A Data-Efficient Approach to Precise and Controlled Pushing
This work addresses data efficiency in robotic manipulation, offering a solution for tasks where modeling is difficult due to unknown parameters, though it is incremental in applying existing methods like Gaussian processes and model predictive control.
The paper tackled the problem of controlling complex pushing tasks with minimal data by learning a dynamical model from only 10 training points, achieving precise and controlled pushing trajectories.
Decades of research in control theory have shown that simple controllers, when provided with timely feedback, can control complex systems. Pushing is an example of a complex mechanical system that is difficult to model accurately due to unknown system parameters such as coefficients of friction and pressure distributions. In this paper, we explore the data-complexity required for controlling, rather than modeling, such a system. Results show that a model-based control approach, where the dynamical model is learned from data, is capable of performing complex pushing trajectories with a minimal amount of training data (10 data points). The dynamics of pushing interactions are modeled using a Gaussian process (GP) and are leveraged within a model predictive control approach that linearizes the GP and imposes actuator and task constraints for a planar manipulation task.