Search

This paper provides a subspace method for closed-loop identification, which clearly specifies the model order from noisy measurement data. The method can handle long I/O data of the target system to be noise-tolerant and determine the model order via nuclear norm minimization. First, the proposed method compresses the long data by projecting them to an appropriate low dimensional subspace, then obtains a low order model whose order is specified by a combination of data compression and nuclear norm minimization. Its effectiveness is demonstrated through detailed numerical examples.

In this paper, we propose a simple framework for closed-loop system identification: the stabilized output error method. Traditional closed-loop system identification methods rely on the linearity of the target system, and they require the identification of noise models or prior knowledge or identifiability of the feedback controllers to obtain unbiased estimates. But in many real-world applications, the noise dynamics and feedback controllers are complex and difficult to identify, and the nonlinearity may not be ignored. The proposed framework introduces a virtual controller that stabilizes the error between model prediction and the output of the target system. This enables us to apply the output error method, which gives unbiased estimates without depending on the noise model and is applicable to a wide range of models, including nonlinear systems, to closed-loop system identification problems. The paper describes the framework and gives the theoretical support and design guidelines for virtual controllers. Through numerical examples, we show the effectiveness of the proposed framework in various situations, which include identification of (a) linear gray box models, (b) systems in the presence of disturbances having realistic complexity, and (c) a nonlinear unstable system in a human-in-the-loop environment.

This paper presents networked controllers for the coordination of multi-robot systems, inspired by the chemotaxis of bacteria. Chemotaxis is a biological phenomenon wherein each organism senses the concentration of a chemical in its environment and moves to the highest (or lowest) concentration point. The problem studied herein is a coverage problem, specifically, the problem of finding networked controllers to deploy robots so that they are located uniformly on a given space. To solve this problem, we decompose a global performance index quantifying the achieved degree of coverage into local indices that can be calculated in a distributed manner over the network of robots. By combining this with a controller causing chemotaxis, we present a solution to the coverage problem wherein each robot performs either a forward movement or random rotation based on the local performance index at each time step. Moreover, we extend this solution to rendezvous at an unspecified point. Simulation and experimental results demonstrate that our solution achieves coverage and rendezvous only via the above two types of robot movements and can handle different tasks simply by changing the global performance index, through the appropriate use of the chemotaxis controller.

This study attempts to analyze and design multi-agent systems in the spatial frequency domain and demonstrates that the spatial frequency-based approach is useful for distributed spatial filtering in sensor networks. First, we take the consensus of multi-agent systems (i.e., letting the states of all agents converge to an identical value) as an example and analyze it using the concept of spatial frequencies. We then show that consensus by typical controllers corresponds to lowpass filtering in the spatial frequency domain. This demonstrates that spatial frequencies can characterize the behavior of multi-agent systems. Second, we present a controller design method in the spatial frequency domain. The designed controllers provide the feedback system with a desired spatial frequency characteristic given in advance. We further derive a sufficient condition for the spatial frequency characteristic to ensure that the designed controllers are distributed. Finally, the effectiveness and applicability of our design method are demonstrated through an example of distributed denoising in a sensor network.

This paper proposes an input design method for identification of linear-parameter-varying (LPV) systems. In particular, this paper focuses on the Bayesian estimation of LPV systems, especially the impulse responses of LPV systems at the scheduling variable of interest. The mutual information is employed as a criterion, and a concrete procedure to obtain the local optimum is given. A numerical example is shown to demonstrate the effectiveness of the proposed input design.

This paper provides a model-free approach based on the Multi-Resolution Simultaneous Perturbation Stochastic Approximation (MR-SPSA) for maximizing power production of wind farms. The main advantage is that the method based on MR-SPSA can achieve fast controller tuning without any plant model by exploiting the information of the wind farm configuration such as turbines location and wind direction. In order to simulate the performance of the model-free scheme, a wind farm model with dynamic characterization of wake interaction between turbines is used and then the proposed method is applied to the Horns Rev wind farm. Simulation results illustrate that the method based on MR-SPSA achieves the maximum total power production with faster convergence compared with other existing model-free methods.