FW-HTF-RL: Co-worker Robots to Impact Seafood Processing (CRISP): Designs, Tools and Methods for Enhanced Worker Experience

FW-HTF-RL：影响海鲜加工的同事机器人 (CRISP)：增强工人体验的设计、工具和方法

• 批准号: 1928654
• 负责人: Taskin Padir
• 金额: $ 250万
• 依托单位: Northeastern University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-09-01 至 2024-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1928654&HistoricalAwards=false
• 关键词: FW; HTF; RL; Co; worker

This Future of Work at the Human-Technology Frontier (FW-HTF) project will investigate the appropriate development and deployment of collaborative robots to transform profitability, productivity, safety, sustainability, and worker quality of life in the seafood processing industry, where harsh conditions and demanding and dangerous tasks challenge the capabilities of humans and robots alike. The result will be designs, tools, methods, and datasets to facilitate seamless human-robot collaboration. Soft robot manipulators will augment the safe and reliable handling of slippery, scaly, and flexible objects. Emerging capabilities in artificial intelligence will assist in identifying and inspecting varieties of fish and shellfish. Critical to the project is understanding how best to allocate specific tasks among robot and human workers, integrating a complex set of desired outcomes, across scales of individual workers, independent businesses, domestic and migrant labor markets, national economic sectors, and global trade, while respecting environmental and ethical constraints. Associated educational and outreach programs will empower engineers, design experts, and social scientists how to address challenges at the convergence of robotics and manipulation, artificial intelligence, human-robot collaboration, ethics, and labor economics. Training programs will develop a new cadre of learners and researchers in data, technology and human literacies. In 2017, the US imported record amounts of seafood, corresponding to a trade deficit of more than $17 billion. Due to low domestic unemployment and obstacles to employing migrant labor, the US seafood processing industry has been unable to meet US consumer demand. This project will advance US leadership in a globally competitive and domestically underserved industry, while simultaneously advancing understanding of key scientific, engineering, and societal challenges.The research goals of the project will impact five interconnected and convergent areas, namely (i) collaborative robotics and shared autonomy, (ii) interaction design with data visualizations, (iii) labor economics in seafood industry, (iv) ethics of autonomy in the workplace, (v) workforce training and new skills learning. To address shortcomings of today?s human-robot co-worker teams, the research plan aims to dramatically enhance productivity in complex environments, while paying specific attention to factors affecting user acceptability. The team will emphasize the integration of robotic systems in the existing socio-technical context -- existing machinery and tools and existing human work practices, including formal and informal modalities ? to develop an approach of ambient and distributed robotics, rather than individual robot interventions. The research plan develops a constellation of design requirements for systems through user-centered design and ideation activities, iteratively advancing model-based and data-driven methods, validating designs through usability studies, and assessing the societal and economic impact of technologies to impact the future of work. The project incorporates and builds upon improvisation in the everyday seafood processing plant workplace, to design and validate robot co-worker competencies. Complementing the technological objectives, the project includes human-subject studies to address ethical issues raised by the adoption of robotics in the workplace. Addressing these issues -- which include challenges specific to a labor context with immigrant labor, low wages, and harsh working conditions -- is essential to successful robotics integration and adoption. Finally, the project will evaluate the economic impact of automation on both workers and companies in this emerging socio-technological landscape to understand both the costs and the benefits of adopting new technologies that will shape the future workplace.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

