• Korean Journal of Agricultural Science
• /
• v.24 no.2
• /
• pp.182-193
• /
• 1997

The normal cement concrete is widely used material to build the construction recently, but it has a fault to increase the dead load on account of its unit weight is large compared with strength. Therefore, many engineers are continuously searching for new materials of construction to provide greater performance at lower density. The main purpose of the work described in this paper were to establish the physical and mechanical properties of synthetic lightweight aggregate concrete using perlite on fine aggregate and expanded clay, pumice stone on coarse aggregate. The test results of this study are summarized that the water-cement ratio was shown 47% using expanded clay, 56% using pumice stone on coarse aggregate, unit weight was shown $l,622kgf/m^3$ using expanded clay, $l,596kgf/m^3$ using pumice stone on coarse aggregate, and the absorption ratio was shown same as 17%. The compressive strength was shown more than $228kgf/cm^2$, tensile and bending strength was more than $27kgf/cm^2$, $58kgf/cm^2$ at all types, and rebound number with schmidt hammer was increased with increase of compressive strength. The static modulus was $1.12{\times}10^5kgf/cm^2$ using expanded clay, $1.09{\times}10^5kgf/cm^2$ using pumice stone on coarse aggregate, and stress-strain curves were shown that increased with increase of stress, and the strain on the maximum stress was shown identical with $2.0{\times}10^{-3}$, approximately.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.5
• /
• pp.179-186
• /
• 2021

Manipulation of complex objects with an anthropomorphic robot hand like a human hand is a challenge in the human-centric environment. In order to train the anthropomorphic robot hand which has a high degree of freedom (DoF), human demonstration augmented deep reinforcement learning policy optimization methods have been proposed. In this work, we first demonstrate augmentation of human demonstration in deep reinforcement learning (DRL) is effective for object manipulation by comparing the performance of the augmentation-free Natural Policy Gradient (NPG) and Demonstration Augmented NPG (DA-NPG). Then three DRL policy optimization methods, namely NPG, Trust Region Policy Optimization (TRPO), and Proximal Policy Optimization (PPO), have been evaluated with DA (i.e., DA-NPG, DA-TRPO, and DA-PPO) and without DA by manipulating six objects such as apple, banana, bottle, light bulb, camera, and hammer. The results show that DA-NPG achieved the average success rate of 99.33% whereas NPG only achieved 60%. In addition, DA-NPG succeeded grasping all six objects while DA-TRPO and DA-PPO failed to grasp some objects and showed unstable performances.