Learning Descriptor of Constrained Task from Demonstration
This work addresses the challenge for robots in adapting to new objects with similar structures but different contexts, though it appears incremental by building on prior kinematic extraction methods.
The paper tackles the problem of robots learning to interact with constrained objects like doors and drawers from demonstrations, proposing a framework that integrates object information into a task descriptor, and demonstrates generalization to novel books with learned constraints.
Constrained objects, such as doors and drawers are often complex and share a similar structure in the human environment. A robot needs to interact accurately with constrained objects to safely and successfully complete a task. Learning from Demonstration offers an appropriate path to learn the object structure of the demonstration for unknown objects for unknown tasks. There is work that extracts the kinematic model from motion. However, the gap remains when the robot faces a new object with a similar model but different contexts, e.g. size, appearance, etc. In this paper, we propose a framework that integrates all the information needed to learn a constrained motion from a depth camera into a descriptor of the constrained task. The descriptor consists of object information, grasping point model, constrained model, and reference frame model. By associating constrained learning and reference frame with the constrained object, we demonstrate that the robot can learn the book opening model and parameter of the constraints from demonstration and generalize to novel books.