11115 research outputs found
Sort by
Optimising human-robot collaboration for efficiency in retail warehousing
No full textThis study investigates the performance of Autonomous Mobile Robot (AMR)–human collaboration in hybrid warehouse order-picking operations. It aims to identify optimal resource allocation strategies and assess the role of warehouse layout design, particularly cross-aisles, in enhancing operational efficiency in high-velocity retail and e-commerce environments
Temporal analysis of the clustering and hypothesized social contagion of mass killing events in the United States
No full textTemporally Multiplexed Spectral LiDAR Testbed in the Near Infrared
No full textA novel multispectral LiDAR testbed is demonstrated utilizing Raman generation in hydrogen filled anti-resonant optical fiber with range and spectral features measured using time-domain pulse processing from a single detector
Engineering safe human-autonomy teaming using STPA-coordination
No full textIn contested and complex air combat environments, safe coordination between decision agents is paramount. While the Department of Defense is developing Artificially Intelligent (AI) agents to perform air combat, there are limited methods to design safe coordination between human pilots and collaborative autonomous wingmen. System-Theoretic Process Analysis extended for Coordination (STPA-Coordination) is a novel analysis method that addresses the limitation. This paper presents research in the application of STPA-Coordination to model and design the Loyal Wingman concept. The research is the first to apply STPA-Coordination to Human-AI teaming in the Air Dominance mission with a focus on weapons engagement decisions. The results showcase a systems-theoretic design framework useful to 1) understand human-AI coordination interactions and 2) to ultimately engineer flawed coordination out of the system before accidents occur
Exploring Emotion Classification of Indonesian Tweets using Large Scale Transfer Learning via IndoBERT
Full text linkBusiness, political, and other social structures create strong motivation to understand the attitudes, motivations, feelings, and emotions of a population of interest. Social media is a rich source of self-disclosed information by individuals from all walks of life about virtually every domain of the human experience, but the vast quantity of data is impossible to effectively analyze without advanced natural language processing algorithms. This research creates a transfer learning based emotion classification model for Indonesian language Twitter data. Transfer learning consists of two steps: pre-training and fine tuning. Three variations of Indonesian Bidirectional Encoder Representations from Transformers (IndoBERT) are tested with hyperparameters tuned via designed experiment. The top IndoBERT model, tested on an open source corpus of 4,401 labeled Indonesian Tweets, outperforms all known prior studies with an F1 score of approximately 0.791. Additionally, this research explores the relationship between training set size and model validity for fine tuning of the transfer learning models; datasets ranging from 100 to 3900 observations are trained and then validated on five unique test sets. Results indicate that as few as 1000 observations can obtain results comparable to using the full training corpus
The Effects of Snow Cover on the Dynamic Pressure of Nuclear Detonation Blast Waves
Full text linkBlast pressure is the primary military targeting metric for nuclear weapons. Any local conditions that affect blast pressure have the potential for altering nuclear plans, both from defensive and offensive standpoints. Understanding the impact of snow to the blast wave, therefore, provides a benefit both to military planners and to warfighters on the ground, for any operation occurring in arctic environments. No existing data provides a quantitative description of how snow on the ground affects a nuclear detonation blast wave passing over it. Similar blast waves passing over dust have experimentally proven to enhance blast pressure in a localized region.1 This research seeks to understand the snow lofting mechanism and determine quantitative results to dynamic and total pressures within the blast wave. The research strategy selected to investigate this topic included conducting several scale experiments using a shock tube. Shock waves, once produced, are very similar regardless of source. If a lab-created shock front transited over snow or a snow-proxy, researchers could determine the pressure delta by comparing those measurements against similar data recorded from a shock front passing over an ideal surface. This result could then improve the understanding of how shock waves created by detonations behave in larger correlating environments
Image Domain Distinct Native Attribute Fingerprinting for Image Forgery Classification
No full textImage forgery is becoming more difficult to detect due to advances in AI image generation. As such, the usefulness — and even requirement — for detection techniques that are affordable (computationally and monetarily) as well as intuitive and simple are equally increasing. This work demonstrates the first adoption of Distinct Native Attribute (DNA) Fingerprinting to image and forgery detection to achieve similar results while mitigating the cost of implementation. General image classification results with accuracy of %C = 98.8% support the overall utility while the ability to detect within-category image forgeries produce an average of %C = 81.8%. Using an intuitive and small set of features, preliminary results show an approximate average classification accuracy difference of only %CΔ = −9% from more complex solutions. This work demonstrates the ability to adopt DNA Fingerprinting for image classification, and image forgery using Image Domain DNA (ID-DNA) that is holistically less resource intensive while requiring less time, money, and expert knowledge
Discovering Reentry Vehicle Characteristics Using Nondimensional Equations of Motion and Dimensional Observations
No full text
Biot Number Error in Low-Temperature Inconel Overall Effectiveness Experiments
No full textTo predict the performance of turbine materials at engine conditions, experiments are often performed at low-temperature laboratory conditions. In order to ensure the low-temperature, laboratory results accurately predict the nondimensionalized surface temperature at engine conditions, several nondimensional parameters must be matched in the experiment, including the Biot number. Matching the Biot number requires that the ratio of the thermal conductivity of the material to the thermal conductivity of the air must be matched between laboratory experiments and engine conditions. With traditional nickel alloys such as Inconel, it is sometimes assumed that the Biot number is matched since Inconel\u27s thermal conductivity variation with temperature scales relatively closely with that of air. However, the thermal conductivity ratio does not scale perfectly and therefore some Biot number error does indeed exist, with the problem exacerbated at lower testing temperatures. To date, there has been no experimentally verified quantification of the error in the overall effectiveness, ϕ, that might be caused by this Biot number error. Ti-6Al-4V is predicted to allow for a better Biot number match, thereby better simulating Inconel at engine conditions in typical low-temperature experiments. In this research, we utilized geometrically identical models constructed of Ti-6Al-4V and Inconel 718 to evaluate the error in overall effectiveness that might occur through simply using an actual engine nickel alloy part at experimental conditions. While the Ti-6Al-4V model has a nearly perfectly matched Biot number, the Inconel model\u27s Biot number was 73% higher than appropriate. The results demonstrate that ϕ measured in low-temperature tests performed on an Inconel turbine component do not suffer markedly from Biot number error. The theoretically more Biot number appropriate Ti-6Al-4V model produced area-averaged overall effectiveness values that differed by only 0.01 from its Inconel counterpart. These results suggest that typical nickel superalloys used in turbine components may be tested at low temperature without the use of a surrogate material to better match Biot number