Learning Context-Adaptive Task Constraints for Robotic Manipulation
This addresses the need for more flexible and automated constraint specification in robotics, though it is incremental as it builds on existing constraint-based control methods.
The paper tackles the problem of manually specifying task constraints for robotic manipulation by automatically deriving and adapting constraints from data, showing improved reproduction accuracy over manually specified constraints on an industrial robot with three dual-arm tasks.
Constraint-based control approaches offer a flexible way to specify robotic manipulation tasks and execute them on robots with many degrees of freedom. However, the specification of task constraints and their associated priorities usually requires a human-expert and often leads to tailor-made solutions for specific situations. This paper presents our recent efforts to automatically derive task constraints for a constraint-based robot controller from data and adapt them with respect to previously unseen situations (contexts). We use a programming-by-demonstration approach to generate training data in multiple variations (context changes) of a given task. From this data we learn a probabilistic model that maps context variables to task constraints and their respective soft task priorities. We evaluate our approach with 3 different dual-arm manipulation tasks on an industrial robot and show that it performs better in terms of reproduction accuracy than constraint-based controllers with manually specified constraints.