Towards Orientation Learning and Adaptation in Cartesian Space

TL;DR

This paper presents a novel method for learning and adapting robot end-effector orientations and angular velocities from demonstrations, enabling smooth and flexible orientation control in Cartesian space, including quaternion-based trajectories.

Contribution

It introduces a kernelized approach for learning orientation trajectories with high-dimensional inputs and extends to quaternion learning with acceleration constraints, improving adaptability and smoothness.

Findings

Effective orientation trajectory learning demonstrated on real robot arms.

Ability to adapt learned orientations to new points and velocities.

Smoother orientation profiles achieved with quaternion constraints.

Abstract

As a promising branch of robotics, imitation learning emerges as an important way to transfer human skills to robots, where human demonstrations represented in Cartesian or joint spaces are utilized to estimate task/skill models that can be subsequently generalized to new situations. While learning Cartesian positions suffices for many applications, the end-effector orientation is required in many others. Despite recent advances in learning orientations from demonstrations, several crucial issues have not been adequately addressed yet. For instance, how can demonstrated orientations be adapted to pass through arbitrary desired points that comprise orientations and angular velocities? In this paper, we propose an approach that is capable of learning multiple orientation trajectories and adapting learned orientation skills to new situations (e.g., via-points and end-points), where both orientation and angular velocity are considered. Specifically, we introduce a kernelized treatment to alleviate explicit basis functions when learning orientations, which allows for learning orientation trajectories associated with high-dimensional inputs. In addition, we extend our approach to the learning of quaternions with angular acceleration or jerk constraints, which allows for generating smoother orientation profiles for robots. Several examples including experiments with real 7-DoF robot arms are provided to verify the effectiveness of our method.