Embodied Zero-Shot Task Planning via Hardware-Aware LLM-VLM Integration for Grounded Robotic Manipulation
Embodied Zero-Shot Task Planning via Hardware-Aware LLM-VLM Integration for Grounded Robotic Manipulation
This research presents a framework that combines large language models (LLMs) and vision-language models (VLMs) for robotic manipulation. The system separates symbolic reasoning from metric execution, using an LLM planner for task information and a VLM verifier for semantic confirmation, while deterministic modules handle physical execution. The framework is evaluated in simulated environments with tasks like pickup and sequential manipulation.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- Environment
- SimulationOpen source S5
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
release verification, and final visual assessment. Execution feedback is returned only as symbolic context for replanning. The framework is evaluated through simulated digital-twin trials involving pickup, occluded-source access, keepout reasoning, relation placement, and sequential manipulation tasks.
Open source S5
Source：arXiv Robotics — research abstracts · arxiv.org