We propose a Tactile perception-based approach to address the Needle Threading task and name it T-NT. This task is divided into two main stages: Tail-end Finding and Tail-end Insertion. In the first stage, the agent traces the contour of the thread twice using vision-based tactile sensors mounted on the gripper fingers, to locate the tail-end of the thread. In the second stage, it employs a tactile-guided RL model to drive the robot to insert the thread into the target needle eyelet. Extensive experiments on real robots are conducted to demonstrate the efficacy of our method.
These curves demonstrate 5 of our RL training using PPO. We record the average reward of every 100 steps for our PPO training. The average reward have reached around 30, which indicates that the agent can finish the insertion with around 100/30 steps, and this number can help us to decide the threshold of steps to determine the success or failure of the insertion stage.