A Learning-based Multi-RRT Approach for Robot Path Planning in Narrow Passages

As an important class of sampling-based path planning methods, the Rapidly-exploring Random Trees (RRT) algorithm has been widely studied and applied in the literature. In RRT, how to select a tree to extend or connect is a critical factor, which will greatly influence the efficiency of path planning. In this paper, a novel learning-based multi-RRTs (LM-RRT) approach is proposed for robot path planning in narrow passages. The LM-RRT approach models the tree selection process as a multi-armed bandit problem and uses a reinforcement learning algorithm that learns action values and selects actions with an improved ε-greedy strategy (ε _t -greedy). Compared with previous RRT algorithms, LM-RRT can not only enhance the local space exploration ability of each tree, but also guarantee the efficiency of global path planning. The probabilistic completeness and combinatory optimality of LM-RRT are proved based on the geometric characteristics of the configuration space. Simulation and experimental results show the effectiveness of the proposed LM-RRT approach in single-query path planning problems with narrow passages.     

On mitigating hotspots to maximize network lifetime in multi-hop wireless sensor network with guaranteed transport delay and reliability

In wireless sensor networks (WSNs), sensor nodes close to the sink consume more energy than others because they are burdened with heavier relay traffic destined for the sink and trend to die early, forming hotspots or energy holes in WSNs. It has a serious impact on network lifetime. In this paper, three optimization algorithms are proposed to mitigate hotspots and prolong network lifetime for adaptive Mary Phase Shift Keying (MPSK) based wireless sensor networks while transport delay and reliability can be still guaranteed. Based on the insight gained into the relationship between nodal data load and energy consumption in different regions, the first algorithm (GlobalSame) can extend considerably the network lifetime by selecting the optimal nodal transmission radius r, bit error rate ε and transmission rate allocations in bits per symbol (BPS) τ. The second algorithm (RingSame) can further improve network lifetime by comparison to the GlobalSame algorithm, which by selecting different ε _i and τ _i for nodes in different regions under constraints of total BER and transport delay . While the third algorithm (NodeDiff) can further improve the network lifetime by adopting different BER ε and BPS τ parameters of the same node for data packets received according to its distance to the sink. Extensive simulation studies show that our algorithms do considerably prolong the network lifetime.     

Multivariate adaptive regression spline (MARS) and least squares support vector machine (LSSVM) for OCR prediction

This article investigates the feasibility of multivariate adaptive regression spline (MARS) and least squares support vector machine (LSSVM) for the prediction of over consolidation ratio (OCR) of clay deposits based on Piezocone Penetration Tests (PCPT) data. MARS uses piece-wise linear segments to describe the non-linear relationships between input and output variables. LSSVM is firmly based on the theory of statistical learning, and uses regression technique. The input parameters of the models are corrected cone resistance (q _t ), vertical total stress (σ_v), hydrostatic pore pressure (u ₀), pore pressure at the cone tip (u ₁), and the pore pressure just above the cone base (u ₂). The developed LSSVM model gives error bar of predicted OCR. Equations have also been developed for prediction of OCR. The performance of MARS and LSSVM models has been compared with the traditional methods for OCR prediction. As the results reveal, the proposed MARS and LSSVM models are robust models for determination of OCR.     

Multi-objective optimal reactive power dispatch considering voltage stability in power systems using HFMOEA

This paper proposes a new hybrid fuzzy multi-objective evolutionary algorithm (HFMOEA) based approach for solving complex multi-objective, mixed integer nonlinear problems such as optimal reactive power dispatch considering voltage stability (ORPD-VS). In HFMOEA based optimization approach, the two parameters like crossover probability (P_C) and mutation probability (P_M) are varied dynamically through the output of a fuzzy logic controller. The fuzzy logic controller is designed on the basis of expert knowledge to enhance the overall stochastic search capability for generating better pareto-optimal solution. Two detailed case studies are presented: Firstly, the performance of HFMOEA is tested on five benchmark test problems such as ZDT1, ZDT2, ZDT3, ZDT4 and ZDT6 as suggested by Zitzler, Deb and Thiele; Secondly, HFMOEA is applied to multi-objective ORPD-VS problem. In both the case studies, the optimization results obtained from HFMOEA are analysed and compared with the same obtained from two versions of elitist non-dominated sorting genetic algorithms such as NSGA-II and MNSGA-II in terms of various performance metrics. The simulation results are promising and confirm the ability of HFMOEA for generating better pareto-optimal fronts with superior convergence and diversity.     

