检索结果-内蒙古大学图书馆

ZeroCAP: Zero-Shot Multi-Robot Context Aware Pattern Formation via Large Language Models

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Venkatesh, Vishnunandan L.N. Min, Byung-Cheol SMART Lab Department of Computer and Information Technology Purdue University West LafayetteIN47907 United States

Incorporating language comprehension into robotic operations unlocks significant advancements in robotics, but also presents distinct challenges, particularly in executing spatially oriented tasks like pattern formation. This paper introduces ZeroCAP, a novel system that integrates large language models with multi-robot systems for zero-shot context aware pattern formation. Grounded in the principles of language-conditioned robotics, ZeroCAP leverages the interpretative power of language models to translate natural language instructions into actionable robotic configurations. This approach combines the synergy of vision-language models, cutting-edge segmentation techniques and shape descriptors, enabling the realization of complex, context-driven pattern formations in the realm of multi robot coordination. Through extensive experiments, we demonstrate the systems proficiency in executing complex context aware pattern formations across a spectrum of tasks, from surrounding and caging objects to infilling regions. This not only validates the system’s capability to interpret and implement intricate context-driven tasks but also underscores its adaptability and effectiveness across varied environments and scenarios. The experimental videos and additional information about this work can be found at https://***/view/zerocap/home. © 2024, CC BY.

关键词： Multipurpose robots

ZeroSCD: Zero-Shot Street Scene Change Detection

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Kannan, Shyam Sundar Min, Byung-Cheol SMART Lab Department of Computer and Information Technology Purdue University West LafayetteIN47907 United States

Scene Change Detection is a challenging task in computer vision and robotics that aims to identify differences between two images of the same scene captured at different times. Traditional change detection methods rely on training models that take these image pairs as input and estimate the changes, which requires large amounts of annotated data, a costly and time-consuming process. To overcome this, we propose ZeroSCD, a zero-shot scene change detection framework that eliminates the need for training. ZeroSCD leverages pre-existing models for place recognition and semantic segmentation, utilizing their features and outputs to perform change detection. In this framework, features extracted from the place recognition model are used to estimate correspondences and detect changes between the two images. These are then combined with segmentation results from the semantic segmentation model to precisely delineate the boundaries of the detected changes. Extensive experiments on benchmark datasets demonstrate that ZeroSCD outperforms several state-of-the-art methods in change detection accuracy, despite not being trained on any of the benchmark datasets, proving its effectiveness and adaptability across different scenarios. © 2024, CC BY.

关键词： Change detection

UPPLIED: UAV Path Planning for Inspection through Demonstration

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Kannan, Shyam Sundar Venkatesh, Vishnunandan L.N. Senthilkumaran, Revanth Krishna Min, Byung-Cheol SMART Lab Department of Computer and Information Technology Purdue University West LafayetteIN47907 United States SMART Lab Department of Computer and Information Technology Purdue University School of Electrical and Computer Engineering Purdue University West LafayetteIN47907 United States

— In this paper, a new demonstration-based path-planning framework for the visual inspection of large structures using UAVs is proposed. We introduce UPPLIED: UAV Path PLanning for InspEction through Demonstration, which utilizes a demonstrated trajectory to generate a new trajectory to inspect other structures of the same kind. The demonstrated trajectory can inspect specific regions of the structure and the new trajectory generated by UPPLIED inspects similar regions in the other structure. The proposed method generates inspection points from the demonstrated trajectory and uses standardization to translate those inspection points to inspect the new structure. Finally, the position of these inspection points is optimized to refine their view. Numerous experiments were conducted with various structures and the proposed framework was able to generate inspection trajectories of various kinds for different structures based on the demonstration. The trajectories generated match with the demonstrated trajectory in geometry and at the same time inspect the regions inspected by the demonstration trajectory with minimum deviation. The experimental video of the work can be found at https://***/YqPx-cLkv04. © 2023, CC BY.

