Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
PathPlanLLM: A LLM based Path-planning framework for Scalable Context Window and Privacy-aware Execution
Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.
Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.ORCID iD: 0000-0002-2173-6039
Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.ORCID iD: 0000-0003-3794-0306
Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering, Signals and Systems.ORCID iD: 0000-0003-0126-1897
2026 (English)In: Mediterranean Conference on Control and Automation (MED), Institute of Electrical and Electronics Engineers Inc. , 2026, p. 515-520Conference paper, Published paper (Refereed)
Abstract [en]

As robots are increasingly deployed in real-world environments, users expect to issue high-level navigation instructions in natural language and have the robot handle the details. A key challenge is that directly applying Large Language Models (LLMs) to path planning over 3D scene graphs exposes three fundamental failure modes: context windows overflow as map size grows (prompt complexity scales as O(|V|), where |V| is the number of graph nodes), geometric hallucinations produce paths over nonexistent edges, and transmitting scene graphs to cloud APIs risks leaking sensitive spatial data about private environments. We present PathPlanLLM, a neurosymbolic framework that resolves all three limitations by shifting the LLM’s role from planner to data engineer. Rather than serializing the scene graph into the prompt, PathPlanLLM provides the LLM only with a lightweight structural schema, the set of object category labels (typically 85-120 for indoor scenes) and uses a three-stage pipeline: a Parser that maps natural language queries to structured intents, a Semantic Linker that resolves categories to unique node identifiers, and a Generative Code Executor (GCE) that prompts the LLM to synthesize, execute, and iteratively debug Python code against a local, sandboxed scene graph. This architecture achieves O(∣Cl) prompt scaling (sublinear in graph size), privacy by design (no instance-level spatial data ever leaves the local system), and formal path validity (every returned path is executionverified or rejected). We evaluate PathPlanLLM on a 404node hierarchical scene graph across 30 test cases spanning eleven query categories, achieving a 97% success rate and outperforming both baselines by a significant margin using the same Qwen2.5-Coder 14B model.

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers Inc. , 2026. p. 515-520
National Category
Computer Sciences
Research subject
Robotics and Artificial Intelligence
Identifiers
URN: urn:nbn:se:ltu:diva-119501DOI: 10.1109/MED70602.2026.11598098Scopus ID: 2-s2.0-105046091019OAI: oai:DiVA.org:ltu-119501DiVA, id: diva2:2094991
Conference
34th Mediterranean Conference on Control and Automation, (MED 2026), Ancona, Italy, 23-26 June, 2026
Note

ISBN for host publication: 979-8-3195-4747-7

Available from: 2026-08-25 Created: 2026-08-25 Last updated: 2026-08-25Bibliographically approved

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full textScopus

Authority records

Kulkarni, NileshKotpalliwar, ShrutiSaradagi, AkshitNikolakopoulos, George

Search in DiVA

By author/editor
Kulkarni, NileshKotpalliwar, ShrutiSaradagi, AkshitNikolakopoulos, George
By organisation
Signals and Systems
Computer Sciences

Search outside of DiVA

GoogleGoogle Scholar

doi
urn-nbn

Altmetric score

doi
urn-nbn
Total: 65 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf