Intelligent Control and Computer Engineering

A new method of intelligent control for closed quantum computation time-independent systems is introduced. The introduced method uses recurrent supervised neural computing to identify certain parameters of the transformed system matrix \( [\tilde{\mathbf{A}}] \). Linear matrix inequality (LMI) is then used to determine the permutation matrix [P] so that a complete system transformation \(\{[\tilde{\mathbf{B}}], [\tilde{\mathbf{C}}], [\tilde{\mathbf{D}}]\}\) is achieved. The transformed model is then reduced using singular perturbation and state feedback control is implemented to enhance system performance. In quantum computation and mechanics, a closed system is an isolated system that can’t exchange energy or matter with its environment and doesn’t interact with other quantum systems. In contrast to an open quantum system, a closed quantum system obeys the unitary evolution and thus is information lossless that implies state reversibility. The experimental simulations show that the new hierarchical control simplifies the model of the quantum computing system and thus uses a simpler controller that produces the desired performance enhancement and system response.

This paper deals with a control of space robot for capturing moving targets. It would be desirable to use the space robot to repair the failed satellite and to remove space debris since the work load to do these tasks by astronauts will be extremely heavy. Extensive studies have been done for the control of space robot. Unfortunately these studies have not incorporated the orbital motion which is essential for space robot. Coplanar motion between space robot and target is discussed in this study. Suboptimal control, which uses piecewise optimized feedback gain by optimal tracking control method, is applied to chase the target. Also Hill’s equation was applied to the relative orbital equations of motions. Based on the above formulation dynamical simulation was conducted to demonstrate the validity of our approach.

This paper investigates the application of Genetic Algorithms (GA) in the design and implementation of Fuzzy Logic Controllers (FLC) for temperature control in an extruder. The importance of FLC is during the process of selecting the membership functions. What is best to determine the membership functions is the first question that has be addressed. It is important therefore to select accurate membership functions but these methods possess one common weakness where conventional FLC use membership functions generated by human operators. In this situation the membership function selection process is done by trial and error and it runs step by step which is too long to arrive at a solution to the problem. This research proposes a method that may help users to determine the membership functions of FLC using GA optimization for the fastest process in solving problems. The data collection is based on simulation results and the results refer to the maximum overshoot. From the results presented, the system arrives at better and more exact results and the value of overshoot is decreased from 1.2800 for FLC without GA, to 1.0011 for FGA.

In this paper, a new technique for designing a robust cascade controller for a quadratic boost converter is proposed. A single performance index in H infinity loop shaping control called stability margin is adopted as the objective function in the proposed optimization control problem; GA is used to solve this problem to evaluate the optimal controller. Necessary conditions of cascade control scheme are adopted as the constraints in the optimization problem. In addition, pre-compensator weight, which is normally difficult to be selected, is simultaneously determined with the controller. Comparative study with the conventional H infinity loop shaping is presented. Finally, simulation results verify the effectiveness of the proposed technique.

The need of an increased availability in the field of robotics is essential. The subjects of this work are the wheeled mobile robots, and their key feature is the movement. The movement has to be realized as safely as possible. The safety implies faultless behavior, which means achieving the specification limits in any situation. A critical part in the wheeled mobile robots movement is the localization module. The first objective of the paper is to provide an availability analysis for wheeled mobile robots regarding the localization module. The first analysis focuses on a distance sensor used to detect static or dynamic obstacles from the environment and the second analysis is related to the entire localization module, the main part of the faults coming from odometry errors. The second objective of the paper is to realize a synthesis of redundant procedures meant to increase the value of the overall availability of the mobile robots.

This chapter proposes a new method of electric motor fault quantification. Higher Order Spectrum (HOS) is a signal processing used as a fault quantification technique. Previous researches have shown that the faults in the stator or rotor generally show sideband frequencies around the main frequency (50 Hz) and its higher harmonics in the spectrum of the Motor Current Signature Analysis (MCSA). However in the present experimental studies such observations are not seen, but the faults in the stator or the rotor may distort the sinusoidal response of the motor RPM and the main frequency. Hence this research proposes the HOS here, namely the Bispectrum of the MCSA, because it relates to both amplitude and phase of number of harmonics in a signal. The Bispectrum with the unwrapped phase angle along its frequency is also analyzed. The tests can show that the proposed method can detect the faults accurately. The proposed method can also show that the severity level of the faults can be measured by observing the change in the heights of the Bispectrum amplitude.

