Daniel J. Simon

An Analysis Of Diploidy And Dominance In Genetic Algorithms

Daniel J. Simon — Wed, 30 Jan 2013 09:06:54 PST

The use of diploidy and dominance in genetic algorithms (GAs) has long been used to improve performance in time-varying optimization problems. Diploidy increases diversity in GAs by allowing recessive genes to survive in a population and become active at some later time when changes in the environment make them more desirable. This paper suggests an intuitive way to implement diploidy and presents some mathematical analyses of fitness proportional selection to justify its use in time-varying problems. An extension of the classical schema theorem for diploid GAs is presented. The mathematical analyses are geared towards the One Max problem, and assume a GA with selection and mutation only (no crossover). The analyses confirm that diploidy increases diversity, and provide some quantitative results for diversity increase as a function of the GA population characteristics.

A Concept for Sled Testing Minuteman III Guidance Systems

Hossny El-Sherief et al. — Wed, 30 Jan 2013 09:06:53 PST

Optimal Neural-Based Navigation Satellite Selection

Daniel J. Simon et al. — Wed, 30 Jan 2013 09:06:51 PST

The application of neural-based methods to optimal satellite subset selection for navigation use is discussed. The optimal satellite subset is chosen by minimizing a quantity known as Geometric Dilution of Precision (GDOP), which is given by the trace of the inverse of the measurement matrix. An artificial neural network learns the functional relationships between the entries of a measurement matrix and the eigen valuse of its inverse, and thus generates GDOP without inverting a matrix.

Global Positioning System Receiver Simulation for Integrated Inertial Navigation Systems

Daniel J. Simon et al. — Wed, 30 Jan 2013 09:06:50 PST

There is much interest in integrated navigation using the Global Positioning System and Inertial Measurement Units. This integration combines the long-term accuracy of GPS with the short-term accuracy of IMUs, thereby realizing the best of both worlds. There is thus a need for realistic simulations of integrated GPS/IMU navigation systems. This paper discusses such a simulation and presents some simulation results. The specific application considered is missile navigation.

Controller Tuning for DC Motor Speed Control Using Genetic Algorithms

Saurabh Jain et al. — Wed, 30 Jan 2013 09:06:49 PST

Suboptimal Robot Joint Interpolation Within User-Specified Knot Tolerances

Daniel J. Simon et al. — Tue, 29 Jan 2013 05:56:33 PST

Approximation of a desired robot path can be accomplished by interpolating a curve through a sequence of joint-space knots. A smooth interpolated trajectory can be realized by using trigonometric splines. But, sometimes the joint trajectory is not required to exactly pass through the given knots. The knots may rather be centers of tolerances near which the trajectory is required to pass. In this article, we optimize trigonometric splines through a given set of knots subject to user-specified knot tolerances. The contribution of this article is the straightforward way in which intermediate constraints (i.e., knot angles) are incorporated into the parameter optimization problem. Another contribution is the exploitation of the decoupled nature of trigonometric splines to reduce the computational expense of the problem. The additional freedom of varying the knot angles results in a lower objective function and a higher computational expense compared to the case in which the knot angles are constrained to exact values. The specific objective functions considered are minimum jerk and minimum torque. In the minimum jerk case, the optimization problem reduces to a quadratic programming problem. Simulation results for a two-link manipulator are presented to support the results of this article.

Variations of Biogeography-based Optimization and Markov Analysis

Haiping Ma et al. — Tue, 29 Jan 2013 05:56:32 PST

Biogeography-based optimization (BBO) is a new evolutionary algorithm that is inspired by biogeography. Previous work has shown that BBO is a competitive optimization algorithm, and it demonstrates good performance on various benchmark functions and real-world optimization problems. Motivated by biogeography theory and previous results, three variations of BBO migration are introduced in this paper. We refer to the original BBO algorithm as partial immigration-based BBO. The new BBO variations that we propose are called total immigration-based BBO, partial emigration-based BBO, and total emigration-based BBO. Their corresponding Markov chain models are also derived based on a previously-derived BBO Markov model. The optimization performance of these BBO variations is analyzed, and new theoretical results that are confirmed with simulation results are obtained. Theoretical results show that total emigration-based BBO and partial emigration-based BBO perform the best for three-bit unimodal problems, partial immigration-based BBO performs the best for three-bit deceptive problems, and all these BBO variations have similar results for three-bit multimodal problems. Performance comparison is further investigated on benchmark functions with a wide range of dimensions and complexities. Benchmark results show that emigration-based BBO performs the best for unimodal problems, and immigration-based BBO performs the best for multimodal problems. In addition, BBO is compared with a stud genetic algorithm (SGA), standard particle swarm optimization (SPSO 07), and adaptive differential evolution (ADE) on real-world optimization problems. The numerical results demonstrate that BBO outperforms SGA and SPSO 07, and performs similarly to ADE for the real-world problems.

