• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.20 no.7
• /
• pp.72-79
• /
• 2021

This study reports on the experiment of using deep learning algorithms to determine the machining process of aluminium and steel. A face cutting milling tool was used for machining and the cutting speed was set between 3 and 4 mm/s. Both materials were machined with a depth to 0.5mm and 1.0mm. To demonstrate the developed deep learning algorithm, simulation experiments were performed using the VGGish algorithm in MATLAB toobox. Downcutting was used to cut aluminum and steel as a machining process for high quality and precise learning. As a result of learning algorithms using audio data, 61%-99% accuracy was obtained in four categories: Al 0.5mm, Al 1.0mm, Steel 0.5mm and Steel 1.0mm. Audio discrimination using deep learning is derived as a probabilistic result.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.20 no.2
• /
• pp.11-25
• /
• 2024

This study aims to analyze research trends in digital twin focusing on Germany, the US, and Korea. In Elsevier's Scopus, we collected 4,657 papers about digital twin published in from 2019 to 2023. Keyword frequency and centrality analysis were conducted on the abstracts of the collected papers. Through the obtained keyword frequencies, we tried to identify keywords with high frequency of occurrence and through centrality analysis, we tried to identify central research keywords for each country. In each country, 'digital_twin', 'machine_learning', and 'iot' appeared as research keywords with the highest interest. As a result of the centrality analysis, research on digital twin, simulation, cyber physical system, Internet of Things, artificial intelligence, and smart manufacturing was conducted as research with high centrality in each country. The implication for Korea is that research on virtual reality, digital transformation, reinforcement learning, industrial Internet of Things, robotics, and data analysis appears to have been conducted with low centrality, and intensive research in related areas appears to be necessary.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.49 no.9
• /
• pp.307-313
• /
• 2012

This paper proposes a hand-eye laser range finder (LRF) based welding plane recognition method for autonomous robotic welding. The robot welding is the process of joining a metal piece and the welding plane along the welding path predefined by the shape of the metal piece. Thus, for successful robotic welding, the position and direction of the welding plane should be exactly detected. If the detected position and direction of the plane is not accurate, the autonomous robotic welding should fail. For precise recognition of the welding plane, a line on the plane is detected by the LRF. For obtaining the line on the plane, the Hough transform is applied to the obtained data from the LRF. Since the Hough transform is based on the voting method, the sensor noise can be reduced. Two lines on the plane are obtained before and after rotation of the robot joint, and then the direction of the plane is calculated by the cross product of two direction vectors of two lines. For verifying the feasibility of the proposed method, the simulation with the robot simulator, RoboticsLab developed by Simlab Co. Ltd., is carried out.

• Transactions on Control, Automation and Systems Engineering
• /
• v.4 no.2
• /
• pp.124-129
• /
• 2002

This paper demonstrates that the largest Lyapunov exponent λ of recurrent neural networks can be controlled efficiently by a stochastic gradient method. An essential core of the proposed method is a novel stochastic approximate formulation of the Lyapunov exponent λ as a function of the network parameters such as connection weights and thresholds of neural activation functions. By a gradient method, a direct calculation to minimize a square error (λ - λ$\^$obj/)$^2$, where λ$\^$obj/ is a desired exponent value, needs gradients collection through time which are given by a recursive calculation from past to present values. The collection is computationally expensive and causes unstable control of the exponent for networks with chaotic dynamics because of chaotic instability. The stochastic formulation derived in this paper gives us an approximation of the gradients collection in a fashion without the recursive calculation. This approximation can realize not only a faster calculation of the gradient, but also stable control for chaotic dynamics. Due to the non-recursive calculation. without respect to the time evolutions, the running times of this approximation grow only about as N$^2$ compared to as N$\^$5/T that is of the direct calculation method. It is also shown by simulation studies that the approximation is a robust formulation for the network size and that proposed method can control the chaos dynamics in recurrent neural networks efficiently.

• International Journal of Control, Automation, and Systems
• /
• v.1 no.3
• /
• pp.339-350
• /
• 2003

In this paper, we propose a simultaneous multithreading (SMT) architecture that improves instruction throughput by exploiting instruction level parallelism (ILP) and thread level parallelism (TLP). The proposed architecture issues and completes instructions belonging to the same thread in exact program order. The issue and completion policy greatly reduces the design complexity and hardware cost of our architecture, compared with others that employ out-of-order issue and completion. On the other hand, when the instructions belong to different threads, the issue and completion orders for those instructions may not necessarily be identical to the fetch order. The processor issues instructions simultaneously from multiple threads to functional units by exploiting ILP and TLP, and by dynamic resource sharing. That parallel execution notably improves performance and resource utilization with minimal additional hardware cost over the conventional superscalar processors. This paper proposes an SMT architecture with grouping as well as one without grouping. Without grouping, all threads dynamically and flexibly share most resources. On the other hand, in the SMT architecture with grouping, in which resources and threads are divided into several groups for design simplification, resources are shared only among threads belonging to the same group as those resources. Simulation results show that our processors with four and eight threads improve performance by three or more times over the conventional superscalar processor with comparable execution resources and policies, and that reasonable grouping reduces the design complexity of SMT processors with little negative effect on performance.

