Automating Transfer of Robot Task Plans using Functorial Data Migrations

TL;DR

This paper presents a category theory-based method for automating robot task plan transfer across different domains using functorial data migrations, enabling universal plan transfer without replanning.

Contribution

It introduces a novel, domain-agnostic framework leveraging functorial data migrations for robot plan transfer, expanding beyond plan-specific methods.

Findings

Successfully transferred plans from Blocksworld to AI2-THOR Kitchen

Proposed benchmarks for evaluating symbolic plan transfer methods

Discussed practical limitations and future scaling directions

Abstract

This paper introduces a novel approach to ontology-based robot plan transfer by leveraging functorial data migrations, a structured mapping method derived from category theory. Functors provide structured maps between planning domain ontologies which enables the transfer of task plans without the need for replanning. Unlike methods tailored to specific plans, our framework applies universally within the source domain once a structured map is defined. We demonstrate this approach by transferring a task plan from the canonical Blocksworld domain to one compatible with the AI2-THOR Kitchen environment. Additionally, we discuss practical limitations, propose benchmarks for evaluating symbolic plan transfer methods, and outline future directions for scaling this approach.