In this paper we propose a newly multiple particle swarm optimizers with diversive curiosity (MPSOα/DC) for enhancing the search performance. It has three outstanding features: (1) Implementing plural particle swarms in parallel to explore; (2) Finding the most suitable solution in a small limited space by a localized random search for correcting the solution found by each particle swarm; (3) Introducing diversive curiosity into the multi-swarm to alleviate stagnation. To demonstrate the proposal’s effectiveness, computer experiments on a suite of benchmark problems are carried out. We investigate its intrinsic characteristics, and compare the search performance with other methods. The obtained results show that the search performance of the MPSOα/DC is superior to that by the PSO/DC, EPSO, OPSO, and RGA/E for the given benchmark problems.

In this contribution GPTIPS, a free, open source MATLAB toolbox for performing symbolic regression by genetic programming (GP) is introduced. GPTIPS is specifically designed to evolve mathematical models of predictor response data that are “multigene” in nature, i.e. linear combinations of low order non-linear transformations of the input variables. The functionality of GPTIPS is demonstrated by using it to generate an accurate, compact QSAR (quantitative structure activity relationship) model of existing toxicity data in order to predict the toxicity of chemical compounds. It is shown that the low-order “multigene” GP methods implemented by GPTIPS can provide a useful alternative, as well as a complementary approach, to currently accepted empirical modelling and data analysis techniques. GPTIPS and documentation is available for download at http://​sites.​google.​com/​site/​gptips4matlab/​.

Pareto-based multi-objective optimization problems (MOPs) are currently best solved using evolutionary algorithms. Nevertheless, the performance of these nature-inspired stochastic search algorithms still depends on the suitability of their parameter settings with respect to specific optimization problems. The tuning of the parameters is a crucial task which concerns resolving the contrary goals of convergence and diversity. To address this issue, we propose a diversity-driven self-adaptive mechanism (SAM) for the simulated binary crossover. This novel technique exploits and optimizes the balance between exploration and exploitation during the evolutionary process. This “explore first and exploit later” approach is addressed through the automated and dynamic adjustment of the distribution index of the simulated binary crossover (SBX) operator. We conducted a series of experiments where SAM is applied to the Non-dominated Sorting Genetic Algorithm to solve the Sphere, Rastrigin, and ZDT optimization problems. Our experimental results have shown that our proposed self adaptation mechanism can produce promising results for both single and multi-objective problem sets.

In this paper, a new Q-Learning method for transfer scheduling of freight cars in a train is proposed. In the proposed method, the number of freight-movements in order to line freights in the desired order is reflected by corresponding evaluation value for each pair of freight-layout and removal-destination at a freight yard. Evaluation values are obtained by the Q-Learning method. The best transfer scheduling can be derived by selecting the removal-action of freight that has the best evaluation value at each freight-layout.

In this paper, we consider the negotiation between two competitive agents that consider both time and cost criteria. Therefore, the negotiation agents are designed to not only optimize price utility but also be successful in optimizing (negotiation) speed utility. To this end, the objective of this work is to find effective strategies for the negotiation. The strategies are coevolved through an evolutionary learning process using two different evolutionary algorithms (EAs)—a genetic algorithm (GA) and an estimation of distribution algorithm (EDA). We present an empirical comparison of GA and EDA in coevolving negotiation strategies with different preference criteria in optimizing the price and (negotiation) speed. The experimental results show that both EAs are successful in finding good solutions with respect to both the price-optimizing (P-Optimizing) and the speed-optimizing (S-Optimizing) negotiation. However, both EAs are not effective in the negotiation for the concurrent optimization of the price and speed (P-S-Optimizing negotiation). This is because in some cases, the original fitness function cannot characterize the difference among P-Optimizing, S-Optimizing, and P-S-Optimizing solutions. Hence, this paper proposes a new fitness function that can better differentiate among the P-Optimizing, S-Optimizing, and P-S-Optimizing solutions. The experiments showed that the EAs using the proposed fitness function can coevolve effective strategies for the exact P-S-Optimizing negotiation.

