• 제어로봇시스템학회논문지
• /
• 제18권10호
• /
• pp.964-969
• /
• 2012

Robot hands capable of grasping or handling various objects are important for service robots to effectively aid humans. In particular, controlling a contact force and providing a compliant motion are essential when the hand is in contact with objects. Many dexterous robot hands equipped with force/torque sensors have been developed to perform force control, but they suffer from the complexity of control and high cost. In this paper, a low-cost robot hand based on SEA (Series Elastic Actuator), which is composed of compression spring, stretch sensor, and wire, is proposed. The grasping force can be estimated by measuring the compression length of spring, which would allow the hand to perform force control. A series of experimentations are carried out to verify the performance of force control of the proposed robot hand, and it is shown that it can successfully control the contact force without any additional force/torque sensors.

• 한국소음진동공학회논문집
• /
• 제21권11호
• /
• pp.991-996
• /
• 2011

This study was carried out vibration attenuation of vibration absorber attached to the cantilever beam structure. Modern tank guns are stabilized to allow fire on the move while traversing uneven terrain. However, as the length of the barrel is extended, to meet required muzzle exit velocities, the terrain induced vibrations lead to increased muzzle pointing errors. Thus, reducing these vibrations should lead to increased accuracy. The vibration absorber includes a compliant energy storage device, such as a spring, and a mass secured to the energy storage device. In this study, it accomplished a research in about gun barrel vibration attenuation using tuned mass damper. The barrel was hung from a bungee cord for free-free condition. It accomplished a vibration experiment for verified attenuation efficiency.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2003년도 춘계학술대회 논문집
• /
• pp.1561-1564
• /
• 2003

In the dynamic analysis model of full vehicles, the ball joint is usually modeled as an ideal joint. Searching a ball joint, the engineering plastic covers metal and the plastic has little compliance. It is expected that the compliance will physically have an influence on load transfer. This thesis presents a dynamic model considering the compliance of a ball joint, and studies an influence related to load transfer. It models the compliance of a ball joint to 3 directional spring. Likewise, it researches the load of a ball joint via a four-post simulation of a full vehicle, comparing with a model considered compliant and the model of an ideal joint. As a result, the difference between the compliance and the ideal joint model was determined. For this reasons, to conduct precision load prediction for durability analysis, dynamic analysis considering the compliance of bali joint should be required.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2011년도 춘계학술대회 논문집
• /
• pp.627-632
• /
• 2011

This study was carried out vibration attenuation of vibration absorber attached to the cantilever beam structure. Modern tank guns are stabilized to allow fire on the move while traversing uneven terrain. However, as the length of the barrel is extended, to meet required muzzle exit velocities, the terrain induced vibrations lead to increased muzzle pointing errors. Thus, reducing these vibrations should lead to increased accuracy. The vibration absorber includes a compliant energy storage device, such as a spring, and a mass secured to the energy storage device. In this study, it accomplished a research in about gun barrel vibration attenuation using tuned mass damper. The barrel was hung from a bungee cord for free - free condition. It accomplished a vibration experiment for verified attenuation efficiency.

• 대한전기학회논문지
• /
• 제39권1호
• /
• pp.92-106
• /
• 1990

Active compliance is often used in the control of robot manipulators for the implementation of complex tasks such as assembly, multi-finger fine motion, legged-vehicle adaptive control,etc. This technique balances the interactive force between the manipulator tip and its working environment with its position and velocity errors to achieve the operation of a damped spring. This paper investigates the effecft of passive compliance on system stability with regard to force feedback implementation for actively compliant motion. Usually it is understood that accurate position control require a stiff system. However, theoretical examination of control experiments on a legged suspension vehicle suggests that, if the control includes discrete-time force feedback, some passive compliance is necessssary at the legs of the vehicle for system stability. This can be an important factor to bl considered in manipulator design and control. A theoretical analysis, numerical simulation, and experimental result, confirming the above conclusion, are introduced in this paper.

• Advances in biomechanics and applications
• /
• 제1권3호
• /
• pp.169-185
• /
• 2014

Computational multibody models of the elbow can provide a versatile tool to study joint mechanics, cartilage loading, ligament function and the effects of joint trauma and orthopaedic repair. An efficiently developed computational model can assist surgeons and other investigators in the design and evaluation of treatments for elbow injuries, and contribute to improvements in patient care. The purpose of this study was to develop an anatomically correct elbow joint model and validate the model against experimental data. The elbow model was constrained by multiple bundles of non-linear ligaments, three-dimensional deformable contacts between articulating geometries, and applied external loads. The developed anatomical computational models of the joint can then be incorporated into neuro-musculoskeletal models within a multibody framework. In the approach presented here, volume images of two cadaver elbows were generated by computed tomography (CT) and one elbow by magnetic resonance imaging (MRI) to construct the three-dimensional bone geometries for the model. The ligaments and triceps tendon were represented with non-linear spring-damper elements as a function of stiffness, ligament length and ligament zero-load length. Articular cartilage was represented as uniform thickness solids that allowed prediction of compliant contact forces. As a final step, the subject specific model was validated by comparing predicted kinematics and triceps tendon forces to experimentally obtained data of the identically loaded cadaver elbow. The maximum root mean square (RMS) error between the predicted and measured kinematics during the complete testing cycle was 4.9 mm medial-lateral translational of the radius relative to the humerus (for Specimen 2 in this study) and 5.30 internal-external rotation of the radius relative to the humerus (for Specimen 3 in this study). The maximum RMS error for triceps tendon force was 7.6 N (for Specimen 3).