A Model Context Protocol server that enables natural language interactive control of Universal Robots collaborative robots, allowing users to control robot motion, monitor status, and execute programs through direct commands to large language models.
English | 简体中文 | 日本語 | 한국어 | Deutsch | Français | Русский язык | ภาษาไทย | Español | العربية | dansk
MCP (Model Context Protocol) is a communication protocol developed by Anthropic (open-sourced in November 2024) that enables large language models (e.g. DeepSeek-V3-0324, DeepSeek-R1, Qwen3) to efficiently access external data/tools/services for delivering more accurate and intelligent responses.
Key Capabilities:
Context Provisioning: Deliver files/database contents as supplementary context Example: Analyzing reports before answering
External Tool Integration: Directly operate local/remote systems (files, APIs, hardware) Example: Automated document organization
Workflow Automation: Combine multiple MCP services for complex tasks
Data Security: Local execution protects sensitive information
Client-Server Architecture:
MCP Client: AI application (initiates requests) MCP Server: Provides data/tool interfaces Communication: JSON-RPC 2.0 compliant
Data Access: Files/databases/memory resources Tool Execution: Predefined functions (SQL queries, file ops) Real-time Updates: Push notifications for data changes Session Management: Connection maintenance
Technical Specification of Tuode Technology's Self-developed nUR_MCP_SERVER Product
Product Overview:
The nUR_MCP_SERVER is an intelligent robot control middleware system built on the MCP (Model Control Protocol) interface, enabling natural language interactive control of industrial robots through integration with large language models (LLMs). Designed with a Client-Server architecture, it supports deep integration with Universal Robots' full range of collaborative robots, revolutionizing the traditional teach pendant programming paradigm for industrial robots.
Core Technical Architecture:
Semantic Parsing Engine
Equipped with a multi-layer Transformer-based NLP processing module, it supports context-aware command parsing (Contextual Command Parsing), achieving end-to-end conversion from natural language to robot control commands with a command recognition accuracy of 98.6%.
Dynamic Script Generation System
An LLM-based code generation framework that automatically converts natural language commands into URScript robot control scripts. It supports real-time syntax validation and safety verification, improving generation efficiency by 12x compared to traditional programming.
Multimodal Control Interface
Core Features:
▶ Natural Language Real-Time Control
Directly drives robot motion (pose control, trajectory planning, I/O operations) via voice/text commands, supporting dynamic parameter injection and real-time motion adjustments.
▶ Intelligent Data Acquisition System
▶ Multi-Robot Collaborative Control
Based on a distributed task scheduling algorithm, it can manage ≤12 UR robots simultaneously when paired with Tuode's MCP-Client, supporting voice cascading commands and cross-device task orchestration.
▶ Adaptive Learning Module
Features an incremental training framework that continuously optimizes command-action mapping through user feedback, with a system iteration cycle of ≤24h.
Technical Specifications:
Users of nUR MCP Server must possess advanced safety awareness and complete UR robot operation training, as LLMs directly control physical robots. Improper operation may cause personal injury or property damage.
If using this software, please cite as follows:
This project adopts a User-Segmented Dual Licensing model.
Basic Principles
Definition of "≤10 people": This refers to situations where the total number of individuals in your organization (including companies, non-profits, government agencies, educational institutions, etc.) who directly or indirectly access, use, or benefit from nUR_MCP_SERVER does not exceed 10. This includes but is not limited to developers, testers, operators, end-users, and users through integrated systems.
MCP Server Development Team, Suzhou Nonead Robotics Technology Co., Ltd.
Tony Ke
Micro Zhu
Anthony Zhuang
Quentin Wang
Discover shared experiences
Shared threads will appear here, showcasing real-world applications and insights from the community. Check back soon for updates!