[1] Aerosol optical properties (aerosol optical thickness, Å ngström exponent, size distribution, and single scattering albedo) over east Asia were examined using long-term measurements of sky radiation at Mandalgovi, Dunhuang, Yinchuan, and Sri-Samrong sites of the Skyradiometer Network (SKYNET). Also included were sky radiation measurements at Anmyon, Gosan in Korea, and Amami-Oshima in Japan during April for examining optical properties of Asian dust. Results show that the seasonal average of aerosol optical thickness (AOT) generally exhibits a maximum in spring and a minimum in autumn over east Asia. At Sri-Samrong and Yinchuan, relatively distinct seasonal cycles are noted, in comparison to the arid desert regions of Dunhuang and Mandalgovi. In general, aerosol size distributions are characterized by a bimodal pattern, with a fine mode around 0.2 mm and a coarse mode around 2À5 mm. Similar to AOT and a, volume spectra are also much dependent on geographical location and season. Dunhuang mostly shows coarse mode particles in all seasons, while Mandalgovi and Sri-Samrong show large seasonal variations in the total volume of fine mode particles. The single scattering albedos of dust particles over east Asia are around 0.9 at 0.5 mm, which are larger than the previously known values of 0.63-0.89 but similar to those found in the Aerosol Robotic Network (AERONET) analysis. It is noted that the optical properties of Asian dust around Korea and Japan are quite similar to those found in dust source regions such as Dunhuang and Mandalgovi. However, the single scattering albedo appears to be smaller than those observed in Dunhuang and Mandalgovi. Furthermore, single scattering albedo tends to become smaller during the dust outbreak period. Considering that aerosols in Korean and Japanese areas are much influenced by anthropogenic aerosols emitted in China particularly under the westerly conditions, the mixing processes between different aerosol species may be the cause of the different optical properties of Asian dust.
As safety is of paramount importance in robotics, reinforcement learning that reflects safety, called safe RL, has been studied extensively. In safe RL, we aim to find a policy which maximizes the desired return while satisfying the defined safety constraints. There are various types of constraints, among which constraints on conditional value at risk (CVaR) effectively lower the probability of failures caused by high costs since CVaR is a conditional expectation obtained above a certain percentile. In this paper, we propose a trust region-based safe RL method with CVaR constraints, called TRC. We first derive the upper bound on CVaR and then approximate the upper bound in a differentiable form in a trust region. Using this approximation, a subproblem to get policy gradients is formulated, and policies are trained by iteratively solving the subproblem. TRC is evaluated through safe navigation tasks in simulations with various robots and a sim-toreal environment with a Jackal robot from Clearpath. Compared to other safe RL methods, the performance is improved by 1.93 times while the constraints are satisfied in all experiments.
This paper aims to solve a safe reinforcement learning (RL) problem with risk measure-based constraints. As risk measures, such as conditional value at risk (CVaR), focus on the tail distribution of cost signals, constraining risk measures can effectively prevent a failure in the worst case. An on-policy safe RL method, called TRC, deals with a CVaR-constrained RL problem using a trust region method and can generate policies with almost zero constraint violations with high returns. However, to achieve outstanding performance in complex environments and satisfy safety constraints quickly, RL methods are required to be sample efficient. To this end, we propose an off-policy safe RL method with CVaR constraints, called off-policy TRC. If offpolicy data from replay buffers is directly used to train TRC, the estimation error caused by the distributional shift results in performance degradation. To resolve this issue, we propose novel surrogate functions, in which the effect of the distributional shift can be reduced, and introduce an adaptive trust-region constraint to ensure a policy not to deviate far from replay buffers. The proposed method has been evaluated in simulation and real-world environments and satisfied safety constraints within a few steps while achieving high returns even in complex robotic tasks.
According that most integral imaging techniques have used rectangular lens array, this integrated distribution of light is recorded in the form of a rectangular grid. However, hexagonal lens array gives a more accurate approximation of ideal circular lens and provides higher pickup/display density than rectangular lens array [4]. Using the parallel processing technique in order to generate the elemental imaging for hexagonal lens array, each pixel that compose the elemental imaging should be determined to belong to the hexagonal lens. This process is output to the screen for every pixel in progress, and many computations are required. In this paper, we have proposed parallel processing method using an OpenCL to generate the elemental imaging for hexagonal lens array in 3D volume date. In the experimental result of proposed method show speed of 20∼60 fps for hexagonal lens array of 20x20 sizes and input data of Male[128x256x256] volume data.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.