This research develops a learning method for the pass selection problem of midfielders in RoboCup Soccer Simulation games. A policy gradient method is applied as a learning method to solve this problem because it can easily represent the various heuristics of pass selection in a policy function. We implement the learning function in the midfielders’ programs of a well-known team, UvA Trilearn Base 2003. Experimental results show that our method effectively achieves clever pass selection by midfielders in full games. Moreover, in this method’s framework, dribbling is learned as a pass technique, in essence to and from the passer itself. It is also shown that the improvement in pass selection by our learning helps to make a team much stronger.

The benefits of buyer coalitions are well-known for electronic marketplaces. However, a few existing buyer coalition schemes over the Internet have focused on forming a buyer coalition with bundles of items. This paper presents an algorithm to form a buyer coalition with bundles of items by using genetic algorithms (GAs). The algorithm called GAGroupBuying finds the best disjoint subsets of all buyers based on the total utility which addresses the situation where a whole group of buyers can be partitioned into smaller sub-groups to obtain more utility than they could accomplish in the whole group. The proposed algorithm is compared with a previous algorithm called GroupPackageString as shown by Boongasame and Sukstrienwong (Emerging Intelligent Computing Technology and Applications, pp. 674–685, 2009). The results of GAGroupBuying simulation are found to be satisfactory with the total discount of a buyer coalition.

Worldwide cooperation among manufacturing companies is increasingly gaining importance to face emerging challenges in manufacturing. The traditional Computer Integrated Manufacturing (CIM) systems cannot satisfy the needs of global market as they are deployed only within an enterprise. Therefore, a more flexible and comprehensive integrating methodology is required to overcome distance barriers, facility sharing problems and communication obstacles. These issues lead to the concept of Virtual CIM (VCIM). In this paper, the limitations of current agent based implementation of VCIM concept are analyzed. It also describes approaches to address those limitations and propose further development on Agent based Resource Scheduling Process in Small and Medium Enterprises in VCIM network.

This paper presents an improvement of our approach called SCRO_II algorithm. SCRO_I algorithm was initially developed in order to retrieve a set of Supreme Court sentences. The goal of SCRO_I is to provide different law issues among those retrieved documents. We create a new ontology using different semantics to study their performances based on diversity measurement. The contribution of this new ontology is compared to the traditional one. A new procedure is embedded in SCRO_II algorithm to identify a set of synonyms and relations. The experiments were done on Thai Succession Law and Bill of Exchange Law. The experimental results show that SCRO_II outperforms SCRO_I algorithm in both data sets.

One central problem of information retrieval is to determine the relevance of documents with respect to the user information needs. The choice of similarity measure is crucial for improving search effectiveness of a retrieval system. Different similarity measures have been suggested to match the query and documents. Some of the popular measures being: cosine, jaccard, dice, okapi etc., each having their own pros and cons. Accordingly one may give better result over other depending on users need, document corpus, organization and indexing of corpus. Therefore it may be justifiable to combine these measures and develop a new similarity measure which can be named as combined similarity measure. Now individual measures can be assigned weights in different proportion in combined similarity measure. In order to optimize ranking of relevant documents, individual weights have to be optimized. In this chapter we suggest a genetic algorithm based model for learning weights of individual components of combined similarity measure. We have considered two different types of functions viz: non-order based and order based fitness functions to evaluate the goodness of the solution. A non-order based fitness function is based on recall-precision values only. However, it has been observed that a better fitness function can be obtained if we also consider the order in which relevant documents are retrieved. This leads to an idea of order based fitness functions. We evaluated the efficacy of a genetic algorithm with various fitness functions. The experiments have been carried out on TREC data collection. The results have been compared with various well-known similarity measures.

This paper presents a Cloud service discovery system (CSDS) that aims to support Cloud users in finding a Cloud service over the Internet. The CSDS interacts with a Cloud ontology to determine the similarities among services. The significance of this project is that it is the first attempt in building an agent-based discovery system that consults an ontology when retrieving information about Cloud services. One of the main contributions of this work is building a Cloud service reasoning agent (CSRA) that enables the CSDS to (1) reason about the relations of Cloud services and (2) rate the search results. Another contribution of this work is designing and constructing a Cloud ontology consisting of a taxonomy of concepts of Cloud services that enables the CSRA to determine the relations of Cloud services using three service reasoning methods: (1) similarity reasoning, (2) equivalent reasoning, and (3) numerical reasoning. Empirical results show that using the Cloud ontology, the CSDS is more successful in finding Cloud services that are closer to users’ requirements. Experiments are also conducted to examine the effect of using different combinations of the three service reasoning method: (1) using only similarity reasoning, (2) using similarity reasoning and equivalent reasoning, and (3) using all three reasoning methods. Additionally, a proof-of-concept example demonstrates the major functionalities of the CSDS.

