Convergence of the auxiliary particle implementation of the PHD filter

Optimal Bayesian multi-target filtering is in general computationally impractical owing to the high dimensionality of the multi-target state. The Probability Hypothesis Density (PHD) filter propagates the first moment of the multi-target posterior distribution. While this reduces the dimensionality of the problem, the PHD filter still involves intractable integrals in many cases of interest. Several authors have proposed Sequential Monte Carlo (SMC) implementations of the PHD filter. However, these implementations are the equivalent of the Bootstrap Particle Filter, and the latter is well known to be inefficient. Drawing on ideas from the Auxiliary Particle Filter (APF), a SMC implementation of the PHD filter which employs auxiliary variables to enhance its efficiency was proposed by Whiteley et. al. Numerical examples were presented for two scenarios, including a challenging nonlinear observation model, to support the claim. This paper studies the theoretical properties of this auxiliary particle implementation. $\mathbb{L}_p$ error bounds are established from which almost sure convergence follows.

DEVELOPMENT OF THE WAVE CONTOURING RAFT.

The wave contouring raft system is the outcome of ideas initiated and developed by Sir Christopher Cockerell from 1972 onwards. His objective was to develop a wave energy device which is within the bounds of current technology. It should consist of simple, relatively small units, amenable to quantity production, which would enable a power generating system to be built up and commissioned in stages according to needs and production capability. This thinking led to the investigation of chains of pontoons, hinged together so that the passage of a wave down the chain causes the pontoons to oscillate relative to one another. Energy is extracted from the sea by applying a torque about the hinges to damp the motion. The work has involved extensive model testing in wave tanks and the building and testing of a 3-unit 1/10 scale power generating installation in the Solent, as well as design studies for a full size installation for Atlantic conditions.

The effect of surface hardening on the elastic shakedown of elliptical contacts

The effect of varying both the aspect ratio and the coefficient of friction of contacts with elliptical geometry on their elastic shakedown performance has been examined theoretically for surfaces with two types of subsurface hardness or strength profiles. In stepwise hardening the hard layer is of uniform strength while in linear hardening its strength reduces from a maximum at the surface to that of the core at the base of the hardened layer. The shakedown load is expressed as the ratio of the maximum Hertzian pressure to the strength of the core material. As the depth of hardening, expressed as a multiple of the elliptical semi-axis, is increased so the potential shakedown load increases from a level that is appropriate to a uniform half-space of unhardened material to a value reflecting the hardness of the surface and near-surface material. In a step-hardened material, the shakedown limit for a surface 'pummelled' by the passage of a sequence of such loads reaches a cut-off or plateau value, which cannot be exceeded by further increases in hardening depth irrespective of the value of the friction coefficient. For a linear-hardened material the corresponding plateau is approached asymptotically. The work confirms earlier results on the upper bounds on shakedown of both point and line contacts and provides numerical values of shakedown loads for intermediate geometries. In general, the case depth required to achieve a given shakedown limit reduces in moving from a transversely moving nominal line load to an axisymmetric point load.

Mineralization and nanomechanical properties in Articular Calcified Cartilage (ACC)

The contribution of the relative volumes of mineral and collagen to the nanomechanical behavior of articular calcified cartilage is explored using nanoindentation, quantitative backscattered electron imaging, and finite element analysis. Elastic modulus generally increases with mineral volume fraction. In highly mineralized tissues, the mineral occupation of water space significantly increases modulus with addition of little mineral. Mineral and organic phases were modeled using Hashin-Shtrikman composite bounds, calculated as a function of mineral volume fraction. Modulus values fall between the Hashin-Shtrikman bounds, indicating some intermediate degree of mineral phase connectivity. Such connectivity in ACC is greater than that achieved in bone and results from uniform collagen orientation and large volume of water space available for mineral occupation.

Modelling effects of axial load on behaviour of pile groups in laterally-spreading soil

Previous research into the behaviour of piled foundations in laterally-spreading soil deposits has concentrated on pile groups that carry small or negligible axial loads. This paper presents dynamic centrifuge test results for 2 x 2 pile groups with bending and geometric properties similar to real 0.5 m diameter tubular steel and solid circular reinforced-concrete field piles. Axial loads applied represented upper-bounds on typical working loads. The simultaneous scaling of the relevant properties controlling both lateral and axial behaviour allows comparisons to be drawn regarding the particular mechanisms of failure that would dominate for each type of pile. Flexible reinforced-concrete piles which tend to carry lower loads were found to be dominated by lateral effects, while steel piles, which are much stiffer and usually carry greater loads are dominated by settlement considerations. © 2006 Taylor & Francis Group.