人类技术前沿（FW-HTF）项目的未来工作将研究协作机器人的适当开发和部署，以改变海产品加工业的盈利能力、生产力、安全性、可持续性和工人的生活质量，在这些行业中，恶劣的条件和苛刻的危险任务对人类和机器人的能力都提出了挑战。其结果将是设计、工具、方法和数据集，以促进无缝的人机协作。软机器人操作器将增加安全可靠地处理湿滑、有鳞和柔性物体。人工智能的新兴能力将有助于识别和检查各种鱼类和贝类。该项目的关键是了解如何最好地在机器人和人类工人之间分配特定任务，在尊重环境和道德约束的同时，在个体工人、独立企业、国内和移民劳动力市场、国家经济部门和全球贸易的规模上整合一套复杂的预期结果。相关的教育和推广项目将使工程师、设计专家和社会科学家能够应对机器人与操作、人工智能、人机协作、伦理和劳动经济学等领域的挑战。培训计划将在数据、技术和人文素养方面培养一批新的学习者和研究人员。2017年，美国进口的海产品数量创历史新高，相当于170多亿美元的贸易逆差。由于美国国内失业率较低，以及在雇佣移民劳工方面存在障碍，美国海产品加工业一直无法满足美国消费者的需求。该项目将促进美国在全球竞争和国内服务不足的行业中的领导地位，同时促进对关键科学、工程和社会挑战的理解。该项目的研究目标将影响五个相互关联和融合的领域，即(i)协作机器人和共享自治，（ii）数据可视化交互设计，（iii）海鲜行业的劳动经济学，（iv）工作场所自治伦理，(v)劳动力培训和新技能学习。解决今天的缺点？在人机协作团队中，该研究计划旨在显著提高复杂环境中的生产力，同时特别关注影响用户可接受性的因素。该团队将强调机器人系统在现有社会技术背景下的整合——现有的机械和工具以及现有的人类工作实践，包括正式和非正式的模式？开发一种环境和分布式机器人方法，而不是单个机器人干预。该研究计划通过以用户为中心的设计和构思活动，迭代推进基于模型和数据驱动的方法，通过可用性研究验证设计，以及评估技术对未来工作的社会和经济影响，为系统开发一系列设计需求。该项目结合并建立在日常海鲜加工厂工作场所的即兴创作基础上，以设计和验证机器人同事的能力。作为技术目标的补充，该项目还包括人体研究，以解决在工作场所采用机器人技术所引起的伦理问题。解决这些问题——包括移民劳工、低工资和恶劣工作条件下的劳工环境的具体挑战——是成功集成和采用机器人的关键。最后，该项目将评估在这个新兴的社会技术环境中自动化对工人和公司的经济影响，以了解采用将塑造未来工作场所的新技术的成本和收益。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

HYPERBOT: A BENCHMARKING TESTBED FOR ACQUISITION OF ROBOT-CENTRIC HYPERSPECTRAL SCENE AND IN-HAND OBJECT DATA

Hyperbot：用于获取以机器人为中心的高光谱场景和手中物体数据的基准测试台

• 发表时间: 2022
• 期刊: WHISPERS
• 作者: Nathaniel Hanson, Tarik Kelestemur

A 3D Reactive Navigation Algorithm for Mobile Robots by Using Tentacle-Based Sampling

基于触手采样的移动机器人 3D 反应式导航算法

• 发表时间: 2020
• 期刊: 2020 Fourth IEEE International Conference on Robotic Computing (IRC
• 作者: Akmandor, N. and

In-Hand Object Recognition with Innervated Fiber Optic Spectroscopy for Soft Grippers

使用受控光纤光谱技术进行软抓手的手持物体识别

• DOI: 10.1109/robosoft54090.2022.9762166
• 发表时间: 2021
• 期刊: 2022 IEEE 5th International Conference on Soft Robotics (RoboSoft
• 作者: Hanson, Nathaniel;Hochsztein, Hillel;Vaidya, Akshay;Willick, Joel;Dorsey, Kristen;Padir, Taskin
• 通讯作者: Padir, Taskin

Model-Based Manipulation of Linear Flexible Objects: Task Automation in Simulation and Real World

基于模型的线性柔性物体操纵：仿真和现实世界中的任务自动化

• DOI: 10.3390/machines8030046
• 发表时间: 2020
• 期刊: Machines
• 影响因子: 2.6
• 作者: Chang, Peng;Padır, Taşkın
• 通讯作者: Padır, Taşkın

SLURP! Spectroscopy of Liquids Using Robot Pre-Touch Sensing

咕噜咕噜！

• DOI: 10.1109/icra48891.2023.10161084
• 发表时间: 2023
• 期刊: IEEE
• 作者: Hanson, Nathaniel;Lewis, Wesley;Puthuveetil, Kavya;Furline, Donelle;Padmanabha, Akhil;Padir, Taşlan;Erickson, Zackory
• 通讯作者: Erickson, Zackory