关键词： Inspection

technology and public interests

学校读者我要写书评

暂无评论

IEEE Potentials 2022年第5期41卷 22-26页

作者： Park, Chanwoon Matson, Eric T. Kim, Miran Purdue University West LafayetteIN47907 United States National R&d Institutes of South Korea Korea Republic of Purdue University Department of Computer and Information Technology West LafayetteIN47907 United States National Assembly Secretariat of South Korea Korea Republic of

Since the inception of human history, people have embraced technologies, from stone axes to plows. Later, the wave of social and technological modernization unleashed by the Industrial Revolution swept the world into the contemporary era. Today, humans are confronting ever more overwhelming technologies. Social and technical discourse on and debates about new technologies with mass popularity or economic ramifications abound. © 1988-2012 IEEE.

关键词： Economics History Social implications of technology Sociotechnical systems

The Evolution of Team Coordination Commitments in the Context of Computational Projects

学校读者我要写书评

暂无评论

The Evolution of Team Coordination Commitments in the Contex...

Frontiers in Education (FIE) Conference

作者： Joreen Arigye Abasiafak Ndifreke Udosen Parth Pravin Joshi Alejandra J. Magana School of Engineering Education Purdue University West Lafayette USA Department of Computer and Information Technology Purdue University West Lafayette USA Department of Computer and Information Technology School of Engineering Education Purdue University West Lafayette USA

This study aims to explore the team coordination commitments reported by undergraduate students pursuing biomedical and agricultural engineering majors after completing three team-based computational modeling assignments throughout the course of a semester. In each post-assignment reflection, the teams were asked to reflect on what aspects could be done better for the next project in terms of team coordination to address the issues that didn't work. Team-based reflections were analyzed using thematic analysis. We identified students' recognition of the importance of scheduling and coordinating meetings, task management and delegation, collaboration and teamwork, and time management towards teams coordination. Furthermore, the results of this study provide the evolution of teams' commitments for three teams, indicating that teams had a shift in their coordination priorities for the different projects.

关键词：

A Matrix Taxonomy of Knowledge, Skills, and Abilities (KSA) Shaping 2030 Labor Market 53

学校读者我要写书评

暂无评论

A Matrix Taxonomy of Knowledge, Skills, and Abilities (KSA) ...

53rd IEEE ASEE Frontiers in Education International Conference, FIE 2023

作者： Caratozzolo, Patricia Azofeifa, Jose Daniel Mejia-Manzano, Luis Alberto Rueda-Castro, Valentina Noguez, Julieta Magana, Alejandra J. Benes, Bedrich Institute for the Future of Education Tecnologico de Monterrey Mexico City Mexico School of Engineering and Sciences Tecnologico de Monterrey Mexico City Mexico School of Engineering Education Purdue University Department of Computer and Information Technology West Lafayette United States Purdue University Department of Computer Science West Lafayette United States

ISBN: (纸本)9798350336429

This paper proposes a dynamic Knowledge, Skills, and Abilities (KSA) matrix-based taxonomy for the Industry 4.0 workforce. The study methodology consisted firstly of identifying the KSAs through a literature review and secondly of a KSA relevance analysis using information from World Economic Forum (WEF) global reports and the Organization for Economic Cooperation and Development (OECD). Finally, we identified the correlation coefficients of the KSA matrix elements concerning the data on jobs and occupations using information from the European Skills, Competencies, and Occupations (ESCO), Occupational information Network (O∗NET), and the strategic intelligence platform of the World Economic Forum. One of the goals was to make the taxonomy compatible with existing and future machine learning methods (i.e., AI-ready) that will enable efficient and effective use of AI in mining and explaining existing and potentially proposing novel trends and strategies. Preliminary results show that the KSA Industry 4.0 Taxonomy can serve as an international reference guide for designing 2030 educational approaches to active and experiential learning in Higher Education Institutions. © 2023 IEEE.

关键词： Abilities (KSA) Educational Innovation Future Skills Higher Education Industry 4.0 Knowledge Labor Market Skills

Digital Forensic Analysis of AGPTEK Smartwatch Application on Android OS

学校读者我要写书评

暂无评论

Digital Forensic Analysis of AGPTEK Smartwatch Application o...