Simulation of solitary wave impact on coastal structures using weakly compressible and incompressible SPH methods

In this paper, we engage a Lagrangian, particle-based CFD method, named Smoothed Particle Hydrodynamic (SPH) to study the solitary wave motion and its impact on coastal structures. Two-dimensional weakly compressible and incompressible SPH models were applied to simulate wave impacting on seawall and schematic coastal house. The results confirmed the accuracy of both models for predicting the wave surface profiles. The incompressible SPH model performed better in predicting the pressure field and impact loadings on coastal structures than the weakly compressible SPH model. The results are in qualitatively agreement with experimental results. Copyright © 2011 by the International Society of Offshore and Polar Engineers (ISOPE).

The limits of flexoelectricity in liquid crystals

The flexoelectric conversion of mechanical to electrical energy in nematic liquid crystals is investigated using continuum theory. Since the electrical energy produced cannot exceed the mechanical energy supplied, and vice-versa, upper bounds are imposed on the magnitudes of the flexoelectric coefficients in terms of the elastic and dielectric coefficients. For conventional values of the elastic and dielectric coefficients, it is shown that the flexoelectric coefficients may not be larger than a few tens of pC/m. This has important consequences for the future use of such flexoelectric materials in devices and the related energetics of distorted equilibrium structures. © 2011 Author(s).

Optimal control with stochastic PDE constraints and uncertain controls

The optimal control of problems that are constrained by partial differential equations with uncertainties and with uncertain controls is addressed. The Lagrangian that defines the problem is postulated in terms of stochastic functions, with the control function possibly decomposed into an unknown deterministic component and a known zero-mean stochastic component. The extra freedom provided by the stochastic dimension in defining cost functionals is explored, demonstrating the scope for controlling statistical aspects of the system response. One-shot stochastic finite element methods are used to find approximate solutions to control problems. It is shown that applying the stochastic collocation finite element method to the formulated problem leads to a coupling between stochastic collocation points when a deterministic optimal control is considered or when moments are included in the cost functional, thereby forgoing the primary advantage of the collocation method over the stochastic Galerkin method for the considered problem. The application of the presented methods is demonstrated through a number of numerical examples. The presented framework is sufficiently general to also consider a class of inverse problems, and numerical examples of this type are also presented. © 2011 Elsevier B.V.

Modelling effects of axial load on behaviour of pile groups in laterally-spreading soil

Previous research into the behaviour of piled foundations in laterally-spreading soil deposits has concentrated on pile groups that carry small or negligible axial loads. This paper presents dynamic centrifuge test results for 2×2 pile groups with bending and geometric properties similar to real 0.5m diameter tubular steel and solid circular reinforced-concrete field piles. Axial loads applied represented upper-bounds on typical working loads. The simultaneous scaling of the relevant properties controlling both lateral and axial behaviour allows comparisons to be drawn regarding the particular mechanisms of failure that would dominate for each type of pile. Flexible reinforced-concrete piles which tend to carry lower loads were found to be dominated by lateral effects, while steel piles, which are much stiffer and usually carry greater loads are dominated by settlement considerations. © 2006 Taylor & Francis Group, London.

Copula-based Kernel Dependency Measures

The paper presents a new copula based method for measuring dependence between random variables. Our approach extends the Maximum Mean Discrepancy to the copula of the joint distribution. We prove that this approach has several advantageous properties. Similarly to Shannon mutual information, the proposed dependence measure is invariant to any strictly increasing transformation of the marginal variables. This is important in many applications, for example in feature selection. The estimator is consistent, robust to outliers, and uses rank statistics only. We derive upper bounds on the convergence rate and propose independence tests too. We illustrate the theoretical contributions through a series of experiments in feature selection and low-dimensional embedding of distributions.

Statistics and dynamics of turbulence-flame alignment in premixed combustion

The geometric alignment of turbulent strain-rate structures with premixed flames greatly influences the results of the turbulence-flame interaction. Here, the statistics and dynamics of this alignment are experimentally investigated in turbulent premixed Bunsen flames using high-repetition-rate stereoscopic particle image velocimetry. In all cases, the statistics showed that the most extensive principal strain-rate associated with the turbulence preferentially aligned such that it was more perpendicular than parallel to the flame surface normal direction. The mean turbulence-flame alignment differed between the flames, with the stronger flames (higher laminar flame speed) exhibiting stronger preferential alignment. Furthermore, the preferential alignment was greatest on the reactant side of the mean flame brush. To understand these differences, individual structures of fluid-dynamic strain-rate were tracked through time in a Lagrangian manner (i.e., by following the fluid elements). It was found that the flame surface affected the orientation of the turbulence structures, with the majority of structures rotating as they approached the flame such that their most extensive principal strain-rate was perpendicular to the flame normal. The maximum change in turbulent structure orientation was found to decrease with the strength of the structure, increase with the strength of the flame, and exhibit similar trends when the structure strength and flame strength were represented by a Karlovitz number. The mean change in orientation decreased from the unburnt to burnt side of the flame brush and appears to be influenced by the overall flame shape. © 2011 The Combustion Institute.