Kalman Filtering with Statistical State Constraints

Tien Li Chia et al. — Tue, 29 Jan 2013 05:56:31 PST

For linear dynamic systems with white process and measurement noise, the Kalman filter is known to be the minimum variance linear state estimator. In the case that the random quantities are Gaussian, then the Kalman filter is the minimim variance state estimator. However, in the application of Kalman filters known signal information is often either ignored or dealt with heuristically. For instance, state variable constraints (which may be based on physical considerations) are often neglected because they do not fit easily into the structure of the optimal filter. Previous work by the authors demonstrated an analytic method of incorporating deterministic state equality constraints in the Kalman filter. This paper extends that work to develop the properties of Kalman filters in the presence of statistical state constraints. That is, given a linear system such that the expected values of the state variables satisfy some linear equality, we can constrain the Kalman filter estimates to satisfy those constraints. This results in a family of constrained filters with each member parameterized by a weighting matrix. This paper derives several interesting properties of the constrained Kalman filters.

Aircraft Turbofan Engine Health Estimation Using Constrained Kalman Filtering

Daniel J. Simon et al. — Tue, 29 Jan 2013 05:56:31 PST

Fuzzy Membership Function Optimization for System Identification Using an Extended Kalman Filter

Srikiran Kosanam et al. — Fri, 25 Jan 2013 10:21:47 PST

The generation of membership functions for fuzzy systems is a challenging problem. In this paper, we use an extended Kalman filter to optimize the membership functions for system modeling, or system identification. We describe the algorithm and then show the result as sub-optimal novel method of system identification. The ideas described in this paper are illustrated for system identification of a nonlinear dynamic system of a permanent magnet synchronous motor. The other interesting observation made is that the proposed system acts as a noise-reducing filter. We demonstrate that the extended Kalman filter can be an effective tool for identifying the parameters of a fuzzy system model.

Kalman Filtering with Uncertain Noise Covariances

Srikiran Kosanam et al. — Fri, 25 Jan 2013 10:21:46 PST

In this paper the robustness of Kalman filtering against uncertainties in process and measurement noise covariances is discussed. It is shown that a standard Kalman filter may not be robust enough if the process and measurement noise covariances are changed. A new filter is proposed which addresses the uncertainties in process and measurement noise covariances and gives better results than the standard Kalman filter. This new filter is used in simulation to estimate the health parameters of an aircraft gas turbine engine.

A Fault-Tolerant Optimal Interpolative Net

Daniel J. Simon et al. — Fri, 25 Jan 2013 10:21:45 PST

The optimal interpolative (OI) classification network is extended to include fault tolerance and make the network more robust to the loss of a neuron. The OI Net has the characteristic that the training data are fit with no more neurons than necessary. Fault tolerance further reduces the number of neurons generated during the learning procedure while maintaining the generalization capabilities of the network. The learning algorithm for the fault tolerant OI Net is presented in a recursive format, allowing for relatively short training times. A simulated fault tolerant OI Net is tested on a navigation satellite selective problem.

Biogeography-Based Optimization of Neuro-Fuzzy System Parameters for Diagnosis of Cardiac Disease

Mirela Ovreiu et al. — Fri, 25 Jan 2013 10:21:44 PST

Cardiomyopathy refers to diseases of the heart muscle that becomes enlarged, thick, or rigid. These changes affect the electrical stability of the myocardial cells, which in turn predisposes the heart to failure or arrhythmias. Cardiomyopathy in its two common forms, dilated and hypertrophic, implies enlargement of the atria; therefore, we investigate its diagnosis through P wave features. In particular, we design a neuro-fuzzy network trained with a new evolutionary algorithm called biogeography-based optimization (BBO). The neuro-fuzzy network recognizes and classifies P wave features for the diagnosis of cardiomyopathy. In addition, we incorporate opposition-based learning in the BBO algorithm for improved training. First we develop a neuro-fuzzy model structure to diagnose cardiomyopathy using P wave features. Next we train the network using BBO and a clinical database of ECG signals. Preliminary results indicate that cardiomyopathy can be reliably diagnosed with these techniques.

