Delay-sensitive Joint Optimal Control and Resource Management in Multi-loop Networked Control Systems

摘要: In the operation of networked control systems, where multiple processes share a resource-limited and time-varying cost-sensitive network, communication delay is inevitable primarily influenced by, first, systems deploying intermittent sensor sampling to reduce cost by restricting non-urgent transmissions, second, network performing resource management minimize excessive traffic eventually data loss. heterogeneous scenario, may tolerate only specific levels sensor-to-controller latency, sensitivities need be considered in design policies achieve desired performance guarantees. We propose cross-layer optimal co-design control, for an NCS consisting stochastic linear time-invariant which close their loops over shared network. Aligned with advanced technology, we assume that offers range latency-varying transmission services given prices. Local samplers decide either pay higher access low-latency channel, or sending state sample at reduced price. A manager residing data-link layer arbitrates channel re-allocates resources if link capacities are exceeded. The local closed-loop measured combination linear-quadratic Gaussian suitable cost, overall objective defined social all three policy makers. derive allocation under different awareness models, including constant parameters, show generally leads enhancement expense computational complexity.