Layered Control Architectures for Complex Networked Systems

未来的嵌入式实时控制系统将越来越多地是无线的、分布式的、大规模的，并且将本质上是混合的，将离散或数字组件与连续时间非线性动力学相结合。这种网络系统的复杂性在通信、计算和控制的交汇处提出了新的挑战。在本提案中，我们将研究网络实时控制系统的设计和分析，处理复杂性，延迟，可靠性和其他问题。我们的目标是开发与应用无关的网络控制系统的控制策略，并使传感和驱动与通信和计算相结合。我们的研究结果的潜在应用包括自动驾驶汽车和自主运动系统、远程建筑、远程手术、在危险和远程环境中工作、监视、搜索和救援机器人、触觉设备和远程教育。我们的研究将集中于发展新的理论工具来分析和控制具有时变延迟的网络系统。在实施方面，我们将在内部设计和建造的系统上进行概念验证型实验，包括远程控制汽车和机器人网络。

Future embedded real time control systems will increasingly be wireless, distributed, large-scale, and will be inherently hybrid, combining discrete or digital components with continuous time nonlinear dynamics. The complexity of such networked systems presents new challenges that lie at the confluence of communication, computing, and control. In this proposal we will investigate the design and analysis of networked real-time control systems treating complexity, delays, reliability, and other issues. Our goal is to develop control strategies for networked control systems that are application independent and enable the convergence of sensing and actuation with communication and computing. Potential applications of our results include autonomous vehicles and autonomous locomotion systems, remote construction, remote surgery, work in hazardous and remote environments, surveillance, search and rescue robots, haptic devices, and distance education. Our research will focus on the development of new theoretical tools for analysis and control of networked systems with time-varying delays. On the implementation side, we will develop proof-of concept type experiments on in-house designed and built systems, including networks of remotely controllable cars and robots.