3 May 2022 in IEEE Transactions on Signal Processing
Ultra Wide Band (UWB) has been emerged as a technology to provide reliable, accurate, and energy-efficient indoor navigation/localization systems. There are, however, several key challenges ahead for its efficient implementation including complexity of the identification/mitigation of Non Line of Sight (NLoS) links, and the limited battery life of UWB beacons, which is especially problematic in practical circumstances with certain beacons located in strategic positions. To address these challenges, we introduce an efficient node selection framework to enhance the location accuracy, without using complex NLoS mitigation methods, while maintaining a balance between the remaining battery life of UWB beacons. Referred to as the Deep Q-Learning Energy-optimized LoS/NLoS (DQLEL) UWB node selection framework, the mobile user is autonomously trained to determine the optimal set of UWB beacons to be localized based on the 2-D Time Difference of Arrival (TDoA) framework. The effectiveness of the proposed DQLEL framework is evaluated in terms of the link condition, the deviation of the remaining battery life of UWB beacons, location error, and cumulative rewards. Based on the simulation results, the proposed DQLEL framework significantly outperformed its counterparts across the aforementioned aspects.
21 March 2022 in IEEE Transactions on Smart Grid
Inspired by the recent surge of interest in security analysis of renewable energies, this article proposes a cyber resilient control framework for large-scale wind parks (WPs) against denial of service (DoS) and deception attacks. A detailed cyber-physical model for a doubly-fed induction generator (DFIG)-based WP is first developed. It is shown that the WP is subject to a stability issue known as the sub-synchronous interaction (SSI). Additionally, two cyber threats in the forms of DoS and deception attack attempt to destabilize the WP from the subsynchronous perspective. To stabilize the WP under the SSI phenomenon and these cyber attacks, an observer-based fuzzy control scheme is proposed which requires only periodic samples of the output. Using a looped-Lyapunov functional (LLF) method for stability analysis, the control and observer gains are computed from the solution of some linear matrix inequality (LMI) conditions. The proposed design framework allows including the desired level of cyber resilience to DoS and deception attacks for computing proper control and observer gains. The effectiveness of the proposed controller against the considered threats is demonstrated via electromagnetic transient (EMT) simulations performed using the EMTP-RV software and a co-simulation platform.
1 September 2007 in Journal of Cognitive Engineering and Decision Making
A lack of guidance for designing complex, dynamic networked systems presents challenges to the development of such systems to maximize overall human-machine system performance. An intelligent adaptive interface (IAI) concept and associated technologies have been developed to address this problem. In order to support effective decision making, a typical IAI is driven by software agents that can change the display and/or control characteristics to react to the changes of mission and operator states in real time. This work investigated the efficacy of IAIs in a multi-uninhabited aerial vehicle (UAV) scenario. The IAI was modeled as part of the UAV tactical workstations found in a maritime patrol aircraft. A performance model was developed to compare the difference in mission activities with and without IAI agents. A prototype IAI experimental environment was implemented for a human-in-the-loop empirical investigation. Both simulation and experiment results revealed that the control of multiple UAVs is a cognitively complex task with high workload. IAIs facilitated a significant reduction in workload and an improvement in situation awareness, thus allowing operators to continue working under high time pressure. This research revealed IAI triggering conditions under different cognitive workload situations.
18 January 2021 in Human-Intelligent Systems Integration
4 March 2016 in Theoretical Issues in Ergonomics Science
This paper provides guidance for adaptive learning systems designers by reviewing the evolution of learning theories and their associated progress, which is enabled by computers, artificial intelligence and augmented cognition. A generic conceptual framework of intelligent adaptive learning systems is discussed in detail for individualised learning. The utility of such design methodologies is illustrated through an example of a customised intelligent tutoring system for improvised explosive device disposal operator training. This testbed approach is the first phase of truly determining the utility and effectiveness of intelligent adaptive learning systems for multiple fields of use in the ever complicated spectrum of society's education and training needs.
8 July 2022 in IEEE Systems Journal
This article proposes a resilient strategy for leaderless and leader–following consensus in general linear multiagent systems under simultaneous presence of false data injection and denial-of-service (DoS) attacks. To save energy, local control updates and communication between the neighboring agents are based on a distributed periodic event-triggered scheme. Sufficient conditions to guarantee the exponential mean square bounded consensus are derived as linear matrix inequality (LMI) conditions obtained from some delay-dependent Lyapunov–Krasovskii functionals. The proposed design framework computes the required consensus control gain based on a desired level of resilience to DoS. The guaranteed level of resilience in our proposed LMI-based method is less conservative (more realistic) compared to the existing analytical solutions. Two physical constraints, namely nonuniform time-varying communication delay and actuator faults, are also included in the problem formulation to enhance the practicability of the proposed solution. Simulations results demonstrate the effectiveness of the proposed framework in the presence of the considered cyberthreats and physical constraints.