Classificiation of Atrial Fibrillation Prone Patients Using Electrocardiographic Parameters in Neuro-Fuzzy Modeling,

Mirela Ovreiu et al. — Fri, 25 Jan 2013 10:21:43 PST

Atrial Fibrillation (AF) is a significant clinical problem and the complications of cardiovascular postoperative AF often lead to longer hospital stays and higher heath care costs. The literature showed that AF may be preceded by changes in electrocardiogram (ECG) characteristics such as premature atrial activity, heart rate variability (HRV), and P-wave morphology. We hypothesize that the limitations of statistics-based attempts to predict AF occurrence may be overcome using a hybrid neuro-fuzzy prediction model that is better capable of uncovering complex, non-linear interactions between ECG parameters. We created a neuro-fuzzy network that was able to classify the patients into the control and AF groups with the performances: 99.42% sensitivity, 99.89% specificity, and 99.74% accuracy for 30 minutes just before AF onset.

http://www.osti.gov/eprints/topicpages/documents/record/120/1879577.html

Modeling and Analysis of Signal Estimation for Stepper Motor Control

Daniel J. Simon — Fri, 25 Jan 2013 10:21:42 PST

Biogeography-Based Optimization and the Solution of the Power Flow Problem

Rick Rarick et al. — Fri, 25 Jan 2013 10:21:41 PST

Biogeography-based optimization (BBO) is a novel evolutionary algorithm that is based on the mathematics of biogeography. Biogeography is the study of the geographical distribution of biological organisms. In the BBO model, problem solutions are represented as islands, and the sharing of features between solutions is represented as immigration and emigration between the islands. This paper presents an application of the BBO algorithm to the power flow problem for an IEEE 30-bus Test Case system. The BBO solution is compared with the solution of the same problem using a genetic algorithm (GA). The results of Monte Carlo simulations indicate that the BBO algorithm consistently performs better than the GA

Neural Network-Based Robot Trajectory Generation

Daniel J. Simon — Fri, 25 Jan 2013 10:21:40 PST

Interpolation of minimum jerk robot joint trajectories through an arbitrary number of knots is realized using a hardwired neural network. The resultant trajectories are numerical rather than analytic functions of time. This application formulates the interpolation problem as a contrained quadratic minimization problem over a continuous joint angle domain and a discrete time domain. Time is discretized according to the robot controller rate. The neuron outputs define the joint angles. An annealing-type method is used to prevent the network from getting stuck in a local minimum. The optimizing neural network and its application to robot path planning are discussed, some simulation results are presented, and the neural network method is compared with other minimum jerk trajectory planning methods

Oppositional Biogeography-Based Optimization for Combinatorial Problems

Mehmet Ergezer et al. — Fri, 25 Jan 2013 10:21:39 PST

In this paper, we propose a framework for employing opposition-based learning to assist evolutionary algorithms in solving discrete and combinatorial optimization problems. To our knowledge, this is the first attempt to apply opposition to combinatorics. We introduce two different methods of opposition to solve two different type of combinatorial optimization problems. The first technique, open-path opposition, is suited for combinatorial problems where the final node in the graph does not have be connected to the first node, such as the graphcoloring problem. The latter technique, circular opposition, can be employed for problems where the endpoints of a graph are linked, such as the well-known traveling salesman problem (TSP). Both discrete opposition methods have been hybridized with biogeography-based optimization (BBO). Simulations on TSP benchmarks illustrate that incorporating opposition into BBO improves its performance. Index Terms—Biogeography-based optimization, opposition, combinatorics, discrete optimization, evolutionary algorithms, graph-coloring problem, traveling salesman problem.

Truth, American Culture, and Fuzzy Logic

Daniel J. Simon — Fri, 25 Jan 2013 10:21:38 PST

This paper examines the history, relationships, and influences that act between concepts of truth, American culture, and fuzzy logic. We see that postmodernism is largely a reaction against the Western overemphasis on crisp mathematics. This overemphasis started with ancient Greece, but became solidified in Western culture with the Renaissance. At first glance, fuzzy logic seems to tie in nicely with postmodernism. But a closer look reveals that fuzzy logic is actually more similar to modernism, because it is based on rigorous mathematics. However, fuzzy logic does make some concessions to postmodernism by acknowledging the possibility of ambiguity, at least to some extent. Fuzzy logic therefore provides a balance in the cultural war between modernism and postmodernism. In order for fuzzy logic to take advantage of this unique position, the fuzzy logic community needs to pursue three objectives: (1) it must acknowledge its own limitations and avoid overselling itself; (2) it must develop a wider perspective on its role in modern-day culture; and (3) it must be more proactive in defining its role.

Hybrid H₂H_∞ Estimation for Phase-Locked Loop Filter Design

Daniel J. Simon et al. — Fri, 25 Jan 2013 10:21:37 PST

A method of combining Kalman filtering and minimax filtering is proposed and demonstrated in an application to phase-locked loop design. Kalman filtering suffers from a lack of robustness to departures from the assumed noise statistics. But minimax filtering has the drawback of ignoring the engineer's (admittedly incomplete) knowledge of the noise statistics. It is shown in this paper that hybrid Kalman/minimax filtering can provide the best of both worlds. Phase-locked loop filter design is used in this paper to demonstrate an application of hybrid estimation.