A simple factor-3 approximation for labeling points with circles

In this paper we present a simple factor-(3+ε), 0<ε<1, approximation algorithm, which runs in O(nlogn+n(1/ε)^O(1/ε2)log(D₃/εD₂)) time, for the problem of labeling a set P of n distinct points with uniform circles. (D₂ is the closest pair of P and D₃ is the minimum diameter of all subsets of P with size three.) This problem is known to be NP-hard. Our bound improves the previous factor of 3.6+ε.     

A systematic estimation model for fraction nonconforming of a wafer in semiconductor manufacturing research

The clustering phenomenon of defects usually occurs in semiconductor manufacturing. However, previous studies did not pay much attention to the influence of clustering phenomenon for estimating fraction nonconforming of a wafer. Thus, this paper presents a systematic estimation model with considering relevant variables about clustering defects for fraction nonconforming of a wafer. The method combines back-propagation neural network (BPNN) with genetic algorithm (GA) to obtain an estimation model. In this study, GA aims to optimize the parameters of BPNN. Five relevant variables: number of defects (ND), squared coefficient of angle variation (SCV_A) for defects, squared coefficient of distance variation (SCV_D) for defects, defect cluster index (CI_M), and the number of cluster groups (NCG) for defects by self-organized map (SOM) are utilized as inputs for GA–BPNN. Finally, a simulation case and a real-world case are used to confirm the effectiveness of proposed method.     

On-line prediction of micro-turning multi-response variables by machine vision system using adaptive neuro-fuzzy inference system (ANFIS)

In this paper, a new attempt has been made in the area of tool-based micromachining for automated, non-contact, and flexible prediction of quality responses such as average surface roughness (R _a), tool wear ratio (TWR) and metal removal rate (MRR) of micro-turned miniaturized parts through a machine vision system (MVS) which is integrated with an adaptive neuro-fuzzy inference system (ANFIS). The images of machined surface grabbed by the MVS could be extracted using the algorithm developed in this work, to get the features of image texture [average gray level (G _a)]. This work presents an area-based surface characterization technique which applies the basic light scattering principles used in other optimal optical measurement systems. These principles are applied in a novel fashion which is especially suitable for in-process prediction and control. The main objective of this study is to design an ANFIS for estimation of R _a, TWR, and MRR in micro-turning process. Cutting speed (S), feed rate (F), depth of cut (D), G _a were taken as input parameters and R _a, TWR, MRR as the output parameters. The results obtained from the ANFIS model were compared with experimental values. It is found that the predicted values of the responses are in good agreement with the experimental values.     

An effective GRASP and tabu search for the 0–1 quadratic knapsack problem

As a generalization of the classical 0-1 knapsack problem, the 0-1 Quadratic Knapsack Problem (QKP) that maximizes a quadratic objective function subject to a linear capacity constraint is NP-hard in strong sense. In this paper, we propose a memory based Greedy Randomized Adaptive Search Procedures (GRASP) and a tabu search algorithm to find near optimal solution for the QKP. Computational experiments on benchmarks and on randomly generated instances demonstrate the effectiveness and the efficiency of the proposed algorithms, which outperforms the current state-of-the-art heuristic Mini-Swarm in computational time and in the probability to achieve optimal solutions. Numerical results on large-sized instances with up to 2000 binary variables have also been reported.     

Normalized difference spectral indices for estimating photosynthetic efficiency and capacity at a canopy scale derived from hyperspectral and CO2 flux measurements in rice

