotics can largely enhance the robot intelligence by offloading tasks to the cloud dynamically. However, the robots differ in their own hardware configuration such as battery and processing capacity, while the transmission frames are also a mixture of different quality of service (QoS) requirements. As the competition for limited channel resource is inevitable, how to optimize the system performance by effective resource allocation is a key problem. The paper proposes a two-tier hierarchical-based MAC (Two-Tier MAC) which means the classification exists not only in frames but also in robots. The Lyapunov optimization technique is used to maximize the time-averaged quality satisfaction. The experiments show the superior performance of the Two-Tier MAC compared with other MAC protocols especially in overloaded networks. Meanwhile, the system also presents a longer lifetime because the Two-Tier MAC takes energy balance into consideration.

中文翻译：

---

云机器人中基于两层分层MAC的Lyapunov优化

通过将任务动态卸载到云中，云机器人技术可以大大增强机器人的智能。但是，机器人在其自身的硬件配置（例如电池和处理能力）方面有所不同，而传输帧也混合了不同的服务质量（QoS）要求。由于不可避免的竞争有限的信道资源，如何通过有效的资源分配来优化系统性能是一个关键问题。本文提出了一种基于两层的基于层次的MAC（两层MAC），这意味着分类不仅存在于帧中，而且还存在于机器人中。Lyapunov优化技术用于最大化时间平均质量满意度。实验表明，与其他MAC协议相比，两层MAC的性能更高，尤其是在过载网络中。与此同时，