Grounding Large Language Models for Robot Task Planning Using Closed‐Loop State Feedback
Bibliographic record
Abstract
Planning algorithms break complex problems into sequential steps for robots. Recent work employs Large Language Models (LLMs) to generate robot policies directly from natural language in simulation and real‐world settings. Models such as GPT‐5 generalize to unseen tasks but often hallucinate because they lack sufficient environmental grounding; supplying state feedback improves robustness. We introduce a task‐planning method that uses two LLMs—one for high‐level planning and one for low‐level control—thereby raising task success rates and goal‐condition recall. Our algorithm, BrainBody‐LLM, is inspired by the human neural system, dividing planning hierarchically across the two LLMs and closing the loop with feedback that learns from simulator errors to fix execution failures. Implemented with GPT‐5, BrainBody‐LLM improves task‐oriented success in the VirtualHome environment by 17% over competitive baselines. We also evaluate seven complex tasks in a realistic physics simulator and on a Franka Research 3 robotic arm, comparing our approach with other state‐of‐the‐art LLM planners. Results show that recent LLMs can use raw simulator or controller errors to revise plans, yielding more reliable robotic task execution. Project resources are available here .
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".