We explored simple and useful spectral indices for estimating photosynthetic variables (radiation use efficiency and photosynthetic capacity) at a canopy scale based on seasonal measurements of hyperspectral reflectance, ecosystem CO₂ flux, and plant and micrometeorological variables. An experimental study was conducted over the simple and homogenous ecosystem of an irrigated rice field. Photosynthetically active radiation absorbed by the canopy (APAR), the canopy absorptivity of APAR (fAPAR), net ecosystem exchange of CO₂ (NEE_CO2) gross primary productivity (GPP), photosynthetic capacity at the saturating APAR (P_max), and three parameters of radiation use efficiency (ε_N: NEE_CO2/APAR; ε_G: GPP/APAR; φ: quantum efficiency) were derived from the data set. Based on the statistical analysis of relationships between these ecophysiological variables and reflectance indicators such as normalized difference spectral indices (NDSI[i,j]) using all combinations of two wavelengths (i and j nm), we found several new indices that would were more effective than conventional spectral indices such as photochemical reflectance index (PRI) and normalized difference vegetation index (NDVI = NDSI[near-infrared, red]). ε_G was correlated well with NDSI[710, 410], NDSI[710, 520], and NDSI[530, 550] derived from nadir measurements. φ was best correlated with NDSI[450, 1330]. NDSI[550, 410] and NDSI[720, 420] had a consistent linear relationships with fAPAR throughout the growing season, whereas conventional indices such as NDVI showed very different relationships before and after heading. Off-nadir measurements were more closely related to the efficiency parameters than nadir measurements. Our results provide useful insights for assessing plant productivity and ecosystem CO₂ exchange, using a wide range of available spectral data as well as useful information for designing future sensors for ecosystem observations.     

A probabilistic solution of robust H∞ control problem with scaled matrices

This paper addresses the robust H_∞ control problem with scaled matrices. It is difficult to find a global optimal solution for this non-convex optimisation problem. A probabilistic solution, which can achieve globally optimal robust performance within any pre-specified tolerance, is obtained by using the proposed method based on randomised algorithm. In the proposed method, the scaled H_∞ control problem is divided into two parts: (1) assume the scaled matrices be random variables, the scaled H_∞ control problem is converted to a convex optimisation problem for the fixed sample of the scaled matrix and a optimal solution corresponding to the fixed sample is obtained; (2) a probabilistic optimal solution is obtained by using the randomised algorithm based on a finite number N optimal solutions, which are obtained in part (1). The analysis shows that the worst case complexity of proposed method is a polynomial.     

Differential evolution based on ε-domination and orthogonal design method for power environmentally-friendly dispatch

This paper proposes a differential evolution algorithm based on ε-domination and orthogonal design method (ε-ODEMO) to solve power dispatch problem considering environment protection and saving energy. Besides the operation costs of thermal power plant, contaminative gas emission is also optimized as an objective. In the proposed algorithm, ε-dominance is adopted to make genetic algorithm obtain a good distribution of Pareto-optimal solutions in a small computational time, and the orthogonal design method can generate an initial population of points that are scattered uniformly over the feasible solution space, these modify the differential evolution algorithm (DE) to make it suit for multi-objective optimization (MOO) problems and improve its performance. A test hydrothermal system is used to verify the feasibility and effectiveness of the proposed method. Compared with other methods, the results obtained demonstrate the effectiveness of the proposed algorithm for solving the power environmentally-friendly dispatch problem.     

SteamTables: An approach of multiple variable sets

Using the IAPWS-95 formulation, an ActiveX component SteamTablesIIE in Visual Basic 6.0 is developed to calculate thermodynamic properties of pure water as a function of two independent intensive variables: (1) temperature (T) or pressure (P) and (2) T, P, volume (V), internal energy (U), enthalpy (H), entropy (S) or Gibbs free energy (G). The second variable cannot be the same as variable 1. Additionally, it calculates the properties along the separation boundaries (i.e., sublimation, saturation, critical isochor, ice I melting, ice III to ice IIV melting and minimum volume curves) considering the input parameter as T or P for the variable 1.SteamTablesIIE is an extension of the ActiveX component SteamTables implemented earlier considering T (190 to 2000 K) and P (3.23×10^?8 to 10000 MPa) as independent variables. It takes into account the following 27 intensive properties: temperature (T), pressure (P), fraction, state, volume (V), density (Den), compressibility factor (Z₀), internal energy (U), enthalpy (H), Gibbs free energy (G), Helmholtz free energy (A), entropy (S), heat capacity at constant pressure (C_p), heat capacity at constant volume (C_v), coefficient of thermal expansion (CTE), isothermal compressibility (Z_iso), speed of sound (VelS), partial derivative of P with T at constant V (dPdT), partial derivative of T with V at constant P (dTdV), partial derivative of V with P at constant T (dVdP), Joule-Thomson coefficient (JTC), isothermal throttling coefficient (IJTC), viscosity (Vis), thermal conductivity (ThrmCond), surface tension (SurfTen), Prandtl number (PrdNum) and dielectric constant (DielCons).     

