Multiple expert classification: a new methodology for parallel decision fusion

Abstract. A new parallel hybrid decision fusion methodology is proposed. It is demonstrated that existing parallel multiple expert decision combination approaches can be divided into two broad categories based on the implicit decision emphasis implemented. The first category consists of methods implementing computationally intensive decision frameworks incorporating a priori information about the target task domain and the reliability of the participating experts, while the second category encompasses approaches implementing group consensus without assigning any importance to the reliability of the experts and ignoring other contextual information. The methodology proposed in this paper is a hybridisation of these two approaches and has shown significant performance enhancements in terms of higher overall recognition rates along with lower substitution rates. Detailed analysis using two different databases supports this claim. Received January 19, 1999 / Revised March 20, 2000     

Time stepping in discontinuous Galerkin method

The time discretization in the Discontinuous Galerkin (DG) scheme has been traditionally based on the Total Variation Diminishing (TVD) second-order Runge-Kutta (RK2) scheme. Computational efficiency and accuracy with the Euler Forward (EF) and the TVD second-order RK2 time stepping schemes in the DG method are investigated in this work. Numerical tests are conducted with the scalar Burgers equation, 1-D and 2-D shallow water flow equations. The maximum Courant number or time step size required for stability for the EF scheme and RK2 scheme with different slope limiters are compared. Numerical results show that the slope limiters affect the stability requirement in the DG method. The RK2 scheme is generally more diffusive than the EF scheme, and the RK2 scheme allows larger time step sizes. The EF scheme is found to be more efficient and accurate than the RK2 scheme in the DG method in computation.     

Adsorptive performance of penta-bismuth hepta-oxide nitrate, Bi?O?NO?, for removal of methyl orange dye

The adsorption of methyl orange dye from aqueous solution onto penta-bismuth hepta-oxide nitrate, Bi(5)O(7)NO(3), synthesized by precipitation method, was studied in a batch adsorption system. The effects of operation parameters such as adsorbent dose, initial dye concentration, pH and temperature were investigated. The adsorption equilibrium and mechanism of adsorption was evaluated by Langmuir and Freundlich isotherm and different kinetic models, respectively. The results indicate that adsorption is highly dependent on all operation parameters. At optimum conditions, the adsorption capacity was found to be 18.9 mg/g. The adsorption data fits well with the Langmuir isotherm model indicating monolayer coverage of adsorbate molecules on the surface of Bi(5)O(7)NO(3). The kinetic studies show that the adsorption process is a second-order kinetic reaction. Although intra-particle diffusion limits the rate of adsorption, the multi-linearity plot of intra-particle model shows the importance of both film and intra-particle diffusion as the rate-limiting steps of the dye removal. Thermodynamic parameters show that the adsorption process is endothermic, spontaneous and favourable at high temperature.     

Implications to stormwater management as a result of lot scale rainwater tank systems: a case study in Western Sydney, Australia

Rainwater tanks are increasingly adopted in Australia to reduce potable water demand and are perceived to reduce the volume of stormwater discharge from developments. This paper investigates the water balance of rainwater tanks, in particular the possible impacts these tanks could have in controlling the stormwater discharge volume. The study collected water quantity data from two sites in the Hawkesbury City Council area, New South Wales, Australia and utilised the collected data in a simple water balance model to assess the effectiveness of rainwater tanks in reducing the stormwater discharge volume. The results indicate that a significant reduction in discharge volume from a lot scale development can be achieved if the rainwater tank is connected to multiple end-uses, but is minimal when using irrigation alone. In addition, the commonly used volumetric runoff coefficient of 0.9 was found to over-estimate the runoff from the roof areas and to thereby under-estimate the available volume within the rainwater tanks for retention or detention. Also, sole reliance on the water in the rainwater tanks can make the users aware of their water use pattern and water availability, resulting in significant reductions in water use as the supply dwindles, through self-imposed water restrictions.     

Vision Based UAV Attitude Estimation: Progress and Insights

Unmanned aerial vehicles (UAVs) are increasingly replacing manned systems in situations that are dangerous, remote, or difficult for manned aircraft to access. Its control tasks are empowered by computer vision technology. Visual sensors are robustly used for stabilization as primary or at least secondary sensors. Hence, UAV stabilization by attitude estimation from visual sensors is a very active research area. Vision based techniques are proving their effectiveness and robustness in handling this problem. In this work a comprehensive review of UAV vision based attitude estimation approaches is covered, starting from horizon based methods and passing by vanishing points, optical flow, and stereoscopic based techniques. A novel segmentation approach for UAV attitude estimation based on polarization is proposed. Our future insightes for attitude estimation from uncalibrated catadioptric sensors are also discussed.     

A correlation-based ant miner for classification rule discovery

In recent years, a few sequential covering algorithms for classification rule discovery based on the ant colony optimization meta-heuristic (ACO) have been proposed. This paper proposes a new ACO-based classification algorithm called AntMiner-C. Its main feature is a heuristic function based on the correlation among the attributes. Other highlights include the manner in which class labels are assigned to the rules prior to their discovery, a strategy for dynamically stopping the addition of terms in a rule’s antecedent part, and a strategy for pruning redundant rules from the rule set. We study the performance of our proposed approach for twelve commonly used data sets and compare it with the original AntMiner algorithm, decision tree builder C4.5, Ripper, logistic regression technique, and a SVM. Experimental results show that the accuracy rate obtained by AntMiner-C is better than that of the compared algorithms. However, the average number of rules and average terms per rule are higher.     

Experimental investigation on the stability of bride girder against tsunami forces

Bridges are the most useful part in the transportation network. Any disruption of the bridge structures may hamper the whole transportation system. In the last recent tsunamis, numerous number of bridges were structurally damaged by the tsunami waves. Lack of proper provisions of tsunami forces in the design guidelines also contributes to the augmentations of the damage level. Therefore, proper evaluation of the tsunami forces on the bridge girder should be introduced in the design promptly. In this study, laboratory experiments were carried out to reveal the damage mechanism of the bridge girder by assessing the exerted tsunami forces. Both broken and unbroken waves were considered for the analysis. The results showed that measured forces were larger for broken waves than those of unbroken waves. Maximum force of the broken waves was 4.59 times as large as the hydrostatic pressure. Also, waves reached the peak value more rapidly for larger wave heights than those of smaller ones. Additionally, a girder that is placed at a higher position is much vulnerable to tsunami hazard.     

A novel direct torque control for interior permanent-magnet synchronous machine drive with low ripple in torque and flux-a speed-sensorless approach

A novel direct torque control (DTC) scheme for an interior permanent magnet synchronous machine (IPMSM) is proposed in this paper, which features low torque and flux ripple and almost fixed switching frequency. The torque and flux ripples have been significantly reduced compared with those of the basic DTC reported in the literature. A speed estimation scheme is integrated with the proposed DTC scheme in order to achieve a fully sensorless high-performance IPMSM drive.     

Adaptive sliding mode approach for learning in a feedforward neural network

An adaptive learning algorithm is proposed for a feedforward neural network. The design principle is based on the sliding mode concept. Unlike the existing algorithms, the adaptive learning algorithm developed does not require a prioriknowledge of upper bounds of bounded signals. The convergence of the algorithm is established and conditions given. Simulations are presented to show the effectiveness of the algorithm.