In this paper, we exploit the properties of mean absolute error (MAE) as a loss function for the deep neural network (DNN) based vector-to-vector regression. The goal of this work is two-fold: (i) presenting performan...
详细信息
In this paper, we show that, in vector-to-vector regression utilizing deep neural networks (DNNs), a generalized loss of mean absolute error (MAE) between the predicted and expected feature vectors is upper bounded by...
详细信息
We address the interpretability of convolutional neural networks (CNNs) for predicting a geo-location from an image. In a pilot experiment we classify images of Pittsburgh vs Tokyo and visualize the learned CNN filter...
详细信息
In this work we present modified iterative methods for computing basic statistical quantities (mean, variance) for use in the calibration process of a system based on V2I (vehicle-to-infrastructure) communication devi...
详细信息
In this work we present modified iterative methods for computing basic statistical quantities (mean, variance) for use in the calibration process of a system based on V2I (vehicle-to-infrastructure) communication devices. Such devices, mounted in the road and urban infrastructure (RSU - road side equipment) may be used as support for autonomous vehicles moving in urban environment. Calibration is necessary to determine the positions of the RSUs in global coordinate system (GCS) and to record this information in their internal memory. The proposed modifications to conventional iterative algorithms aim at adapting these methods to the application in devices with low computational abilities or directly at the transistor level in specialized integrated circuits.
This study aims to delineate the origin, purpose, process, and results of the research, focusing on investigating the impact of different exterior wall materials on outdoor thermal indicators in Nagoya, Japan. The ult...
This study aims to delineate the origin, purpose, process, and results of the research, focusing on investigating the impact of different exterior wall materials on outdoor thermal indicators in Nagoya, Japan. The ultimate goal is to mitigate the Urban Heat Island (UHI) effect and improve urban outdoor thermal comfort. Computational Fluid Dynamics (CFD) is employed to evaluate two types of exterior wall materials: diffuse highly reflective (DHR) and retro-reflective (RR) materials. The research process involves assessing various outdoor thermal indices, such as wet-bulb globe temperature (WBGT), standard effective temperature (SET*), and universal thermal climate index (UTCI), in order to comprehend the thermal performance of each material. The results of the study indicate that the RR material consistently outperforms the DHR material in terms of outdoor thermal comfort, despite having slightly lower solar reflectance. This is supported by lower surface temperatures and surface heat flow associated with the RR material, suggesting reduced indoor cooling loads and significant energy savings for cooling purposes. These findings highlight the potential of RR material in mitigating the UHI effect and enhancing urban outdoor thermal conditions. By establishing a critical relationship between exterior wall materials and outdoor thermal indicators, this study proposes the adoption of RR materials for urban building exteriors as a sustainable and economically viable solution to improve urban outdoor thermal comfort and enhance the quality of life for residents.
Solutions proposed in this work are related to development of vehicle-to-infrastructure (V2I) communication in the context of its support for Traffic Sign Recognition (TSR) algorithms used in advanced driver assistanc...
详细信息
Solutions proposed in this work are related to development of vehicle-to-infrastructure (V2I) communication in the context of its support for Traffic Sign Recognition (TSR) algorithms used in advanced driver assistance systems (ADAS). One of the ideas of the application of the V2I communication, proposed in the literature, is to equip traffic signs (TS) with devices, capable of communicate their meaning to passing vehicles equipped with ADAS functions or autonomous vehicles. We propose grouping TSs and covering groups with single road side unit (RSU) devices. It will facilitate implementation and maintenance of the overall system. This will also reduce the amount of data sent over the network.
Large, well described gaps exist in both what we know and what we need to know to address the biodiversity crisis. Artificial intelligence (AI) offers new potential for filling these knowledge gaps, but where the bigg...
Large, well described gaps exist in both what we know and what we need to know to address the biodiversity crisis. Artificial intelligence (AI) offers new potential for filling these knowledge gaps, but where the biggest and most influential gains could be made remains unclear. To date, biodiversity-related uses of AI have largely focused on tracking and monitoring of wildlife populations. Rapid progress is being made in the use of AI to build phylogenetic trees and species distribution models. However, AI also has considerable unrealized potential in the re-evaluation of important ecological questions, especially those that require the integration of disparate and inherently complex data types, such as images, video, text, audio and DNA. This Review describes the current and potential future use of AI to address seven clearly defined shortfalls in biodiversity knowledge. Recommended steps for AI-based improvements include the re-use of existing image data and the development of novel paradigms, including the collaborative generation of new testable hypotheses. The resulting expansion of biodiversity knowledge could lead to science spanning from genes to ecosystems — advances that might represent our best hope for meeting the rapidly approaching 2030 targets of the Global Biodiversity Framework.
Augmented Reality and mobile robots are gaining much attention within industries due to the high potential to make processes cost and time efficient. To facilitate augmented reality, a calibration between the Augmente...
详细信息
Human robot collaboration is aspiring to establish hybrid work environments in accordance with specific strengths of humans and robots. We present an approach of flexibly integrating robotic handover assistance into c...
详细信息
ISBN:
(纸本)9781728126234
Human robot collaboration is aspiring to establish hybrid work environments in accordance with specific strengths of humans and robots. We present an approach of flexibly integrating robotic handover assistance into collaborative assembly tasks through the use of natural communication. For flexibly instructed handovers, we implement recent Convolutional Neural Networks in terms of object detection and grasping of arbitrary objects based on an RGB-D camera equipped to a robot following the eye-in-hand principle. In order to increase fluency and efficiency of the overall assembly process, we investigate the human ability to instruct the robot predictively with voice commands. We conduct a user study quantitatively and qualitatively evaluating the predictive instruction in order to achieve just-in-time handovers of tools needed for following subtasks. We compare our predictive strategy with a pure manual assembly having all tools in direct reach and a step-by-step reactive handover. The results reveal that the human is able to predict the handover comparable to algorithm-based predictors. Nevertheless, human prediction does not rely on extensive prior knowledge and is thus suitable for more flexible usage. However, the cognitive workload for the worker is increased compared to manual or reactive assembly.
Nanoelectromechanical system (NEMS) sensors and actuators could be of use in the development of next generation mobile, wearable, and implantable devices. However, these NEMS devices require transducers that are ultra...
详细信息
暂无评论