Characteristics of blood flow in microchannels and relevant impact on modelling blood behaviour in biochip separators

This paper reports a perspective investigation of computational modelling of blood fluid in microchannel devices as a preparation for future research on fluid-structure interaction (FSI) in biofluid mechanics. The investigation is carried out through two aspects, respectively on physical behaviours of blood flow in microchannels and appropriate methodology for modelling. The physics of blood flow is targeted to the challenges for describing blood flow in microchannels, including rheology of blood fluid, suspension features of red blood cells (RBCs), laminar hydrodynamic influence and effect of surface roughness. The analysis shows that due to the hyperelastic property of RBC and its comparable dimension with microchannels, blood fluid shows complex behaviours of two phase flow. The trajectory and migration of RBCs require accurate description of RBC deformation and interaction with plasma. Following on a discussion of modelling approaches, i.e. Eulerian method and Lagrangian method, the main stream modelling methods for multiphase flow are reviewed and their suitability to blood flow is analysed. It is concluded that the key issue for blood flow modelling is how to describe the suspended blood cells, modelled by Lagrangian method, and couple them with the based flow, modelled by Eulerian method. The multiphase flow methods are thereby classified based on the number of points required for describing a particle, as follows: (i) single-point particle methods, (ii) mutli-point particle methods, (iii) functional particle methods, and (iv) fluid particle methods. While single-point particle methods concentrate on particle dynamic movement, multipoint and functional particle methods can take into account particle mechanics and thus offer more detailed information for individual particles. Fluid particle methods provide good compromise between two phases, but require additional information for particle mechanics. For furthermore detailed description, we suggest to investigate the possibility using two domain coupling method, in which particles and base flow are modelled by two separated solvers. It is expected that this paper could clarify relevant issues in numerical modelling of blood flow in microchannels and induce some considerations for modelling blood flow using multiphase flow methods. © 2012 IEEE.

Minimum sensor duty cycle with guaranteed estimator performance

In this paper, we consider Kalman filtering over a network and construct the optimal sensor data scheduling schemes which minimize the sensor duty cycle and guarantee a bounded error or a bounded average error at the remote estimator. Depending on the computation capability of the sensor, we can either give a closed-form expression of the minimum sensor duty cycle or provide tight lower and upper bounds of it. Examples are provided throughout the paper to demonstrate the results. © 2012 IEEE.

Sensor graphs for guaranteed cooperative localization performance

A group of mobile robots can localize cooperatively, using relative position and absolute orientation measurements, fused through an extended Kalman filter (ekf). The topology of the graph of relative measurements is known to affect the steady-state value of the position error covariance matrix. Classes of sensor graphs are identified, for which tight bounds for the trace of the covariance matrix can be obtained based on the algebraic properties of the underlying relative measurement graph. The string and the star graph topologies are considered, and the explicit form of the eigenvalues of error covariance matrix is given. More general sensor graph topologies are considered as combinations of the string and star topologies, when additional edges are added. It is demonstrated how the addition of edges increases the trace of the steady-state value of the position error covariance matrix, and the theoretical predictions are verified through simulation analysis.

The interaction of elasticity and rocking in flexible structures allowed to uplift

Numerous structures uplift under the influence of strong ground motion. Although many researchers have investigated the effects of base uplift on very stiff (ideally rigid) structures, the rocking response of flexible structures has received less attention. Related practical analysis methods treat these structures with simplified 'equivalent' oscillators without directly addressing the interaction between elasticity and rocking. This paper addresses the fundamental dynamics of flexible rocking structures. The nonlinear equations of motion, derived using a Lagrangian formulation for large rotations, are presented for an idealized structural model. Particular attention is devoted to the transition between successive phases; a physically consistent classical impact framework is utilized alongside an energy approach. The fundamental dynamic properties of the flexible rocking system are compared with those of similar linear elastic oscillators and rigid rocking structures, revealing the distinct characteristics of flexible rocking structures. In particular, parametric analysis is performed to quantify the effect of elasticity on uplift, overturning instability, and harmonic response, from which an uplifted resonance emerges. The contribution of stability and strength to the collapse of flexible rocking structures is discussed. © 2012 John Wiley & Sons, Ltd.