Terrestrial robots, i.e., unmanned ground vehicles (UGVs), and aerial robots, i.e., unmanned aerial vehicles (UAVs), operate in separate spaces. To exploit their complementary features (e.g., fields of views, communication links, computing capabilities), a promising paradigm termed integrated robotics network emerges, which provides communications for cooperative UAVs-UGVs applications. However, how to efficiently deploy UAVs and schedule the UAVs-UGVs connections according to different UGV tasks become challenging. In this paper, we propose a sum-rate maximization problem, where UGVs plan their trajectories autonomously and are dynamically associated with UAVs according to their planned trajectories. Although the problem is a NP-hard mixed integer program, a fast polynomial time algorithm using alternating gradient descent and penalty-based binary relaxation, is devised. Simulation results demonstrate the effectiveness of the proposed algorithm.

翻译：地面机器人（即无人地面车辆，UGVs）与空中机器人（即无人飞行器，UAVs）运行于不同的空间领域。为充分利用两者的互补特性（如视场、通信链路及计算能力），一种称为"集成机器人网络"的前沿范式应运而生，可为合作型无人机-无人车应用提供通信支持。然而，如何根据不同的无人车任务高效部署无人机并调度无人机-无人车连接链路成为挑战。本文提出以和速率最大化为目标的问题模型，其中无人车自主规划轨迹，并根据规划轨迹动态关联无人机。尽管该问题属于NP难混合整数规划问题，我们设计了一种采用交替梯度下降与基于惩罚的二元松弛优化的快速多项式时间算法。仿真结果验证了所提算法的有效性。