LLM-driven agent for speech-enabled control of industrial robots: A case study in snow-crab quality inspection
Bibliographic record
Abstract
This study investigates the integration of large language models (LLMs) into a voice- and vision-based robotic control system in autonomous industrial applications. The main objective is to demonstrate that an LLM-based agent can interpret natural instructions, dynamically plan movements, and execute robotic actions without domain-specific supervised learning, thereby enabling autonomous robotic planning. The proposed system relies on a voice interface, an LLM agent, and tools for real-time robot control. A dedicated communication module was developed to ensure the full control of a KUKA industrial robot using WebSocket, without resorting to proprietary solutions. To validate the approach, a case study was conducted using a robotic cell, which was applied to snow-crab sorting, where computer vision provides real-time perception. The experimental evaluation covered a wide range of commands, including movement instructions, complex planning tasks (e.g., trajectory generation), and visual queries based on crab quality, size, and anatomy. The results showed that the model exhibited robust interpretation capabilities with an overall success rate of 98.46%. These performances highlight the potential of LLMs to facilitate human-robot interaction in real industrial environments, while reducing programming complexity and increasing system autonomy. • A speech-enabled system controls an industrial KUKA robot using a large language model. • Python tools enable real-time motion planning and robotic execution. • Visual inspection of snow crabs guides automated sorting tasks. • Achieves 98.46% overall success rate across complex and noisy commands. • Demonstrates real-world deployment in an industrial robotic cell.
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 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".