• Journal of the Korea Society of Computer and Information
• /
• v.15 no.10
• /
• pp.129-136
• /
• 2010

In this paper proposed the sensor fusion algorithm between a gyroscope and an accelerometer to maintain the inverted posture with two wheels which can make the robot body move to the desired destination. Mobile inverted robot fall down to the forward or reverse direction to converge to the stable point. Therefore, precise information of tilt angles and quick posture control by using the information are necessary to maintain the inverted posture, hence this paper proposed the sensor fusion algorithm between a gyroscope to obtain the angular velocity and a accelerometer to compensate for the gyroscope. Kalman Filter is normally used for the algorithm and numerous research is progressing at the moment. However, a high-performing DSP and systems are needed for the algorithm. This paper realized the robot control method which is much simpler but able to get desired performance by using the sensor fusion algorithm and PID control.

• Journal of IKEEE
• /
• v.23 no.4
• /
• pp.1360-1365
• /
• 2019

In this paper, we describe a sound source localization(SSL) system which can be applied to mobile robot and automatic control systems. Usually the SSL method finds the Interaural Time Difference, the Interaural Level Difference, and uses the geometrical principle of microphone array. But here we proposed another approach based on the deep neural network to obtain the horizontal directional angle(azimuth) of the sound source. We pick up the sound source signals from the two microphones attached symmetrically on both sides of the robot to imitate the human ears. Here, we use difference of spectral distributions of sounds obtained from two microphones to train the network. We train the network with the data obtained at the multiples of 10 degrees and test with several data obtained at the random degrees. The result shows quite promising validity of our approach.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.517-517
• /
• 2000

Welding automation is one of the most important manufacturing issues in shipbuilding in order to lower the cost, increase the quality, and avoid the labor problems. Generally the on-line teaching is utilized on the robot that is used in the welding automation system, but it requires much effort and long time to program. Especially, if the system is composed of more than two cooperating robots, it demands much more skill to program the robots' motion. Thus, a convenient programming tool is required for efficient utilization of welding automation system. In this study, a convenient programming tool is developed for welding automation in which gantry-robot system is used. The system is composed of a gantry transporter and two robots mounted on the gantry to cover the wide work range in the ship building application. As a programming tool, an off-line programming software based on PC is developed. By using this software, field operator does not need to concern about coding of task programs for three control units, one is for gantry and two are for robots. The task programs are automatically generated by assembling the program modules in database according to geometrical information of workpiece and welding condition, which become the only concern of field operator, The feasibility of the generated programs can be verified via a motion simulator previously to on-line running.

• Journal of Service Research and Studies
• /
• v.4 no.1
• /
• pp.49-59
• /
• 2014

Context-aware computing is an emerging paradigm to achieve ubiquitous computing environments by enabling computer systems to understand their situational contexts. A context-aware system uses context to provide relevant information and services to the user depending on the user's task. In this paper, we propose an ontology-based context-aware modeling methodology that transmits low-level contexts acquired by directly accessing various sensors in the physical environments to high-level contexts. With these high-level contexts, context-aware application can provides proactive and intelligent services using ECA (Event-Condition-Action) rules. We implemented a home robot service in smart office environment.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.60 no.7
• /
• pp.1417-1426
• /
• 2011

Radiofrequency ablation through cardiac catheterization is one of minimally invasive intervention procedures used in drug resistant arrhythmia treatment. To facilitate more accurate and precise catheter navigation, systems for robotic cardiac catheter navigation have been developed and commercialized. The authors have been developing a novel robotic catheter navigation system. The system is a network-based master-slave configuration 3-DOF (Degree-Of-Freedom) robotic manipulator for operation with conventional cardiac ablation catheter. The catheter manipulation motion is composed of the translation (forward/backward) and the roll movements of the catheter and knob rotation for the catheter tip articulation. The master manipulator comprises an operator handle compartment for the knob and the roll movement input, and a base platform for the translation movement input. The slave manipulator implements a robotic catheter platform in which conventional cardiac catheter is mounted and the 3-DOF motions of the catheter are controlled. The system software that runs on a realtime OS based PC, implements the master-slave motion synchronization control in the robot system. The master-slave motion synchronization performance tested with step, sinusoidal and arbitrarily varying motion commands showed satisfactory results with acceptable level of steady state error. The developed system will be further improved through evaluation of safety and performance in in vitro and in vivo tests.