Convergence analysis of recursive identification algorithms with forgetting factor

A convergence analysis of a modified version of the least-squares recursive identification algorithm with forgetting factor is given. It is shown that the parametric distance converges to a zero mean random variable. It is also shown that, under persistent excitation condition on both system input and output, the condition number of the adaptation gain matrix is bounded. The variance of the parametric distance is bounded by the product of the noise variance times the upper bound of the condition number of the gain matrix. This is done by normalizing the measurement vector entering in the identification algorithm and by using a forgetting factor verifying λ_t ? 1 ? ε; ε >0.     

Modelling the nonlinear dynamic behaviour of a boiler‐turbine system using a radial basis function neural network

Building an appropriate mathematical model that describes the system behaviour with a certain degree of satisfaction is quite challenging owing to the uncertain and volatile nature of thermodynamic constants and geometric parameters. In this paper, we present a technique to approximate and validate the dynamic behaviour of the Aström–Bell boiler‐turbine power plant based on an RBFNN over a large operating range. The proposed RBFNN is applied to solve the parametric identification problem for nonlinear and complex systems using an optimiser based on a hybrid genetic algorithm. This optimiser is composed of the gradient descent optimiser and a genetic algorithm for fast convergence. Two simulations were performed to show the effectiveness of the proposed technique under different situations with several boiler‐turbine input variables. The optimal structure and parameters of the obtained RBFNN‐based model emulates well the dynamic behaviour of the Aström–Bell boiler‐turbine system. Copyright © 2013 John Wiley & Sons, Ltd.     

Deriving maximal light use efficiency from coordinated flux measurements and satellite data for regional gross primary production modeling

Remote sensing models based on light use efficiency (LUE) provide promising tools for monitoring spatial and temporal variation of gross primary production (GPP) at regional scale. In most of current LUE-based models, maximal LUE (ε_max) heavily relies on land cover types and is considered as a constant, rather than a variable for a certain vegetation type or even entire eco-region. However, species composition and plant functional types are often highly heterogeneous in a given land cover class; therefore, spatial heterogeneity of ε_max must be fully considered in GPP modeling, so that a single cover type does not equate to a single ε_max value. A spatial dataset of ε_max accurately represents the spatial heterogeneity of maximal light use would be of significant beneficial to regional GPP models. Here, we developed a spatial dataset of ε_max by integrating eddy covariance flux measurements from 14 field sites in a network of coordinated observation across northern China and satellite derived indices such as enhanced vegetation index (EVI) and visible albedo to simulate regional distribution of GPP. This dynamic modeling method recognizes the spatial heterogeneity of ε_max and reduces the uncertainties in mixed pixels. Further, we simulated GPP with the spatial dataset of ε_max generated above. Both ε_max and growing season GPP show complex patterns over northern China that reflect influences of humidity, green vegetation fractions, and land use intensity. “Green spots” such as oasis meadow and alpine forests in dryland and “brown spots” such as build-up and heavily degraded vegetation in the east are clearly captured by the simulation. The correlation between simulated GPP and EC measured GPP indicate that the simulated GPP from this new approach is well matched with flux-measured GPP. Those results have demonstrated the importance of considering ε_max as both a spatially and temporally variable values in GPP modeling.     

Probability of failure for uncertain control systems using neural networks and multi-objective uniform-diversity genetic algorithms (MUGA)

There are much research effort in the literature using Monte Carlo simulation (MCS) which is a direct and simple numerical method, however, it can be computationally very expensive as the governing dynamic equations of the system to be simulated for each random sample using the MCS. In this paper, polynomial meta-models based on the evolved group method of data handling (GMDH) neural networks are obtained to simply calculate the probability of failure in the MCS, instead of direct solution of dynamic equation of system. In this way, some input–output data consisting of uncertain parameters of system and controller parameters as inputs and probability of failure of some cost functions as output are used for training such GMDH-type neural networks which replace the very time consuming direct solution of dynamical systems during the MCS. A multi-objective genetic algorithm is also used for Pareto optimal design of PI and PID controllers for both first and second order uncertain system with time delays using methodology of this paper. The objective functions that are considered for such Pareto multi-objective optimization are namely, probability of failure of settling time (P_TS) and probability of failure of overshoot (P_OS). The comparisons of the obtained results using the method of this paper with those obtained using direct method shows a significant reduction in computational time, whilst the accuracy is maintained.     

