• Smart Structures and Systems
• /
• v.8 no.1
• /
• pp.39-52
• /
• 2011

We are developing a biomimetic robot based on the Sea Lamprey. The robot consists of a cylindrical electronics bay propelled by an undulatory body axis. Shape memory alloy (SMA) actuators generate propagating flexion waves in five undulatory segments of a polyurethane strip. The behavior of the robot is controlled by an electronic nervous system (ENS) composed of networks of discrete-time map-based neurons and synapses that execute on a digital signal processing chip. Motor neuron action potentials gate power transistors that apply current to the SMA actuators. The ENS consists of a set of segmental central pattern generators (CPGs), modulated by layered command and coordinating neuron networks, that integrate input from exteroceptive sensors including a compass, accelerometers, inclinometers and a short baseline sonar array (SBA). The CPGs instantiate the 3-element hemi-segmental network model established from physiological studies. Anterior and posterior propagating pathways between CPGs mediate intersegmental coordination to generate flexion waves for forward and backward swimming. The command network mediates layered exteroceptive reflexes for homing, primary orientation, and impediment compensation. The SBA allows homing on a sonar beacon by indicating deviations in azimuth and inclination. Inclinometers actuate a bending segment between the hull and undulator to allow climb and dive. Accelerometers can distinguish collisions from impediment to allow compensatory reflexes. Modulatory commands mediate speed control and turning. A SBA communications interface is being developed to allow supervised reactive autonomy.

• 제어로봇시스템학회:학술대회논문집
• /
• 1993.10a
• /
• pp.233-238
• /
• 1993

In this paper, two diagnosis algorithms using the simple fuzzy, cognitive map (FCM) that is an useful qualitative model are proposed. The first basic algorithm is considered as a simple transition of Shiozaki's signed directed graph approach to FCM framework. And the second one is an extended version of the basic algorithm. In the extension, three important concepts, modified temporal associative memory (TAM) recall, temporal pattern matching algorithm and hierarchical decomposition are adopted. As the resultant diagnosis scheme takes short computation time, it can be used for on-line fault diagnosis of large scale and complex processes that conventional diagnosis methods cannot be applied. The diagnosis system can be trained by the basic algorithm and generates FCM model for every experienced process fault. In on-line application, the self-generated fault model FCM generates predicted pattern sequences, which are compared with observed pattern sequences to declare the origin of fault. In practical case, observed pattern sequences depend on transport time. So if predicted pattern sequences are different from observed ones, the time weighted FCM with transport delay can be used to generate predicted ones. The fault diagnosis procedure can be completed during the actual propagation since pattern sequences of tvo different faults do not coincide in general.

• The Journal of the Acoustical Society of Korea
• /
• v.14 no.1
• /
• pp.13-21
• /
• 1995

In this paper, we improve an image classifier algorithm based on neural network learning. It consists of two steps. The first is input pattern generation and the second, the global neural network implementation using an improved back-propagation algorithm. The feature vector for pattern recognition consists of the codebook data obtained from self-organization feature map learning. It decreases the input neuron number as well as the computational cost. The global neural network algorithm which is used in classifier inserts a control part and an address memory part to the back-propagation algorithm to control weights and unit-offsets. The simulation results show that it does not fall into the local minima and can implement easily the large-scale neural network. And it decreases largely the learning time.

• Journal of information and communication convergence engineering
• /
• v.7 no.2
• /
• pp.119-124
• /
• 2009

In this paper, we propose a novel approach to estimating the real-time moving trajectory of an object is proposed in this paper. The object's position is obtained from the image data of a CCD camera, while a state estimator predicts the linear and angular velocities of the moving object. To overcome the uncertainties and noises residing in the input data, a Extended Kalman Filter(EKF) and neural networks are utilized cooperatively. Since the EKF needs to approximate a nonlinear system into a linear model in order to estimate the states, there still exist errors as well as uncertainties. To resolve this problem, in this approach the Kohonen networks, which have a high adaptability to the memory of the input-output relationship, are utilized for the nonlinear region. In addition to this, the Kohonen network, as a sort of neural network, can effectively adapt to the dynamic variations and become robust against noises. This approach is derived from the observation that the Kohonen network is a type of self-organized map and is spatially oriented, which makes it suitable for determining the trajectories of moving objects. The superiority of the proposed algorithm compared with the EKF is demonstrated through real experiments.

