• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10b
• /
• pp.1476-1479
• /
• 1996

The ring rolling process is one of the best known ring production method. The present model based control system was designed for rings with rectangle cross-section yet. An Adaptive Fuzzy Control for Closed-Die Ring-Rolling was developed in order to enhance the flexibility of the radial-axial ring rolling machine and to produce the rings with highly complex cross-section profile, roller bearing rings. A fuzzy method was implemented because of its simple application and to utilize the known process knowledge. The quality of the control system was estimated by die filling grad, which is strong dependent on the rising time of the controller. The rolling process parameters were also varied to determine their influence on filling of the ring profile. Die filling met the requirement of the industry.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.11
• /
• pp.4268-4289
• /
• 2020

Advanced traffic signal timing method plays very important role in reducing road congestion and air pollution. Reinforcement learning is considered as superior approach to build traffic light timing scheme by many recent studies. It fulfills real adaptive control by the means of taking real-time traffic information as state, and adjusting traffic light scheme as action. However, existing works behave inefficient in complex intersections and they are lack of feasibility because most of them adopt traffic light scheme whose phase sequence is flexible. To address these issues, a novel adaptive traffic signal timing scheme is proposed. It's based on actor-critic reinforcement learning algorithm, and advanced techniques proximal policy optimization and generalized advantage estimation are integrated. In particular, a new kind of reward function and a simplified form of state representation are carefully defined, and they facilitate to improve the learning efficiency and reduce the computational complexity, respectively. Meanwhile, a fixed phase sequence signal scheme is derived, and constraint on the variations of successive phase durations is introduced, which enhances its feasibility and robustness in field applications. The proposed scheme is verified through field-data-based experiments in both medium and high traffic density scenarios. Simulation results exhibit remarkable improvement in traffic performance as well as the learning efficiency comparing with the existing reinforcement learning-based methods such as 3DQN and DDQN.

• Journal of Information Processing Systems
• /
• v.16 no.6
• /
• pp.1424-1436
• /
• 2020

Although huge progress has been made in current image segmentation work, there are still no efficient segmentation strategies for tree image which is taken from natural environment and contains complex background. To improve those problems, we propose a method for tree image segmentation combining adaptive mean shifting with image abstraction. Our approach perform better than others because it focuses mainly on the background of image and characteristics of the tree itself. First, we abstract the original tree image using bilateral filtering and image pyramid from multiple perspectives, which can reduce the influence of the background and tree canopy gaps on clustering. Spatial location and gray scale features are obtained by step detection and the insertion rule method, respectively. Bandwidths calculated by spatial location and gray scale features are then used to determine the size of the Gaussian kernel function and in the mean shift clustering. Furthermore, the flood fill method is employed to fill the results of clustering and highlight the region of interest. To prove the effectiveness of tree image abstractions on image clustering, we compared different abstraction levels and achieved the optimal clustering results. For our algorithm, the average segmentation accuracy (SA), over-segmentation rate (OR), and under-segmentation rate (UR) of the crown are 91.21%, 3.54%, and 9.85%, respectively. The average values of the trunk are 92.78%, 8.16%, and 7.93%, respectively. Comparing the results of our method experimentally with other popular tree image segmentation methods, our segmentation method get rid of human interaction and shows higher SA. Meanwhile, this work shows a promising application prospect on visual reconstruction and factors measurement of tree.

• ETRI Journal
• /
• v.37 no.1
• /
• pp.43-53
• /
• 2015

In this paper, a selective coefficient updating (SCU) approach at each branch of the per-tone equalization (PTEQ) structure has been applied for insufficient cyclic prefix (CP) length. Because of the high number of adaptive filters and their complex adaption process in the PTEQ structure, SCU has been proposed. Using this method leads to a reduction in the computational complexity, while the performance remains almost unchanged. Moreover, the use of set-membership filtering with variable step size is proposed for a sufficient CP case to increase convergence speed and decrease the average number of calculations. Simulation results show that despite the aforementioned algorithms having similar performance in comparison with conventional algorithms, they are able to reduce the number of calculations necessary. In addition, compensation of both the channel effect and the transmitter/receiver in-phase/quadrature-phase imbalances are achievable by these algorithms.

