Bootstrapped LLM Semantics for Context-Aware Path Planning
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866912710544850944 |
|---|---|
| author | Amani, Mani Beheshti, Behrad Akhavian, Reza |
| author_facet | Amani, Mani Beheshti, Behrad Akhavian, Reza |
| contents | Prompting robots with natural language (NL) has largely been studied as what task to execute (goal selection, skill sequencing) rather than how to execute that task safely and efficiently in semantically rich, human-centric spaces. We address this gap with a framework that turns a large language model (LLM) into a stochastic semantic sensor whose outputs modulate a classical planner. Given a prompt and a semantic map, we draw multiple LLM "danger" judgments and apply a Bayesian bootstrap to approximate a posterior over per-class risk. Using statistics from the posterior, we create a potential cost to formulate a path planning problem. Across simulated environments and a BIM-backed digital twin, our method adapts how the robot moves in response to explicit prompts and implicit contextual information. We present qualitative and quantitative results. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_11967 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Bootstrapped LLM Semantics for Context-Aware Path Planning Amani, Mani Beheshti, Behrad Akhavian, Reza Robotics Prompting robots with natural language (NL) has largely been studied as what task to execute (goal selection, skill sequencing) rather than how to execute that task safely and efficiently in semantically rich, human-centric spaces. We address this gap with a framework that turns a large language model (LLM) into a stochastic semantic sensor whose outputs modulate a classical planner. Given a prompt and a semantic map, we draw multiple LLM "danger" judgments and apply a Bayesian bootstrap to approximate a posterior over per-class risk. Using statistics from the posterior, we create a potential cost to formulate a path planning problem. Across simulated environments and a BIM-backed digital twin, our method adapts how the robot moves in response to explicit prompts and implicit contextual information. We present qualitative and quantitative results. |
| title | Bootstrapped LLM Semantics for Context-Aware Path Planning |
| topic | Robotics |
| url | https://arxiv.org/abs/2511.11967 |