This paper introduces a method of navigation in a large family of tilings of the hyperbolic plane and looks at the question of possible applications in the light of the few ones which were already obtained. (This paper is a revised and slightly extended version of a paper presented by the author at IMECS’2010, see Margenstern (Proceedings of the International MultiConference of Engineers and Computer Scientists 2010, pp. 367–382, 2010).)

In this chapter, I present a new method called Setvectors to predict cache contention introduced by co-scheduled applications on a multicore processor system. Additionally, I propose a new metric to compare cache contention prediction methods. Applying this metric, I demonstrate that the Setvector method predicts cache contention with about the same accuracy as the most accurate state-of-the-art method. However, the Setvector method executes nearly 4000 times as fast. This chapter is a revised and extended version of Zwick et al. (Proceedings of the International MultiConference of Engineers and Computer Scientists 2010, pp. 244–251, 2010), presented at the MultiConference of Engineers and Computer Scientists in Hong Kong.

A material model to describe large deformation of pressure sensitive adhesive (PSA) is presented. A relationship between stress and strain of PSA includes viscoelasticity and rubber-elasticity. Therefore, we propose the material model for describing viscoelasticity and rubber-elasticity and formulate the presented material model for finite element analysis. And we validate the present formulation by using one axis tensile calculation.

Ethernet Passive Optical Networks (EPONs) are designed to deliver multiple services and applications, such as voice communications, standard (SDTV) and high-definition video (HDTV). To support these applications and their various requirements, EPONs require Quality-of-Service (QoS) mechanisms to be built in service. For this purpose, a scalable Interleaved Dynamic Bandwidth Allocation (IDBA) mechanism for sharing the uplink bandwidth among optical network units (ONUs) is proposed in this paper. The modus operandi of IDBA is to divide the cycle time by partitioning the ONUs into two groups with some timing overlap to execute interleaved bandwidth allocation, which cooperates with Limited Bandwidth Allocation (LBA), Excess Bandwidth Reallocation (EBR) and accurate prediction mechanism in EPONs. The proposed IDBA mechanism has two advantages, namely it eliminates the idle period problem in the traditional DBA mechanism, and guarantees QoS services by dynamically adjusting the bandwidth within the group of subscribers. This will not only support the differentiated services architecture but also offer various QoS levels. Simulation results obtained show that the proposed IDBA mechanism achieves desirable system performance relative to packet delay, jitter performance, throughput, ratio of packet loss and fairness.

Why are n-player games much more complex than two-player games? Is it much more difficult to cooperate or to compete? The game of n-player Shove is the n-player version of Shove, a two-player combinatorial game. In multi-player games, because of the possibility to form alliances, cooperation between players is a key-factor to determine the winning coalition and, as a consequence, n-player Shove played on a set of finite strips is \(\mathcal{PSPACE}\)-complete.

In recent years the vestibular-sympathetic reflex has received an increasing amount of attention due to the role it could play in the human organism in different types of scenarios. Despite this, quantitative models of this reflex mechanism are still lacking. In this context, the current paper aims at taking a first step towards the modeling of the vestibular-sympathetic reflex by developing a model of the Vestibular Nucleus – the central part of the vestibular-sympathetic reflex. After a careful analysis of the limitations and uncertainties of the available experimental data from the literature, a three step modeling methodology for the Vestibular Nucleus is presented. After a description of the operations involved in each step, in the end, some preliminary results are shown.

This is a review paper of our quest in developing and implementing an automated three-dimensional (3D) lung delineation method capable of handling single photon emission computed tomography (SPECT) lung scans with defective contours and/or varying maximum count value (MCV) and total count value (TCV). Six clinically significant datasets consisting of simulations and real subject scans are used consistently throughout our studies. We first develop a dynamic thresholding method which allows removal of background noise in a 3D volumetric fashion. Next, we implement 3D image processing techniques to enhance the SPECT lung contours. Finally, we develop 3D active contours to perform actual delineation. Quantitative validation using known-volume simulations and qualitative verification via experienced physicians are done to evaluate the methods. We achieve over 90% agreement on average throughout all six datasets.