Learning Spatial Bimanual Action Models Based on Affordance Regions and Human Demonstrations

TL;DR

This paper introduces a method for learning bimanual manipulation actions from human demonstrations by extracting and modeling spatial affordance constraints, enabling robots to understand object interactions for tasks like pouring and rolling.

Contribution

It presents a novel approach to learn and optimize spatial affordance constraints from demonstrations, improving robot understanding of bimanual object interactions.

Findings

The approach successfully models object interactions in simulation for pouring and rolling tasks.

Different definitions of affordance constraints impact the effectiveness of the learned models.

Optimization of object configurations enhances task execution based on learned affordance constraints.

Abstract

In this paper, we present a novel approach for learning bimanual manipulation actions from human demonstration by extracting spatial constraints between affordance regions, termed affordance constraints, of the objects involved. Affordance regions are defined as object parts that provide interaction possibilities to an agent. For example, the bottom of a bottle affords the object to be placed on a surface, while its spout affords the contained liquid to be poured. We propose a novel approach to learn changes of affordance constraints in human demonstration to construct spatial bimanual action models representing object interactions. To exploit the information encoded in these spatial bimanual action models, we formulate an optimization problem to determine optimal object configurations across multiple execution keypoints while taking into account the initial scene, the learned affordance constraints, and the robot's kinematics. We evaluate the approach in simulation with two example tasks (pouring drinks and rolling dough) and compare three different definitions of affordance constraints: (i) component-wise distances between affordance regions in Cartesian space, (ii) component-wise distances between affordance regions in cylindrical space, and (iii) degrees of satisfaction of manually defined symbolic spatial affordance constraints.