Achieving Scalable Robot Autonomy via neurosymbolic planning using lightweight local LLM

TL;DR

Gideon enables scalable, on-device neurosymbolic planning for robots by leveraging lightweight local LLMs with extended context, improving multi-domain adaptability and efficiency over traditional remote models.

Contribution

The paper introduces Gideon, a framework that adapts neurosymbolic planning to small, local LLMs with a novel dataset generator, enhancing scalability and multi-domain support in robot autonomy.

Findings

66.1% valid plan rate with Qwen-2.5 1.5B model on single domain

70.6% planning validity in multi-domain scenarios

120x smaller model with significant inference efficiency advantages

Abstract

PDDL-based symbolic task planning remains pivotal for robot autonomy yet struggles with dynamic human-robot collaboration due to scalability, re-planning demands, and delayed plan availability. Although a few neurosymbolic frameworks have previously leveraged LLMs such as GPT-3 to address these challenges, reliance on closed-source, remote models with limited context introduced critical constraints: third-party dependency, inconsistent response times, restricted plan length and complexity, and multi-domain scalability issues. We present Gideon, a novel framework that enables the transition to modern, smaller, local LLMs with extended context length. Gideon integrates a novel problem generator to systematically generate large-scale datasets of realistic domain-problem-plan tuples for any domain, and adapts neurosymbolic planning for local LLMs, enabling on-device execution and extended context for multi-domain support. Preliminary experiments in single-domain scenarios performed on Qwen-2.5 1.5B and trained on 8k-32k samples, demonstrate a valid plan percentage of 66.1% (32k model) and show that the figure can be further scaled through additional data. Multi-domain tests on 16k samples yield an even higher 70.6% planning validity rate, proving extensibility across domains and signaling that data variety can have a positive effect on learning efficiency. Although long-horizon planning and reduced model size make Gideon training much less efficient than baseline models based on larger LLMs, the results are still significant considering that the trained model is about 120x smaller than baseline and that significant advantages can be achieved in inference efficiency, scalability, and multi-domain adaptability, all critical factors in human-robot collaboration. Training inefficiency can be mitigated by Gideon's streamlined data generation pipeline.