Manual Maneuverability: Metrics for Analysing and Benchmarking Kinesthetic Robot Guidance

2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)(2022)

引用 0|浏览1
暂无评分
摘要
Kinesthetic teaching of collaborative robots is applied for intuitive and flexible robot programming by demonstration. This enables non-experts to program such robots on the task-level. Multiple strategies exist to teach velocity- or torque-controlled robots and, thus, the maneuverability among commercial robots differs significantly. However, currently there exists no metric that quantifies how "well" the robot can be guided, e.g., how much effort is required to initiate a motion. In this paper, we propose standardized procedures to quantitatively assess robot manual maneuverability. First, we identify different motion phases during kinesthetic teaching. For each phase, we then propose metrics and experimental setups to evaluate them. The experimental protocols are applied to the proprietary teaching schemes of five commercial robots, namely the KUKA LWR iiwa 14, Yuanda Yu+, Franka Emika robot, and Universal Robot's UR5e and UR10e. The experimental comparison highlights distinct differences between the robots and shows that the proposed methods are a meaningful contribution to the performance and ergonomics assessment of collaborative robots.
更多
查看译文
关键词
robot
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要