Force/Torque(F/T) sensing technology enables a dexterous robot control such as direct teaching, master-slave system, and pick-and-place task. In general, 6-axis F/T sensor is attached to the end-effector of the robot manipulator to assist in utilizing advanced robot systems. However, in actual applications, various tools such as robotic grippers, robotic hand, grinders are attached to the sensor and it causes F/T offsets with respect to the gravity. In this letter, Autonomous Weight Compensating(AWC) technique for 6-axis F/T sensor is presented. The proposed AWC technique can reduce the F/T offsets by estimating the F/T offsets through installed Inertial Measurement Unit(IMU) sensor. In this study, the 6-axis F/T are measured based on capacitance sensing scheme and to estimate the orientation of the sensor, a 9-axis IMU sensor is installed inside of the sensor. Then, the F/T offsets are calibrated via Artificial Neural Network(ANN) model. Finally, the performance of the proposed method is demonstrated through comparing the F/T data with both trained data and untrained data.

中文翻译：

---

具有适用于机器人应用的新型自主重量补偿功能的 6 轴力/扭矩传感器

Force/Torque(F/T)传感技术实现了机器人的灵巧控制，如直接示教、主从系统和取放任务。通常，6 轴 F/T 传感器连接到机器人机械手的末端执行器，以协助利用先进的机器人系统。然而，在实际应用中，机器人抓手、机器人手、磨床等各种工具都连接到传感器上，这会导致相对于重力的 F/T 偏移。在这封信中，介绍了用于 6 轴 F/T 传感器的自主重量补偿 (AWC) 技术。所提出的 AWC 技术可以通过安装的惯性测量单元 (IMU) 传感器估计 F/T 偏移来减少 F/T 偏移。在本研究中，基于电容传感方案测量 6 轴 F/T 并估计传感器的方向，传感器内部安装了一个 9 轴 IMU 传感器。然后，通过人工神经网络 (ANN) 模型校准 F/T 偏移。最后，通过将 F/T 数据与训练数据和未训练数据进行比较，证明了所提出方法的性能。