14 September 2015 in IEEE Transactions on Automatic Control
This note presents a design technique for the delay-based controller called proportional integral retarded (PIR), which solves the regulation problem of a general class of stable second-order LTI systems. Using spectral analysis, the technique yields a tuning strategy for the PIR by placing a triple real dominant root for the closed-loop system. This result ultimately guarantees a desired exponential decay rate σ d while achieving the PIR tuning as an explicit function of σ d and system parameters.
3 January 2017 in SIAM Journal on Control and Optimization
This paper explores the deliberate introduction of delays in feedback loops and their positive impact on linear time-invariant, single-input, single-output (LTI-SISO) systems. The delay, considered...
1 September 2015 in ISA Transactions
16 February 2024 in IEEE Transactions on Systems, Man, and Cybernetics: Systems
This article considers the containment of heterogeneous linear time-invariant multi-human multi-agent systems with a distributed dynamic state feedback control law. The unique aspect of this problem lies in the presence of multiple human operators. Each operator commands a subset of the agents while observing the same or a different subset. Commanding based on their observation approach of the human operators makes them integral components of the network rather than simple providers of reference signals. This setup requires a new set of conditions distinct from the standard containment problem literature. Considering fixed-time delays in human reactions and information exchange between human operators and agents, it is shown that the containment problem is solved if the delayed closed-loop system is exponentially stable under the new set of conditions. Encompassing different graphs, system scales, and human parameters, we evaluate the largest delay without losing the closed-loop stability under various scenarios through numerical examples. The results highlight the potential for effective coordination of a large fleet of autonomous vehicles facilitated by a limited number of human operators.
21 August 2018 in IEEE Transactions on Automatic Control
A distributed Proportional-Retarded (PR) protocol is proposed for a class of large-scale consensus system with single-integrator agent dynamics. The protocol, a part of which utilizes intentional delays, enables fast consensus mainly by placing system' dominant poles and separating the rest of the spectrum from these poles as desired. The PR protocol rests upon spectral analysis and salient features of Lambert W functions, and ultimately helps compose the graph Laplacian of the network as a complete graph.
21 December 2018 in International Journal of Robust and Nonlinear Control