Entity

Time filter

Source Type

Engineering, Singapore

Liao F.,National University of Singapore | Liao F.,Temasek Laboratories | Wang J.L.,Nanyang Technological University
AIAA Guidance, Navigation, and Control Conference 2012 | Year: 2012

In this paper, an unified consensus-based feedback control strategy is proposed for formation and reconfiguration control of nonholonomic UAVs with velocity constraints. A state-dependent network topology is used to model the communication links among UAVs. Based on the conception of consensus, proportional and integral (PI) control of the relative position between vehicles is introduced to achieve and maintain a desired formation configuration. A bounded time-varying control parameter is designed for each UAV to ensure that its velocity always satisfy the given velocity constraints. The proposed control strategy is shown to guarantee that the multi-UAV system converge to a desired formation configuration subject to the given velocity constraints. An example is used to demonstrate the proposed approach. © 2012 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved. Source


Wang B.,National University of Singapore | Khoo B.C.,National University of Singapore | Khoo B.C.,Temasek Laboratories
Journal of Computational Physics | Year: 2013

In this paper, we present a hybridizable discontinuous Galerkin (HDG) method for solving the Stokes interface problems with discontinuous viscosity and variable surface tension. The jump condition of the stress tensor across the interface is naturally incorporated into the HDG formulation through a constraint on the numerical flux. The most important feature of HDG method compared to other DG methods is that it reduces the number of globally coupled unknowns significantly when high order approximate polynomials are used. For problems with polygonal interfaces, it provides optimal convergence rates of order k + 1 in L2-norm for the velocity, pressure and as well as the gradient of velocity. Furthermore, a new approximate velocity can be obtained by an element-by-element postprocessing which converges with order k + 2 in the L2-norm. For Stokes interface problems with curved interfaces, we use general curvilinear element to ensure the optimal convergence rates. An error estimate is given for the approximation of the interface. It indicates that curvilinear elements of degree 2k + 1 should be used for optimal convergence rate of order k + 1. © 2013 Elsevier Inc. Source


Zhao Z.,National University of Singapore | Zhao Z.,Northwestern Polytechnical University | Zhao Z.,Temasek Laboratories | Li J.-M.,National University of Singapore | And 6 more authors.
AIAA Journal | Year: 2015

The shock wave behavior generated from a single shot of pulsed nanosecond dielectric-barrier-discharge plasma actuator with varying pulse voltages in quiescent air was studied by experiments and numerical simulations. The experiments included using the schlieren technique, a fast response pressure transducer, and a two velocity-component particle image velocimetry system to measure the propagation of the shock wave, the shock overpressure, and the shock induced flow, respectively. For the numerical simulation, a simple "phenomenological approach" was employed by modeling the plasma region over the encapsulated electrode as a jump-heated and pressurized gas layer. The present investigation revealed that the behaviors of the shock wave generated by the nanosecond pulsed plasma were fundamentally a microblast wave, and their speed and strength were found to increase with higher input voltages. The blast wave occured about 1 to 3 μs after the discharge of the nanosecond pulse, which was dependent on the input voltages, and then it decayed quickly from a speed of around 400 m/s (Mach = 1.14) to a speed of around 360 m/s (Mach = 1.03) within about 5 μs (2-3mmfrom the actuator surface). The shock-induced burst perturbations (overpressure and induced velocity) were found to be restricted to a very narrow region (about 1 mm) behind the shock front and lasted only for a few microseconds. In a relatively long time period after the discharge of the plasma, a fairly weak induced vortex structure was observed. These results indicated that the pulsed nanosecond dielectric-barrier-discharge plasma actuator had stronger local effects in time and spatial domain. Source


Yu K.,National University of Singapore | Shen K.,National University of Singapore | Shao S.,National University of Singapore | Ng W.C.,National University of Singapore | And 2 more authors.
IEEE Transactions on Biomedical Engineering | Year: 2011

Searching for target images in large volume imagery is a challenging problem and the rapid serial visual presentation (RSVP) triage is potentially a promising solution to the problem. RSVP triage is essentially a cortically-coupled computer vision technique that relies on single-trial detection of event-related potentials (ERP). In RSVP triage, images are shown to a subject in a rapid serial sequence. When a target image is seen by the subject, unique ERP characterized by P300 are elicited. Thus, in RSVP triage, accurate detection of such distinct ERP allows for fast searching of target images in large volume imagery. The accuracy of the distinct ERP detection in RSVP triage depends on the feature extraction method, for which the common spatial pattern analysis (CSP) was used with limited success. This paper presents a novel feature extraction method, termed common spatio-temporal pattern (CSTP), which is critical for robust single-trial detection of ERP. Unlike the conventional CSP, whereby only spatial patterns of ERP are considered, the present proposed method exploits spatial and temporal patterns of ERP separately, providing complementary spatial and temporal features for high accurate single-trial ERP detection. Numerical study using data collected from 20 subjects in RSVP triage experiments demonstrates that the proposed method offers significant performance improvement over the conventional CSP method (corrected p-value < 0.05, Pearson r=0.64) and other competing methods in the literature. This paper further shows that the main idea of CSTP can be easily applied to other methods. © 2011 IEEE. Source


Lei L.,Temasek Laboratories | Lie J.P.,Temasek Laboratories | Gershman A.B.,TU Darmstadt | See C.M.S.,Temasek Laboratories
ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings | Year: 2010

Two new approaches to adaptive beamforming in sparse subarray-based sensor arrays are proposed. Each subarray is assumed to be well calibrated but the intersubarray gain and/or phase mismatches are assumed to remain unknown or imperfectly known. Our first approach is based on a worst-case beamformer design that, unlike the existing worst-case designs, exploits a structured ellipsoidal uncertainty model for the signal steering vector. Our second approach exploits the idea of estimating the signal steering vector by maximizing the output power of the minimum variance beamformer. Several modifications of our second approach are developed for the cases of gain-and-phase and phase-only intersubarray distortions. ©2010 IEEE. Source

Discover hidden collaborations