• Journal of IKEEE
• /
• v.24 no.2
• /
• pp.486-491
• /
• 2020

ARM Cortex-M0 DesignStart provided by ARM is cost-free design development suit targeting for designing and prototyping SoC with Cortex-M0 core. In this paper, we presents a method how to implement a custom system design using ARM Cortex-M0 DesignStart. First, hardware elements for ARM Cortex-M0 DesginStart is analyzed focusing on bus and memory map, and next software toolchain is explained to clarify the translating process from high level language to binary machine language. As an example of the custom system, UART system operated with Cortex-M0 is designed and simulated.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.10 no.2
• /
• pp.101-106
• /
• 2010

We propose an approach to estimate the real-time moving trajectory of an object in this paper. The object's position is obtained from the image data of a CCD camera, while a state estimator predicts the linear and angular velocities of the moving object. To overcome the uncertainties and noises residing in the input data, a Extended Kalman Filter(EKF) and neural networks are utilized cooperatively. Since the EKF needs to approximate a nonlinear system into a linear model in order to estimate the states, there still exist errors as well as uncertainties. To resolve this problem, in this approach the Kohonen networks, which have a high adaptability to the memory of the inputoutput relationship, are utilized for the nonlinear region. In addition to this, the Kohonen network, as a sort of neural network, can effectively adapt to the dynamic variations and become robust against noises. This approach is derived from the observation that the Kohonen network is a type of self-organized map and is spatially oriented, which makes it suitable for determining the trajectories of moving objects. The superiority of the proposed algorithm compared with the EKF is demonstrated through real experiments.

• Journal of Korean Society for Geospatial Information Science
• /
• v.17 no.4
• /
• pp.61-66
• /
• 2009

This Paper describes a new program called DEM Generator need to process DEM from LiDAR data or digital map data. It is difficult to generate raster DEM from LiDAR mass point data sets and digital maps too large to fit into memory. The DEM Generator was designed to process DEM and shaded relief image of GeoTiff format in order of streaming meshes; I/O minimize tag, delaunay triangle, natural neighborhood or TIN, temporary files and grid. It is expected that we can be improved the precision of DEM and solved the time consuming problem of DEM generating of a wider area.

• BMB Reports
• /
• v.54 no.6
• /
• pp.295-304
• /
• 2021

Olfactory neuropathology is a cause of olfactory loss in Alzheimer's disease (AD). Olfactory dysfunction is also associated with memory and cognitive dysfunction and is an incidental finding of AD dementia. Here we review neuropathological research on the olfactory system in AD, considering both structural and functional evidence. Experimental and clinical findings identify olfactory dysfunction as an early indicator of AD. In keeping with this, amyloid-β production and neuroinflammation are related to underlying causes of impaired olfaction. Notably, physiological features of the spatial map in the olfactory system suggest the evidence of ongoing neurodegeneration. Our aim in this review is to examine olfactory pathology findings essential to identifying mechanisms of olfactory dysfunction in the development of AD in hopes of supporting investigations leading towards revealing potential diagnostic methods and causes of early pathogenesis in the olfactory system.

• Journal of Korea Multimedia Society
• /
• v.24 no.11
• /
• pp.1552-1559
• /
• 2021

Convolution neural networks (CNNs) show notable performance in image processing and are used as representative core models. CNNs extract and learn features from large amounts of train dataset. In general, it has a structure in which a convolution layer and a fully connected layer are stacked. The core of CNN is the convolution layer. The size of the kernel used for feature extraction and the number that affect the depth of the feature map determine the amount of weight parameters of the CNN that can be learned. These parameters are the main causes of increasing the computational complexity and memory usage of the entire neural network. The most computationally expensive components in CNNs are fully connected and spatial convolution computations. In this paper, we propose a Fourier Convolution Neural Network that performs the operation of the convolution layer in the Fourier domain. We work on modifying and improving the amount of computation by applying the fast fourier transform method. Using the MNIST dataset, the performance was similar to that of the general CNN in terms of accuracy. In terms of operation speed, 7.2% faster operation speed was achieved. An average of 19% faster speed was achieved in experiments using 1024x1024 images and various sizes of kernels.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.3
• /
• pp.755-778
• /
• 2024

In recent years, the number of devices being connected to the internet has grown enormously, as has the intrusive behavior in the network. Thus, it is important for intrusion detection systems to report all intrusive behavior. Using deep learning and machine learning algorithms, intrusion detection systems are able to perform well in identifying attacks. However, the concern with these deep learning algorithms is their inability to identify a suitable network based on traffic volume, which requires manual changing of hyperparameters, which consumes a lot of time and effort. So, to address this, this paper offers a solution using the extended compact genetic algorithm for the automatic tuning of the hyperparameters. The novelty in this work comes in the form of modeling the problem of identifying attacks as a multi-objective optimization problem and the usage of linkage learning for solving the optimization problem. The solution is obtained using the feature map-based Convolutional Neural Network that gets encoded into genes, and using the extended compact genetic algorithm the model is optimized for the detection accuracy and latency. The CIC-IDS-2017 and 2018 datasets are used to verify the hypothesis, and the most recent analysis yielded a substantial F1 score of 99.23%. Response time, CPU, and memory consumption evaluations are done to demonstrate the suitability of this model in a fog environment.