• Physical Therapy Korea
• /
• v.28 no.3
• /
• pp.215-226
• /
• 2021

Background: Robot-assisted gait training (RAGT) is an effective method for walking rehabilitation. Additionally, the body weight support (BWS) system reduces muscle fatigue while walking. However, no previous studies have investigated the effects of RAGT with BWS on isokinetic strength of quadriceps and hamstring muscles. Objects: The purpose of this study was to investigate the effects of torque, work, and power on the quadriceps and hamstring muscles during RAGT, using the BWS of three conditions in healthy subjects. The three different BWS conditions were BWS 50%, BWS 20%, and full weight bearing (FWB). Methods: Eleven healthy subjects (7 males and 4 females) participated in this study. The Walkbot_S was used to cause fatigue of the quadriceps and hamstring muscles and the Biodex Systems 4 Pro was used to measure the isokinetic torque, work, and power of them. After RAGT trials of each of the three conditions, the subjects performed isokinetic concentric knee flexion and extension, five at an angular velocity of 60°/s and fifteen at an angular velocity of 180°/s. One-way repeated analysis of variance was used to determine significant differences in all the variables. The least significant difference test was used for post-hoc analysis. Results: On both sides, there were significant differences in peak torque (PT) of knee extension and flexion between the three BWS conditions at an angular velocity of 60°/s and 180°/s conditions. A post-hoc comparison revealed that the PT in the BWS 50% was significantly greater than in the BWS 20% and the FWB and the PT in the BWS 20% was significantly greater than in the FWB. Conclusion: The results of this study suggest that the lower BWS during RAGT seems to lower the isokinetic torque, work, and power of the quadriceps and hamstring muscles because of the muscle fatigue increase.

• Journal of Drive and Control
• /
• v.20 no.2
• /
• pp.15-23
• /
• 2023

In this study, two-dimensional location of crops for auto weeding was detected using deep learning. To construct a dataset for soybean detection, an image-capturing system was developed using a mono camera and single-board computer and the system was mounted on a weeding robot to collect soybean images. A dataset was constructed by extracting RoI (region of interest) from the raw image and each sample was labeled with soybean and the background for classification learning. The deep learning model consisted of four convolutional layers and was trained with a weakly supervised learning method that can provide object localization only using image-level labeling. Localization of the soybean area can be visualized via CAM and the two-dimensional position of the soybean was estimated by clustering the pixels associated with the soybean area and transforming the pixel coordinates to world coordinates. The actual position, which is determined manually as pixel coordinates in the image was evaluated and performances were 6.6(X-axis), 5.1(Y-axis) and 1.2(X-axis), 2.2(Y-axis) for MSE and RMSE about world coordinates, respectively. From the results, we confirmed that the center position of the soybean area derived through deep learning was sufficient for use in automatic weeding systems.

• International Journal of Control, Automation, and Systems
• /
• v.1 no.1
• /
• pp.142-148
• /
• 2003

Reinforcement learning is considered as an important tool for robotic learning in unknown/uncertain environments. In this paper, we propose an evaluation function expressed in a vector form to realize multi-dimensional reinforcement learning. The novel feature of the proposed method is that learning one behavior induces parallel learning of other behaviors though the objectives of each behavior are different. In brief, all behaviors watch other behaviors from a critical point of view. Therefore, in the proposed method, there is cross-criticism and parallel learning that make the multi-dimensional learning process more efficient. By ap-plying the proposed learning method, we carried out multi-dimensional evaluation (reward) and multi-dimensional learning simultaneously in one trial. A special neural network (Q-net), in which the weights and the output are represented by vectors, is proposed to realize a critic net-work for Q-learning. The proposed learning method is applied for behavior planning of mobile robots.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1998.06a
• /
• pp.158-163
• /
• 1998

This paper is our first attempt to construct a information processing system such as the living creatures' brain based on artificial life technique. In this paper, we propose a method of constructing neural networks using bio-inspired emergent and evolutionary concept, Ontogeny of living things is realized by cellular automata model and Phylogeny that is living things adaptation ability themselves to given environment, are realized by evolutionary algorithms. Proposing evolving cellular automata neural systems are calledin a word ECANS. A basic component of ECANS is 'cell' which is modeled on chaotic neuron with complex characteristics, In our system, the states of cell are classified into eight by method of connection neighborhood cells. When a problem is given, ECANS adapt itself to the problem by evolutionary method. For fixed cells transition rule, the structure of neural network is adapted by change of initial cell' arrangement. This initial cell is to become a network b developmental process. The effectiveness and the capability of proposed scheme are verified by applying it to pattern classification and robot control problem.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.5 no.6
• /
• pp.1163-1168
• /
• 2001

The ranging systems measure range data in three-dimensional coordinate from target surface. These non-contact remote ranging systems is widely used in various automation applications, including military equipment, construction field, navigation, inspection, assembly, and robot vision. The active ranging systems using time of flight technique or light pattern illumination technique are complex and expensive, the passive systems based on stereo or focusing principle are time-consuming. The proposed algorithm, that is based on cross correlation of projection profile of vertical edge, provides advantages of fast and simple operation in the range acquisition. The results of experiment show the effectiveness of the proposed algorithm.