• 우주기술과 응용
• /
• 제4권1호
• /
• pp.27-47
• /
• 2024

뉴스페이스(New Space) 시대에 이르러 궤도상 서비싱(OOS, on orbit servicing) 임무를 수행하기 위한 인공위성들이 개발되고 있다. 궤도상 서비싱을 위한 다양한 임무는 고장수리, 재급유, 견인, 구성품 교체, 우주 상 건설 등의 여러 임무가 있으며, 이를 수행하기 위해 로봇팔 탑재체가 탑재되어야 한다. 로봇팔 탑재체는 기존 인공위성의 탑재체와 달리 고정된 상태로 움직이지 않는 것이 아니라 임무 수행을 위해 지속적으로 움직여야 하는 탑재체라는 특징이 있으며, 또한 인공위성의 구조체 내부에 존재하는 것이 아닌 우주공간에 직접적으로 노출된 상태로 임무 수행을 해야 한다는 특징이 있다. 이러한 탑재체의 특징으로 인해 극한의 우주 열환경에서 이상 없이 운용될 수 있는 열 설계 및 해석은 필수적이나, 로봇팔 열 설계 및 해석에 대한 논문은 그리 많지 않은 실정이다. 본 논문에서는 현재까지 개발된 로봇팔 탑재체에 대한 열 설계, 해석에 대한 사례들을 소개 및 정리하였으며, 마지막에는 앞으로 개발할 로봇팔 탑재체의 열 설계 및 해석에 대한 방향을 제시해 보고자 한다.

• 로봇학회논문지
• /
• 제11권4호
• /
• pp.293-299
• /
• 2016

In this article art performing applications of industrial dual-arm robots are introduced. It was real collaboration among robot researchers and artist. Artist designed the performance to use dual-arm robot. Robot researchers collaborated with artist by providing robotic constraints and configuring robot motion. Two art performances were configured with two industrial dual-arm robots. In both performance robots carry objects to be used as moving screens. Both performances rely on the high power and high precision of robots. In addition human-like appearance make those performances be familiar to public.

• 전기학회논문지
• /
• 제65권7호
• /
• pp.1270-1276
• /
• 2016

This paper proposes the controller design for regulation control of two-link robot arm using sum of squares (SOS) control method that takes into account the input constraint. The existing studies of two link robotic arm system used a linear model of all the non-linearity of the system is linearized. For a linear controller, since the model of the system is simplified, it is possible to design a controller in consideration of constraints on the disturbance. However, there is a limit to the performance using a linearized model for a system with a complex nonlinear properties. To compensate for this in the case of using a fuzzy LMI method, it is necessary to have a large number of linear models and thus there is a disadvantage that the system becomes complicated. To solve these problems, we represents a two-link robot arm system with a polynomial model using a Taylor series expansion and design the controller considering the case where the magnitude of the control input is limited using SOS method. We demonstrate by simulations the feasibility of the proposed algorithm.

• Physical Therapy Rehabilitation Science
• /
• 제8권2호
• /
• pp.93-98
• /
• 2019

Objective: The purpose of this study was to investigate the effect of robot arm reach training on upper extremity functional movement in chronic stroke survivors. Design: One group pretest-posttest design. Methods: Thirteen chronic stroke survivors participated in this study. Robot arm reach training was performed with a Whole Arm Manipulator (WAM) and a 120-inch projective display to provide visual and auditory feedback. During the robotic arm reach training, WAM provided gravity compensation and assist-as-needed (AAN) force according to the robot control mode. When a participant could not move the arm toward the target for more than 2 seconds, WAM provided AAN force to reach the desired targets. All patients participated in the training for 40 minutes per day, 3 times a week, for 4 weeks. Main outcome measures were the Fugl-Meyer Assessment (FMA), Action Research Arm Test (ARAT) and Box and Block Test (BBT) to assess upper extremity functional movement. Results: After 4 weeks, significant improvement was observed in upper extremity functional movement (FMA: 42.15 to 46.23, BBT: 12.23 to 14.00, p<0.05). In the subscore analysis of the FMA upper extremity motor function domains, significant improvement was observed in upper extremity and coordination/speed units (p<0.05). However, there were no significant differences in the ARAT. Conclusions: This study showed the positive effects of robot arm reach training on upper extremity functional movement in chronic stroke survivors. In particular, we confirmed that robot arm reach training could have a positive influence by leading to improvement of motor recovery of the proximal upper extremity.

• 한국전자통신학회논문지
• /
• 제19권1호
• /
• pp.85-92
• /
• 2024

본 논문에서는 로봇팔의 선단을 잡고 원하는 위치로 이동시켜서 작업을 직접 교시하는 직감적인 교시 및 제어를 시스템을 개발하였다. 개발된 시스템은 로봇팔 선단부의 위치 방향 및 자세 방향 힘을 측정하는 6축힘 센서, 선단부에서 측정된 힘에 의한 로봇팔 관절 속도제어 명령어 생성 알고리즘, 자체 제작한 6축 로봇팔 및 제어 시스템으로 구성된다. 선단부 핸들러에 부착된 힘센서를 통해 로봇팔 조작자가 핸들러를 조종하는 위치 자세의 6차원의 힘/토크를 감지하고 이를 선단부 조종속도 명령으로 변환하여 6축 로봇팔을 제어한다. 연구 방법의 검증은 자체 제작된 6축 로봇으로 실행하였으며, 조종자의 핸들러 조정을 통한 작업교시에 의한 실험을 통해 제안한 힘 센서기반 로봇 선단 제어 방법이 성공적으로 동작함을 확인하였다.

• 대한기계학회논문집
• /
• 제19권1호
• /
• pp.22-37
• /
• 1995

Industrial robot has played a central role in the production automation such as welding, assembling, and painting. There has been, however, little effort to the application of robots in machining work(grinding, cutting, milling, etc.) which is typical 3D work. The machining automation requires a high stiffness robot arm to reduce deformation and vibration. Conventional articulated robots have serially connecting links from the base to the gripper. So, they have very weak structure for he machining work. Stewart Platform is a typical parallel robotic mechanism with a very high stiffness but it has a small work space and a large installation space. This research proposes a new machining robot arm with a double parallel mechanism. It is composed of two platforms and a central axis. The central axis will connect the motions between the first and the second platforms. Therefore, the robot has a large range of work space as well as a high stiffness. This paper will introduce the machining work using the robot and design the proposed robot arm.

• 한국생산제조학회지
• /
• 제23권6호
• /
• pp.580-589
• /
• 2014

Demands for automation in the cell production process of IT products are becoming increasingly sophisticated. In particular, the dual-arm robot has drawn attention as a solution because it has a flexibility and works similarly to humans. In this paper, we propose an automation system for cellular phone packing processes using two dual-arm robots. Applied robots are designed with specifications to meet the requirements of cellular phone packing jobs. In addition, a robotic cell production system is proposed by applying a method of task allocation for efficient packing of cellular phones. Specifically, a task is assigned to reduce takt-time and to avoid collision between two robots. Finally, we discuss some experimental results that include the packing job of five unit boxes with seven kinds of accessories.

• 로봇학회논문지
• /
• 제15권2호
• /
• pp.197-204
• /
• 2020

In this paper, we present a learning platform for robotic grasping in real world, in which actor-critic deep reinforcement learning is employed to directly learn the grasping skill from raw image pixels and rarely observed rewards. This is a challenging task because existing algorithms based on deep reinforcement learning require an extensive number of training data or massive computational cost so that they cannot be affordable in real world settings. To address this problems, the proposed learning platform basically consists of two training phases; a learning phase in simulator and subsequent learning in real world. Here, main processing blocks in the platform are extraction of latent vector based on state representation learning and disentanglement of a raw image, generation of adapted synthetic image using generative adversarial networks, and object detection and arm segmentation for the disentanglement. We demonstrate the effectiveness of this approach in a real environment.

• 대한골관절종양학회지
• /
• 제13권1호
• /
• pp.14-21
• /
• 2007

Bone is a common site for metastatic spread from many kinds of malignancies. The morbidity associated with this metastatic spread can be significant, including severe pain. When it comes to spinal metastasis, occupying nearly 40% of skeletal metastases, the risks of complications, such as vertebral body collapse, nerve root impingement, or spinal cord compression, are also significant. Because of the necessity of preserving the integrity of the spinal column and the proximity of critical structures, surgical treatment has limitations when durable local control is desired. Radiotherapy, therefore, is often used as an adjunct treatment or as a sole treatment. A considerable limitation of standard radiotherapy is the reported recurrence rate or ineffective palliation of pain, either clinically or symptomatically. This may be due to limited radiation doses to tumor itself because of the proximity of critical structures. CyberKnife is an image-guided robotic radiosurgical system. The image guidance system includes a kilovoltage X-ray imaging source and amorphous silica detectors. The radiation delivery device is a mobile X-band linear accelerator (6 MV) mounted on a robotic arm. Highly conformal fields and hypofractionated radiotherapy schedules are increasingly being used as a means to achieve biologic dose escalation for body tumors. Therefore, we can give much higher doses to the targeted tumor volume with minimizing doses to the surrounding critical structures, resulting in more effective local control and less severe side effects, compared to conventional fractionated radiotherapy. A description of this technology and a review of clinical applications to bone metastases are detailed herein.

• 로봇학회논문지
• /
• 제14권1호
• /
• pp.40-49
• /
• 2019

Reinforcement learning has been applied to various problems in robotics. However, it was still hard to train complex robotic manipulation tasks since there is a few models which can be applicable to general tasks. Such general models require a lot of training episodes. In these reasons, deep neural networks which have shown to be good function approximators have not been actively used for robot manipulation task. Recently, some of these challenges are solved by a set of methods, such as Guided Policy Search, which guide or limit search directions while training of a deep neural network based policy model. These frameworks are already applied to a humanoid robot, PR2. However, in robotics, it is not trivial to adjust existing algorithms designed for one robot to another robot. In this paper, we present our implementation of Guided Policy Search to the robotic arms of the Baxter Research Robot. To meet the goals and needs of the project, we build on an existing implementation of Baxter Agent class for the Guided Policy Search algorithm code using the built-in Python interface. This work is expected to play an important role in popularizing robot manipulation reinforcement learning methods on cost-effective robot platforms.