黄海丰, 刘培森, 李擎, 于欣波. 协作机器人智能控制与人机交互研究综述[J]. 工程科学学报, 2022, 44(4): 780-791. DOI: 10.13374/j.issn2095-9389.2021.08.31.001
引用本文: 黄海丰, 刘培森, 李擎, 于欣波. 协作机器人智能控制与人机交互研究综述[J]. 工程科学学报, 2022, 44(4): 780-791. DOI: 10.13374/j.issn2095-9389.2021.08.31.001
HUANG Hai-feng, LIU Pei-sen, LI Qing, YU Xin-bo. Review: Intelligent control and human-robot interaction for collaborative robots[J]. Chinese Journal of Engineering, 2022, 44(4): 780-791. DOI: 10.13374/j.issn2095-9389.2021.08.31.001
Citation: HUANG Hai-feng, LIU Pei-sen, LI Qing, YU Xin-bo. Review: Intelligent control and human-robot interaction for collaborative robots[J]. Chinese Journal of Engineering, 2022, 44(4): 780-791. DOI: 10.13374/j.issn2095-9389.2021.08.31.001

协作机器人智能控制与人机交互研究综述

Review: Intelligent control and human-robot interaction for collaborative robots

  • 摘要: 协作机器人是一类能够在共享空间中与人类交互或在人类附近安全工作的新型工业机器人,由于其轻质、安全的特点,在柔性制造、社会服务、医疗健康、防灾抗疫等多个领域展现出了良好的应用前景,受到工业界和学术界的广泛关注,成为当前机器人领域的研究热点之一。协作机器人需要具备良好的控制性能确保与人交互的安全性,集成多种传感器感知外部环境并应用智能控制理论与方法来确保高效的协作行为。在我国,人机协作已列入《智能制造2025》和《新一代人工智能发展规划》重点支持研究计划。本文主要介绍了国内外几款常见的协作机器人,机器人基于感知信息的控制、高精度跟踪控制、交互控制等智能控制方法,并围绕机器人与人执行协作任务的高效性,对机器人的人类意图估计和技能学习方法进行了讨论。最终对协作机器人未来的发展方向进行了展望。

     

    Abstract: The aggravating trend of an aging population impacts industrial production and social services. Robots are expected to be able to work not only in highly structured manufacturing environments but also in human-inhabited environments, and hence, need to have more sophisticated cognitive abilities. They have to be able to operate safely and efficiently in unstructured, populated environments and achieve high-level collaboration and communication with humans. Collaborative robots, also referred to as cobots, are a new class of industrial robots that can interact with humans in shared spaces or work safely in the vicinity of humans. Collaborative robots are generally lightweight and edge-rounded with multiple degrees of freedom. Besides, multiple sensors must be integrated and limitations of speed and force must be set to ensure their behavior safety. Collaborative robots have shown good application prospects in many fields, such as flexible manufacturing, social services, medical care, disaster prevention, and antiepidemic. They have received wide attention in the industry and academia. Collaborative robots require the integration of multimodal sensory information and intelligent control methods to ensure efficient collaborative behavior. Human-robot collaboration (HRC) considers key issues attached to how safe and efficient collaboration between cobots and humans can be achieved, involving robotics, cognitive sciences, machine learning, artificial intelligence, philosophy, and others. HRC has been included in the key support research programs such as Smart Manufacturing 2025 and the Development Plan of New Generation Artificial Intelligence, recently becoming an important research direction in the field of intelligent robotics with a wide range of applications. This paper introduces several domestic and foreign collaborative robots and intelligent control methods of collaborative robots, including control methods based on perception information, high accuracy tracking control methods, and interaction control methods. It also discusses human intention estimation and robot skill learning methods for efficient human-robot collaboration. Finally, future directions of collaborative robots are explored.

     

/

返回文章
返回