This paper explores the design of a composite prescribed-time controller for two-time-scale systems (TTSSs). First, due to numerical issues caused by the small perturbation parameter, standard control design technique...
详细信息
This paper explores the design of a composite prescribed-time controller for two-time-scale systems (TTSSs). First, due to numerical issues caused by the small perturbation parameter, standard control design techniques for prescribed-time stabilization of regular linear systems are typically unsuitable for TTSSs. To address this, we use time-scale separation to decouple the system into slow and fast subsystems. To avoid strict requirements on the input matrix, we apply a state transformation to convert the state equations of the slow and fast subsystems into block forms. Next, to prevent input saturation and achieve a directly and arbitrarily adjustable upper bound on the convergence time, we use time-varying scaling functions to design prescribed-time controllers for both subsystems. The composite prescribed-time controller ensures the prescribed-time convergence of the TTSS. Finally, two comparative examples and one RC ladder circuit system example are presented to demonstrate the effectiveness and advantages of the proposed approach.
In the context of Domain Incremental Learning for Semantic Segmentation, catastrophic forgetting is a significant issue when a model learns new geographical domains. While replay-based approaches have been commonly us...
详细信息
Ensemble learning aggregates outputs from multiple base learners for better performance. Bootstrap aggregating (bagging) and boosting are two popular such approaches. They are suitable for integrating unstable base le...
详细信息
We address the problem of resilient consensus in second-order multiagent networks (MANs) with a time-varying set of malicious agents. Existing research typically assumes a static set of malicious agents. However, in r...
详细信息
We address the problem of resilient consensus in second-order multiagent networks (MANs) with a time-varying set of malicious agents. Existing research typically assumes a static set of malicious agents. However, in real-world scenarios, network attacks and recoveries may lead to dynamic changes in the roles of agents. To overcome this challenge, a novel adaptive weighted subsequence reduction (AWSR) is proposed for resilient consensus in MANs with a dynamic malicious agent set. Based on the AWSR approach, a resilient consensus algorithm is designed for second-order MANs. Graph-theoretic conditions in the context of network robustness are derived to ensure resilient consensus in the considered MAN. We also extend the algorithm and analysis to MANs with communication delays. Finally, simulation experiments are conducted to demonstrate the effectiveness of the proposed algorithms.
This study introduces an innovative approach for gesture recognition in smart wearable devices using a deep domain adaptation model, focusing on the challenges posed by heterogeneous user environments and the need for...
详细信息
The robustness of multiple Euler-Lagrange systems (MELSs) measures the capability of maintaining coordination control performance when experiencing disturbance or faults. This article investigates a leader-follower ro...
详细信息
The robustness of multiple Euler-Lagrange systems (MELSs) measures the capability of maintaining coordination control performance when experiencing disturbance or faults. This article investigates a leader-follower robust consensus problem of MELSs, which experience time-varying input disturbance and uncertain communication link faults (CLFs). First of all, based on an adaptive control theory, we design fully distributed observers for estimating the dynamic and state of leader, which have robustness to CLFs. Then, an observer-based proportional-integral (PI) control protocol is designed to achieve consensus of MELSs with robustness to time-varying input disturbance. Different from the existing related results, this robust observer-based PI controller is fully distributed and model independent, which is irrelevant to any prior information (i.e., the structures and features) of agent dynamic or global network information. At last, the validity of the proposed theoretical results is confirmed by a simulation example.
Many real-world systems interact with one another through dependency links, which reduces the system robustness. Most previous studies on the robustness of interdependent networks focus on undirected networks, and the...
详细信息
Many real-world systems interact with one another through dependency links, which reduces the system robustness. Most previous studies on the robustness of interdependent networks focus on undirected networks, and the related studies on directed networks are limited to random or targeted attacks. However, some failure scenarios cannot be described by these two kinds of attacks, such as earthquakes, floods, and epidemics, where systems are attacked in a local range. In this work, we develop a theoretical framework for analyzing the robustness of interdependent directed networks under localized attacks. We find that for degree homogeneous networks, network robustness under localized attacks is similar to that under random attacks. There are four phase transitions in the phase diagram of the network, and a four-phase transition point and a two-phase transition point are found. For degree heterogeneous networks, localized attacks are more likely to lead to collapse than random attacks. As the coupling strength between networks increases, the interdependent networks first show a second order phase transition, and then a hybrid phase transition, and a first order phase transition at last. Furthermore, as the degree heterogeneity increases, the robustness of networks first increases and then decreases, showing a local robustness maximum. The findings could help understand network robustness and enable better design of robust interdependent systems.
For large-scale heterogeneous nonlinear multi-agent systems (MASs) consisting of abundant first-order and second-order agents, this paper presents a novel framework based on partial differential equations (PDEs) to fa...
详细信息
For large-scale heterogeneous nonlinear multi-agent systems (MASs) consisting of abundant first-order and second-order agents, this paper presents a novel framework based on partial differential equations (PDEs) to facilitate their practically finite-time deployment in 2D or 3D space. First, through designing appropriate network communication protocols (NCPs), a heterogeneous nonlinear PDE model composed of a heat equation and a damped wave equation is constructed to characterize the collective dynamics of considered heterogeneous nonlinear MASs. Second, a single-point control strategy and a double-boundary control strategy are proposed, which could not only ensure the well-posedness of the closed-loop heterogeneous PDEs but also enable the finite-time deployment of multi agents. Notably, to better align with real MASs and operating environment, the network topologies and controllers are designed to be semi-Markov switched, while adhering to multiple asynchronous switching rules. Third, with the designed NCPs and control schemes, several sufficient conditions are derived to guarantee the practically finite-time stability of error systems. Finally, two numerical examples and an application example are conducted to validate effectiveness and practicability of the developed approaches.
This article investigates global exponential synchronization of complex networks with reaction diffusions and finite distributed delays coupling (RDDCCNs). Some restrictions on the coupling configuration matrix and ti...
详细信息
This article investigates global exponential synchronization of complex networks with reaction diffusions and finite distributed delays coupling (RDDCCNs). Some restrictions on the coupling configuration matrix and timevarying delays are removed. By constructing comparison functions, utilizing analytic method and inequality techniques, a criterion for exponential synchronization of RDDCCNs is attained under adaptive intermittent control. A corresponding outcome without RDs is also acquired. At last, two simulations are proposed to illustrate validity of the obtained criteria.
Affective associative memory is one method by which agents acquire knowledge, experience, and skills from natural surroundings or social activities. Using neuromorphic circuits to implement affective associative memor...
详细信息
Affective associative memory is one method by which agents acquire knowledge, experience, and skills from natural surroundings or social activities. Using neuromorphic circuits to implement affective associative memory aids in developing brain-inspired intelligence. In this article, a feature-affective associative memory (FAAM) network model and its memristive circuit are proposed for real-time and mutual associative memory and retrieval between multiple features and emotions. With the context of fear conditioning, FAAM network circuit is verified to enable the acquisition and extinction of associations. Different from other works, the proposed temporal-rate mixed coding circuit encodes stimulus intensity and arousal level as different pulses, allowing the associative learning rate and emotion degree can vary with stimulus intensity and arousal level. Furthermore, the bidirectional and multifeature-to-multiemotion association model allows the circuit to be extended to associative memory network containing 10 neurons and 90 synapses, with capabilities such as emotion generation and modulation, associative generalization and differentiation, which are applied to feature binding, situational memory, and inference decision. This work enables advanced cognitive functions and is expected to enable intelligent robot platforms for real-time learning, reasoning decisions, and emotional companionship in dynamic environments.
暂无评论