LaSeSOM: A Latent and Semantic Representation Framework for Soft Object Manipulation

TL;DR

LaSeSOM introduces a flexible, scalable semantic representation framework for soft object manipulation, enabling shape planning across various geometries and properties, advancing robotic handling capabilities.

Contribution

It presents a novel latent and semantic shape representation framework that is generic and scalable for soft object manipulation tasks.

Findings

Effective shape classification across different geometries.

Enables high-level shape planning for soft objects.

Validated with robotic manipulator experiments.

Abstract

Soft object manipulation has recently gained popularity within the robotics community due to its potential applications in many economically important areas. Although great progress has been recently achieved in these types of tasks, most state-of-the-art methods are case-specific; They can only be used to perform a single deformation task (e.g. bending), as their shape representation algorithms typically rely on "hard-coded" features. In this paper, we present LaSeSOM, a new feedback latent representation framework for semantic soft object manipulation. Our new method introduces internal latent representation layers between low-level geometric feature extraction and high-level semantic shape analysis; This allows the identification of each compressed semantic function and the formation of a valid shape classifier from different feature extraction levels. The proposed latent framework makes soft object representation more generic (independent from the object's geometry and its mechanical properties) and scalable (it can work with 1D/2D/3D tasks). Its high-level semantic layer enables to perform (quasi) shape planning tasks with soft objects, a valuable and underexplored capability in many soft manipulation tasks. To validate this new methodology, we report a detailed experimental study with robotic manipulators.