TieBot: Learning to Knot a Tie from Visual Demonstration through a Real-to-Sim-to-Real Approach
This work addresses a specific, complex manipulation problem for robotics, with incremental improvements in learning from demonstration methods.
The paper tackles the challenge of teaching robots to knot a tie, a task involving high deformation and long-horizon manipulation, by developing TieBot, a Real-to-Sim-to-Real learning system from visual demonstrations, achieving a 50% success rate in real-world experiments.
The tie-knotting task is highly challenging due to the tie's high deformation and long-horizon manipulation actions. This work presents TieBot, a Real-to-Sim-to-Real learning from visual demonstration system for the robots to learn to knot a tie. We introduce the Hierarchical Feature Matching approach to estimate a sequence of tie's meshes from the demonstration video. With these estimated meshes used as subgoals, we first learn a teacher policy using privileged information. Then, we learn a student policy with point cloud observation by imitating teacher policy. Lastly, our pipeline applies learned policy to real-world execution. We demonstrate the effectiveness of TieBot in simulation and the real world. In the real-world experiment, a dual-arm robot successfully knots a tie, achieving 50% success rate among 10 trials. Videos can be found https://tiebots.github.io/.