Merge pull request #15 from AlicaQiao/llm_interaction

Add a new branch for LLM interaction

Author: eisDNV

.gitignore

@@ -1,3 +1,5 @@
+C:\Users\HONQIA\Documents\python_workspace\AI_devs\py-crane\llm_interaction\secret_keys.txt
+
 # Byte-compiled / optimized / DLL files
 __pycache__/
 *.py[cod]
@@ -158,3 +160,7 @@ examples/**/*.nc
 
 # Cmake generated files
 CMakeUserPresets.json
+
+llm_interaction/llm_keys.txt
+
+results/*

llm_interaction/ReadMe.md

@@ -0,0 +1,30 @@
+## Project Overview
+LLM interaction for crane simulation,three major capabilities :
+(1) Control crane motion with natural language via LLM
+(2) Generate crane configurations from natural language descriptions
+(3) Run an interactive server for real-time command processing with live visualization
+
+## Features & Where to Find Them
+* **Natural-language motion control**: Parse LLM commands and execute crane physics in real-time with matplotlib visualization
+  Files: `llm_parser.py`, `physics_executor.py`, `demo_llm_control.py`, `llm_command.py`
+
+* **LLM-based crane configuration**: Build crane geometry from text specifications and display animation
+  File: `crane_configuration.py`
+
+* **Interactive server**: Send commands to running server; real-time matplotlib animation displays motion; `/debug` endpoint provides execution status
+  Server: `crane_server.py`
+  Test script: `interactive_test.py`
+
+## Real-time Visualization
+All flows use matplotlib with FuncAnimation to display crane motion in real-time. Watch the animated matplotlib window as commands execute.
+
+## Configuration
+Set environment variables for LLM backend:
+- **Azure**: `AZURE_KEY`, `AZURE_ENDPOINT`
+- **Qwen**: `QWEN_API_URL`, `QWEN_API_KEY`
+
+## Quick Commands
+* **Offline demo**: `python llm_interaction/demo_llm_control.py`
+* **Build crane from description**: `python llm_interaction/crane_configuration.py`
+* **Start server**: `python llm_interaction/crane_server.py`
+* **Test server**: `python llm_interaction/interactive_test.py`