FW-HTF-P/Collaborative Research: Anthropocentric Robot Collaboration in Construction

劳动密集型建筑业占美国经济的很大一部分。然而，建筑业面临着严重的职业伤害/死亡、生产力停滞、缺乏熟练工人和劳动力老龄化等问题。作为解决这些问题的一种手段，建筑行业正在逐步采用机器人自动化，特别是在人机协作中，人类和机器人在非结构化和动态的建筑环境中一起工作。尽管最近机器人的功能和能力取得了进步，但人机协作中的许多基本问题仍未得到解答。这些包括：1；机器人如何与人类工人一起工作，并在同一空间中建立和维持信任？2. 为这种人机协作设计未来建筑任务和工作环境的最佳策略是什么？3. 在这个前所未有的新工作环境下，建造业如何再培训现有工人并吸引新工人？为了回答这些问题，本项目开展了一项以人为中心的调查，在该调查中，人类工人对建筑中人机协作的不同场景的反应是非侵入性的，并持续监控，以最大限度地提高人机协作的整体性能。调查的结果有可能为我们如何为未来的建设准备现有和新的劳动力建立基础知识。研究人员已经确定了人机构建的三个主要开放研究领域。首先是深入了解人类在人机协作中的身体、认知和情感反应，这是允许机器人的适应性行为被校准为人类反应的关键知识。二是探索如何重新设计施工任务、操作和作业地点，以更好地服务于人机协作。研究这种设计可能涉及在现实世界中进行大量可能不切实际或不可行的测试，因此需要创新的测试技术和方法。三是为未来人机协同建设培训新老工人的策略。在本项目中，研究人员将在上述研究领域的背景下，测试以下三种研究活动的可行性：1)测试可穿戴生物传感器，以无创、连续地测量人类工人在人机协作中的情绪反应；2)研究虚拟现实作为某些真实构建测试的替代方案，用于重新设计人机协作的任务和工作场所的可行性；3)组织两场研讨会，收集不同利益相关者和专家的广泛观点，以完善未来建筑领域人机合作的研究重点。该项目的最终目标是发展必要的研究人员、研究基础设施和基础工作，以扩大在FW-HTF全面研究计划的水平上研究未来技术、未来工人和未来工作的机会。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。

Labor-intensive construction accounts for a significant portion of the U.S. economy. However, construction suffers from significant occupational injuries/deaths, stagnant productivity, a lack of skilled laborers, and an aging workforce. As one means to address these issues, the construction industry is gradually adopting robotic automation, particularly in human-robot collaboration where humans and robots work together in unstructured and dynamic construction environments. Despite recent advancement of the functionality and capability of robots, many fundamental questions in human-robot collaboration remain unanswered. These include: 1. How can a robot work with a human worker, and build and maintain trust when they work together in the same space? 2. What are the best strategies to design future construction tasks and work environments for such human-robot collaboration? 3. How can the construction industry retrain existing workers and attract new ones in this new and unprecedented working environment? To answer these questions, this project carries out a human-centered investigation where a human worker's response to different scenarios of human-robot collaboration in construction is non-invasively and continuously monitored in order to maximize the overall performance of human-robot collaboration. The outcome of the investigation has the potential to build foundational knowledge on how we can prepare our existing and new workforce for future construction. Researchers have identified three major open research areas in human-robot construction. First is the in-depth understanding of human's physical, cognitive and emotional response in human-robot collaboration, a critical knowledge in allowing a robot's adaptable behaviors to be calibrated to a human response. Second is to explore how to redesign construction tasks, operations, and job sites to better serve human-robot collaboration. Studying such designs could involve a significant number of tests in the real world which may be impractical or infeasible, thus innovative testing technologies and methods are needed. Third involves strategies for training new and existing workers for future human-robot collaborative construction. In this project, researchers will test the feasibility of the three following research activities in the context of the above research areas: 1) testing wearable biosensors to non-invasively and continuously measure a human worker's emotional response in human-robot collaboration; 2) investigating the feasibility of virtual reality as an alternative to certain real construction tests for redesigning tasks and job sites for human-robot collaboration; and 3) organizing two workshops to collect broad perspectives from different stakeholders and experts to refine research thrusts for future human-robot collaboration in construction. The ultimate goal of this project is to develop the necessary research personnel, research infrastructure, and foundational work to expand the opportunities for studying future technology, future workers, and future work at the level of a FW-HTF full research proposal.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.