55453 research outputs found
Sort by
Collaborative Robots for the Automation of Additive Manufacturing Workflows
No full textCollaborative robots have the potential to transform the world, being easier to program, safer around humans, and far cheaper than their industrial counterparts. This makes them uniquely qualified to operate autonomously in environments where the presence of humans is necessary, either as an operator themselves or as observers of active processes. One of the fields where collaborative robots have significant opportunity for utility is in metal additive manufacturing, in which near net-shape parts can be autonomously fabricated using materials like stainless steel, titanium, or nickel super-alloys. These processes, especially laser powder bed fusion and directed energy deposition, use a fine powder feedstock to incrementally print incredibly complex metal parts; however, this powder has the downside in that it poses a severe hazard to any humans unfortunate enough to exist unprotected around it, damaging eyes and lungs. This risk is maximized once the part is fully printed and is being taken out of the machine, where all the leftover powder must be removed prior to post-processing, which includes any additional processes like heat treatment, surface finishing, and support removal. Currently, this is facilitated almost entirely by humans, despite the hazards, simply because most machines are not capable of automating these steps. This leads to my research, which investigates the development of a framework providing the fundamental tools required for automation. The ubiquity of additive manufacturing is precisely why automation is so challenging; unlike in traditional manufacturing, where the inputs and outputs of the processes are incredibly consistent, additive processes are typically much lower throughput and more varied, so automation solutions must be both (1) efficient, such that implementations of automation are accessible, and (2) flexible and robust enough to handle the intrinsic variation of additive technologies. Paramount to this is the development of the end effectors required for collaborative robots to effectively interact with these parts, the design of an efficient, powerful computer 3D vision algorithm, and the culmination of a significant aspect of the post-processing workflow in a powder cleanup routine. These developments will prove foundational to further expansion of automation within an additive manufacturing workflow, with collaborative robots leading the way as a critical enabler
Administration, Sophia Gumenick
No full textA photographic print of VCU administrator Sophia Gumenick.https://scholarscompass.vcu.edu/rg60/1061/thumbnail.jp
Faculty, Richard Keenan, 1990 August
No full textA photographic print of VCU professor Dr. Richard Keenan using a pulse oximeter.https://scholarscompass.vcu.edu/rg60/1062/thumbnail.jp
Faculty, Bruce Fuchs
No full textA photographic print of VCU professor Dr. Bruce Fuchs sitting down and holding a writing utensil, with a piece of paper on a table.https://scholarscompass.vcu.edu/rg60/1079/thumbnail.jp
Administration/Faculty
No full textA photographic print of a VCU adminsitrator or faculty member only identified by the name Sergei.https://scholarscompass.vcu.edu/rg60/1086/thumbnail.jp
Administration/Faculty
No full textA photographic print of an unnamed VCU administrator or faculty member.https://scholarscompass.vcu.edu/rg60/1104/thumbnail.jp
Administration/Faculty
No full textA contact sheet of multiple photographs of an unnamed VCU administrator or faculty member sitting and conversing with two other individuals across a table.https://scholarscompass.vcu.edu/rg60/1108/thumbnail.jp
Indiana Prison Writers Workshop: The Power of Group Cohesion and Creative Expression among Incarcerated Individuals
Get PDFObjective: The present study describes the development of the Indiana Prison Writers Workshop (IPWW), a creative writing workshop, based in group theory and designed for incarcerated settings. Methods: Quantitative evaluation data from IPWW workshops conducted from 2023-2024 were analyzed using IBM SPSS V. 29. Prior to the workshop, 149 men responded to questions on class expectations and goals, and 115 men responded at workshop conclusion. Open-ended qualitative responses from post-workshop data were analyzed using thematic analysis. Results: Quantitative data demonstrates the effectiveness of IPWW in teaching writing skills and reflects participants’ agreement that the workshops provide opportunities to share their work and receive feedback from peers and the facilitator. Five themes emerged from the qualitative data: 1) self-expression, 2) skill development, 3) self-reflection, 4) perspective, and 5) therapeutic. Conclusions: We discuss how these results influence group cohesion among incarcerated men and explore implications for future research and IPWW program development
PyZoBot: A Platform for Conversational Information Extraction and Synthesis from Curated Zotero Reference Libraries through Advanced Retrieval-Augmented Generation.
No full textThis dissertation presents a systematic evaluation of PyZoBot, an AI-powered platform for literature- based question answering, using the Retrieval-Augmented Generation Assessment Scores (RAGAS) framework. The study focuses on a subset of 49 cardiology-related questions extracted from the BioASQ benchmark dataset. PyZoBot\u27s performance was assessed across 32 configurations, including standard Retrieval-Augmented Generation (RAG) and GraphRAG pipelines, implemented with both OpenAI-based models (GPT-3.5-Turbo, GPT-4o) and open- source models (LLaMA 3.1, Mistral).
To establish a comparative benchmark, responses generated by PyZoBot were evaluated alongside answers manually written by six PhD students and recent graduates from the pharmacotherapy field, using a curated Zotero library containing BioASQ-referenced documents. The evaluation applied four key RAGAS metrics—faithfulness, answer relevancy, context recall, and context precision—along with a composite harmonic score to determine overall performance.
The findings reveal that 22 PyZoBot configurations surpassed the highest-performing human participant, with the top pipeline (GPT-3.5-Turbo + layout-aware chunking, k=10) achieving a harmonic RAGAS score of 0.6944. Statistical analysis using Kruskal-Wallis and Dunn’s post hoc tests confirmed significant differences across all metrics, especially in faithfulness and time efficiency.
These results validate PyZoBot’s ability to support high-quality biomedical information synthesis and demonstrate the system’s potential to meet or exceed human performance in complex, evidence-based academic tasks