• International Journal of Control, Automation, and Systems
• /
• v.4 no.2
• /
• pp.204-216
• /
• 2006

Latest advances in hardware technology and state of the art of mobile robot and artificial intelligence research can be employed to develop autonomous and distributed monitoring systems. And mobile service robot requires the perception of its present position to coexist with humans and support humans effectively in populated environments. To realize these abilities, robot needs to keep track of relevant changes in the environment. This paper proposes a localization of mobile robot using the images by distributed intelligent networked devices (DINDs) in intelligent space (ISpace) is used in order to achieve these goals. This scheme combines data from the observed position using dead-reckoning sensors and the estimated position using images of moving object, such as those of a walking human, used to determine the moving location of a mobile robot. The moving object is assumed to be a point-object and projected onto an image plane to form a geometrical constraint equation that provides position data of the object based on the kinematics of the intelligent space. Using the a priori known path of a moving object and a perspective camera model, the geometric constraint equations that represent the relation between image frame coordinates of a moving object and the estimated position of the robot are derived. The proposed method utilizes the error between the observed and estimated image coordinates to localize the mobile robot, and the Kalman filtering scheme is used to estimate the location of moving robot. The proposed approach is applied for a mobile robot in ISpace to show the reduction of uncertainty in the determining of the location of the mobile robot. Its performance is verified by computer simulation and experiment.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.1926-1930
• /
• 2004

We suggest a turning gait planning of a quadruped walking robot with an articulated spine. Robot developer has tried to implement a gait more similar to that of natural animals with high stability margin. Therefore, so many types of walking robot with reasonable gait have been developed. But there is a big difference with a natural animal walking motion. A key point is the fact that natural animals use their waist-oint(articulated spine) to walk. For example, a crocodile which has short legs relative to a long body uses their waist to walk more quickly and to turn more effectively. The other animals such as tiger, dog and so forth, also use their waist. Therefore, this paper proposes discontinuous turning gait planning for a newly modeled quadruped walking robot with an articulated spine which connects the front and rear parts of the body. Turning gait is very important as same as straight gait. All animals need a turning gait to avoid obstacle or to change walking direction. Turning gait has mainly two types of gaits; circular gait and spinning gait. We apply articulated spine to above two gaits, which shows the majority of an articulated spine more effectively. Firstly, we describe a kinematic relation of a waist-joint, the hip, and the center of gravity of body, and then apply a spinning gait. Next, we apply a waist-joint to a circular gait. We compare a gait stability margin with that of a conventional single rigid body walking robot. Finally, we show the validity of a proposed gait with simulation.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.1706-1710
• /
• 2004

This paper presents a feedback-controlled, MEMS-fabricated microaccelerometer (${\mu}$XL). The ${\mu}$XL has received much commercial attraction, but its performance is generally limited. To improve the open-loop performance, a feedback controller is designed and experimentally evaluated. The feedback controller is applied to the x/y-axis ${\mu}$XL fabricated by sacrificial bulk micromachining (SBM) process. Even though the resolution of the closed-loop system is slightly worse than open-loop system, the bandwidth, linearity, and bias stability are significantly improved. The noise equivalent resolution of open-loop system is 0.615 mg and that of closed-loop system is 0.864 mg. The bandwidths of open-loop and closed-loop system are over 100 Hz. The input range, non-linearity and bias stability are improved from ${\pm}$10 g to ${\pm}$18 g, from 11.1 %FSO to 0.86 %FSO, and from 0.221 mg to 0.128 mg by feedback control, respectively

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.1066-1070
• /
• 2003

Recently technologies have created new principle and theory but the PID control system remains its popularity as the PID controller contains simple structure, including maintenance and parameter adjustment being so simple. Thus, this paper proposes auto tune PID by fuzzy logic controller based on FPGA which to achieve real time and small size circuit board. The digital PID controller design to consist of analog to digital converter which use chip TDA8763AM/3 (10 bit high-speed low power ADC), digital to analog converter which use two chip DAC08 (8 bit digital to analog converters) and fuzzy logic tune digital PID processor embedded on chip FPGA XC2S50-5tq-144. The digital PID processor was designed by fundamental PID equation which architectures including multiplier, adder, subtracter and some other logic gate. The fuzzy logic tune digital PID was designed by look up table (LUT) method which data storage into ROM refer from trial and error process. The digital PID processor verified behavior by the application program ModelSimXE. The result of simulation when input is units step and vary controller gain ($K_p$, $K_i$ and $K_d$) are similarity with theory of PID and maximum execution time is 150 ns/action at frequency are 30 MHz. The fuzzy logic tune digital PID controller based on FPGA was verified by control model of level control system which can control level into model are correctly and rapidly. Finally, this design use small size circuit board and very faster than computer and microcontroller.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.1654-1659
• /
• 2003

In this paper, the performance evaluation of the JADE-MUSIC estimation based on the indoor channel is presented. By means of the JADE-MUSIC algorithm, DOA and time delay can be obtained simultaneously. In the JADE-MUSIC method, the channel impulse response is first estimated from the received samples and then this impulse response is employed to estimate DOAs and time delays of multipath waves. Moreover, according to the JADE-MUSIC characteristics, it can work in cases when the number of impinging waves is more than the number of antenna elements, unlike the traditional parametric subspace-based method, such a case is not true. Therefore, we employ the JADE-MUSIC algorithm applying for the real indoor environment where is rich of the multipath propagation waves and can imply that the number of waves is very possibly higher than that of the array element. The experiment is carried out in our laboratory considered to be the real indoor environment. The performance of the JADE-MUSIC algorithm is evaluated in terms of the comparison between the simulation and experiment results by using the simulated channel model and the real indoor channel model, respectively. It is clear that the joint angle and delay estimation using the simulated channel model are in good agreement with the estimation using the real indoor channel model. Therefore, we can say that the JADE-MUSIC algorithm accomplishes the high performance to jointly estimate the angle and delay of the arriving signal for the indoor environment.