International Symposium on Networks, computers and Communications (ISNCC)

作者： Claire Rightley Umit Karabiyik Department of Computer and Information Technology Purdue University West Lafayette Indiana USA

As wearable IoT devices become more popular, the likelihood for digital forensic investigators and analysts to encounter them in the field increases. Smartwatches are one of the most popular wearable IoT devices today, and many brands and devices have yet to be analyzed. Analysis of such devices plays an integral role in criminal investigations, as data collected by these devices may be used as inculpatory or exculpatory evidence. This paper provides a forensic analysis of the AGPTEK smartwatch application (FitCloudPro) on Android OS for both a non-rooted and rooted smartphone. The smartwatch was worn for a data population period while paired with the smartphone. The phone was then imaged using Magnet AXIOM Process and analyzed using Magnet AXIOM Examine. Directories, files, and databases of interest are documented. The smartwatch application creates one major database containing user and activity data (appDatabase). The logs for the application contain an abundant amount of descriptive data but only are retained for approximately three days.

关键词：

Cooperative Learning and Co-Regulation: Exploring Students' Teamwork Strategies in Higher Education

学校读者我要写书评

暂无评论

Cooperative Learning and Co-Regulation: Exploring Students' ...

Frontiers in Education (FIE) Conference

作者： Sakhi Aggrawal Jorge A. Cristancho Devang A. Patel Alejandra J. Magana Department of Computer and Information Technology Purdue University West Lafayette Indiana School of Engineering Education Purdue University West Lafayette Indiana Department of Computer and Information Technology School of Engineering Education Purdue University West Lafayette Indiana

This research paper investigates the effectiveness of cooperative learning and co-regulation strategies in promoting teamwork and enhancing students' performance in higher education. The study was conducted in an in-person intermediate-level information system design course with 152 students divided into 31 teams. The students utilized the Scrum framework to manage a semester-long project with three milestones. Retrospective data were collected at the end of each milestone, and the first milestone data were analyzed in this study. Through a thematic analysis of retrospective data collected after the first milestone, the study examines students' planning, monitoring, and reflection strategies. The findings reveal that students demonstrated adaptive planning, equitable contribution, and task allocation based on individual strengths and preferences. In terms of monitoring, students adopted a proactive approach, displayed relational competence, and utilized both synchronous and asynchronous communication channels. Regarding reflection, students valued effective planning and execution but struggled with time management. They developed concrete improvement strategies for the next milestone, emphasizing realistic deadlines, improved communication, and a better understanding of team members' strengths. This research contributes to the understanding of cooperative learning and co-regulation in promoting effective teamwork in higher education. The findings have implications for pedagogical practices and suggest the importance of integrating cooperative learning and co-regulation strategies in team-based learning environments. Future research can further explore the application of these strategies in different educational contexts and investigate their long-term effects on students' performance and engagement.

关键词：

Toward Mitigating Sex Bias in Pilot Trainees’ Stress and Fatigue Modeling

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Pfeifer, Rachel Vhaduri, Sudip Wilson, Mark Keller, Julius Computer Science Department Computer and Information Technology Department School of Health Sciences School of Aviation and Transportation Technology Purdue University IN47907 United States

While researchers have been trying to understand the stress and fatigue among pilots, especially pilot trainees, and to develop stress/fatigue models to automate the process of detecting stress/fatigue, they often do not consider biases such as sex in those models. However, in a critical profession like aviation, where the demographic distribution is disproportionately skewed to one sex, it is urgent to mitigate biases for fair and safe model predictions. In this work, we investigate the perceived stress/fatigue of 69 college students, including 40 pilot trainees with around 63% male. We construct models with decision trees first without bias mitigation and then with bias mitigation using a threshold optimizer with demographic parity and equalized odds constraints 30 times with random instances. Using bias mitigation, we achieve improvements of 88.31% (demographic parity difference) and 54.26% (equalized odds difference), which are also found to be statistically significant. © 2024, CC BY.

关键词： Population statistics