High order stable infinite impulse response filter design using cuckoo search algorithm

In this paper, an efficient technique for optimal design of digital infinite impulse response (IIR) filter with minimum passband error (e _p ), minimum stopband error (e _s ), high stopband attenuation (A _s ), and also free from limit cycle effect is proposed using cuckoo search (CS) algorithm. In the proposed method, error function, which is multi-model and non-differentiable in the heuristic surface, is constructed as the mean squared difference between the designed and desired response in frequency domain, and is optimized using CS algorithm. Computational efficiency of the proposed technique for exploration in search space is examined, and during exploration, stability of filter is maintained by considering lattice representation of the denominator polynomials, which requires less computational complexity as well as it improves the exploration ability in search space for designing higher filter taps. A comparative study of the proposed method with other algorithms is made, and the obtained results show that 90% reduction in errors is achieved using the proposed method. However, computational complexity in term of CPU time is increased as compared to other existing algorithms.     

Real coded chemical reaction based optimization for short-term hydrothermal scheduling

This paper presents a real coded chemical reaction based (RCCRO) algorithm to solve the short-term hydrothermal scheduling (STHS) problem. Hydrothermal system is highly complex and related with every problem variables in a nonlinear way. The objective of the hydro thermal scheduling is to determine the optimal hourly schedule of power generation for different hydrothermal power system for certain intervals of time such that cost of power generation is minimum. Chemical reaction optimization mimics the interactions of molecules in term of chemical reaction to reach a low energy stable state. A real coded version of chemical reaction optimization, known as real-coded chemical reaction optimization (RCCRO) is considered here. To check the effectiveness of the RCCRO, 3 different test systems are considered and mathematical remodeling of the algorithm is done to make it suitable for solving short-term hydrothermal scheduling problem. Simulation results confirm that the proposed approach outperforms several other existing optimization techniques in terms quality of solution obtained and computational efficiency. Results also establish the robustness of the proposed methodology to solve STHS problems.     

ANN-based GA for generating the sizing curve of stand-alone photovoltaic systems

Recent advances in artificial intelligence techniques have allowed the application of such technologies in real engineering problems. In this paper, an artificial neural network-based genetic algorithm (ANN-GA) model was developed for generating the sizing curve of stand-alone photovoltaic (SAPV) systems. Due to the high computing time needed for generating the sizing curves and complex architecture of the neural networks, the genetic algorithm is used in order to find the optimal architecture of the ANN (number of hidden layers and the number of neurons within each hidden layer). Firstly, a numerical method is used for generating the sizing curves for different loss of load probability (LLP) corresponding to 40 sites located in Algeria. The inputs of ANN-GA are the geographical coordinates and the LLP while the output is the sizing curve represented by C_A = f(C_S) (i.e., 30-points were taken from each sizing curve). Subsequently, the proposed ANN-GA model has been trained by using a set of 36 sites, whereas data for 4 sites (randomly selected) which are not included in the training dataset have been used for testing the ANN-GA model. The results obtained are compared and tested with those of the numerical method in order to show the effectiveness of the proposed approach.     

Quantized H∞ Control for Networked Systems with Communication Constraints

The problem of quantized H_∞ control for networked control systems (NCSs) subject to time‐varying delay and multiple packet dropouts is investigated in this paper. Both the control input and the measurement output signals are quantized before being transmitted and the quantized errors are described as sector bound uncertainties. The measurement channel and the control channel packet dropouts are considered simultaneously, and the stochastic variables satisfying Bernoulli random binary distribution are utilized to model the random multiple packet dropouts. Sufficient conditions for the existence of an observer‐based controller are established to ensure the exponential mean‐square stablility of the closed‐loop system and achieve the optimal H_∞ disturbance attenuation level. By using a globally convergent algorithm involving convex optimization, the nonconvex feasibility can be solved successfully. Finally, a numerical example is given to illustrate the effectiveness and applicability of the proposed method.