轻型机械臂物理性人机交互碰撞行为及安全反应策略研究

项目名称： 轻型机械臂物理性人机交互碰撞行为及安全反应策略研究

项目编号： No.61263045

项目类型： 地区科学基金项目

立项/批准年度： 2013

项目学科： 自动化技术、计算机技术

项目作者： 熊根良

作者单位： 南昌大学

项目金额： 45万元

中文摘要： 轻型机械臂由于其负载/自重比大、质量轻、具有柔顺性而被广泛的用作服务机器人，应用于非结构化环境以及人类领域，从而不可避免的与人发生身体上的交互。因此，为了保证物理性人机交互过程中人的安全，本项目将以机器人学、生物力学、医学、计算机科学、认知科学等多学科交叉理论研究为基础，建立物理性人机交互碰撞伤害指数模型以及心理伤害指数模型，运用有限元方法和多体动力学仿真相结合分析复杂碰撞情况下的人体变形。同时以创新的基于假人、动物、人体等对象的碰撞实验数据采集和医学影像检查方法为实验手段，修正碰撞伤害指数模型；利用关节力矩传感器以及变阻抗驱动控制方法实现机械臂在物理性人机交互碰撞下的安全反应策略；利用触觉传感器，研究认知性人机交互与物理性人机交互之间相互作用的影响；最后基于虚拟现实技术建立虚拟仿真平台，评估用户在物理性人机交互中的舒适感。本研究对促进机器人物理性人机交互安全技术的发展有重要的意义。

中文关键词： 轻型机械臂；碰撞行为；安全反应策略；物理性人机交互；力矩传感器

英文摘要： Due to the features of high load-to-weight-ratio, lightweight and compliance，lightweight robots are widely used as service robot and applied to the unstructured environment and anthropic domains which inevitably occur the physical interaction with human. Therefore, in order to ensure the safety of human during the physical human-robot interaction. The project will be interdisciplinary theoretical resaerch based on robotics, biomechanics, medicine, computer science, cognitive science, create a physical human-robot interaction collision injury index model as well as psychological injury index model, and combination of the finite element method and multi-body dynamics simulation analysis of the human body deformation in the case of complex collision. Innovative crash test dummies, animals, human and other object-based data acquisition and medical imaging detection methods for the experimental means to correct the collision injury index model. The safe reaction strategies of the robotic arm were achieved by using joint torque sensors and variable impedance actuation control method under the physical human-robot interaction. The interaction between cognitive human-robot interaction and physical human-robot interaction was studied based on tactile sensor. Finally, based on virtual reality technology to create a virtua

英文关键词： lightweight robot arm；collision behavior；safe reaction strategy；physical human-roobot interaction；torque sensor