Robotic in-hand manipulation with relaxed optimization

TL;DR

This paper introduces a learning-based approach for robotic in-hand manipulation that uses motion primitive dictionaries learned from human demonstrations to implicitly satisfy stability constraints without explicit formalization.

Contribution

It presents a novel optimization method that combines motion primitives to generate stable fingertip trajectories for in-hand object manipulation.

Findings

Robotic hand can manipulate objects like humans.

The approach implicitly embeds stability constraints.

It does not require explicit formalization of constraints.

Abstract

Dexterous in-hand manipulation is a unique and valuable human skill requiring sophisticated sensorimotor interaction with the environment while respecting stability constraints. Satisfying these constraints with generated motions is essential for a robotic platform to achieve reliable in-hand manipulation skills. Explicitly modelling these constraints can be challenging, but they can be implicitly modelled and learned through experience or human demonstrations. We propose a learning and control approach based on dictionaries of motion primitives generated from human demonstrations. To achieve this, we defined an optimization process that combines motion primitives to generate robot fingertip trajectories for moving an object from an initial to a desired final pose. Based on our experiments, our approach allows a robotic hand to handle objects like humans, adhering to stability constraints without requiring explicit formalization. In other words, the proposed motion primitive dictionaries learn and implicitly embed the constraints crucial to the in-hand manipulation task.