This paper introduces a contact force estimation method of the DMJ that utilizes the KF fusion based on the motor current and the tactile sensor information. The method incorporates the coupled dynamic model of the manipulator system and tactile feedback, demonstrating excellent performance in tracking both large and small external force. It outperforms using current or tactile sensors estimation methods alone, thereby improving the precision of joint contact force estimation.However, this paper does not consider the estimation of small contact force. In the future, we will conduct more in-depth research on contact force estimation for information fusion of collaborative robots and apply it to the flexible control of robot arms, which allows robots to interact more flexibly and reliably with humans or their environment.