• Journal of Electrical Engineering and Technology
• /
• v.10 no.6
• /
• pp.2393-2405
• /
• 2015

This paper considers a decentralized multiple faults detection and isolation (FDI) scheme for reconfigurable manipulators. Inspired by their modularization property, a global sliding mode (GSM) based stable adaptive fuzzy decentralized controller is investigated for the system in fault free, while for the system suffering from multiple faults (actuator fault and sensor fault), the decentralized sliding mode observer (DSMO) is employed to detect their occurrence. Hereafter, the time and location of faults can be determined by a fault isolation scheme via a bank of DSMOs. Finally, the effectiveness of the proposed schemes in controlling, detecting and isolating faults is illustrated by the simulations of two 3-DOF reconfigurable manipulators with different configurations successfully.

• 제어로봇시스템학회:학술대회논문집
• /
• 1993.10a
• /
• pp.113-118
• /
• 1993

This paper represents mechanical compliance & ZMP(Zero Moment Point) control algorithm for IWR(Inha Walking Robot) system. In case of walking in different environments, a biped walking robot must vary its gait(walking period or step length, etc.) according to the environments. However, most of biped walking robots do not have the capability to change their gaits or need more complex control algorithm, because ZMP cannot be defined in their control algorithm. Therefore new linear type with balancing joint is proposed which is used as an aid in balancing & ZMP control itself. In IWR system, ZMP can be defined by solving differential equations and it does not need to be predefined ZMP trajectory. Furthermore we can input the desired ZMP position. In parallel with the development, we also considered a mechanical compliance for reducing the inverse kinematics, dynamics and the control complexity. It will figure out some powerful adaptation with 3D irregular terrains.

• 제어로봇시스템학회:학술대회논문집
• /
• 1994.10a
• /
• pp.257-262
• /
• 1994

This paper describes an autotuning fuzzy PID controller for a position control of DC serve motor. Because ZNM(Ziegler-Nichols Method) with relay feedback has the difficulty in re-tuning the PID parameters and adaptive method has complex algorithm, a new method to overcome those problems is required. The proposed scheme determines the initial PID gains by using ZNM with relay feedback, and then re-tunes the optimal PID parameters by using fuzzy expert system whenever control conditions are changed. To show the validity of the proposed method, a position control of DC servo motor is illustrated by computer simulation and is experimented by a designed controller.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.1628-1631
• /
• 2005

Ship Autopilots are usually designed based on the PD and Pill controllers because of simplicity, reliability and easy to construct. However their performance in various environmental conditions is not as good as desired. This disadvantage can be overcome by adjusting works or constructing adaptive controllers. But those methods are complex and not easy to do. This paper presents a new method for constructing a Ship Autopilot based on the combination of Fuzzy Logic Control (FLC) and Linear Control Theory (Pill control). The new Ship Autopilot has the advantages of both the Pill and FLC control methodologies: easy to construct, and optimal control laws can be established based on ship masters' knowledge. Therefore, the new ship autopilot can be well adapted with parameter variations and strong environment effects. Simulation using MATLAB software for a ship with real parameters shows high effectiveness of the Fuzzy Pill autopilot in course keeping and course changing manoeuvres in comparison with the ordinary Pill ship autopilots.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.124.1-124
• /
• 2001

Control of pH neutralization process plays a very important role in some chemical process. Because of their high nonlinearity, frequent disturbance, and time-varying characteristics, it is difficult to control and estimate pH processes. For the adaptive control of pH processes, a lot of researchers have made an efforts in the modeling and control of pH processes. It is very difficult to obtain information of influent stream such as concentrations and dissociation constants and the titration curve equation is very complex. Therefore, several simple models, which hate small number of unknown parameters and estimate the titration curve, have been available, These models were considered here and were transformed into forms that can applied the linear least square method.

• 제어로봇시스템학회:학술대회논문집
• /
• 1990.10b
• /
• pp.1173-1177
• /
• 1990

Cerebellar Model Linear Associator Net(CMLAN), a kind of neuro-net based adaptive control function generator, was applied to the problem of direct inverse calibration of three and six d.o.f. POMA 560 robot. Since CMLAN autonomously maps and generalizes a desired system function via learning on the sampled input/output pair nodes, CMLAN allows no knowledge in system modeling and other error sources. The CMLAN based direct inverse calibration avoids the complex procedure of identifying various system parameters such as geometric(kinematic) or nongeometric(dynamic) ones and generates the corresponding desired compensated joint commands directly to each joint for given target commands in the world coordinate. The generated net outputs automatically handles the effect of unknown system parameters and dynamic error sources. On-line sequential learning on the prespecified sampled nodes requires only the measurement of the corresponding tool tip locations for three d.o.f. manipulator but location and orientation for six d.o.f. manipulator. The proposed calibration procedure can be applied to any robot.