This is a MATLAB-based reinforcement learning framework that includes the Proximal Policy Optimization (PPO) algorithm and its multi-agent extension (MAPPO). It supports GPU acceleration and parallel computing, making it suitable for research and engineering applications in control systems.

This project has implemented a servo controller based on the CAN2.0A to FTSCS1.0 protocol, supporting all control commands for FEETECH servo motors.

X5Shrink is a system image compression tool specifically designed for **RDK X5** development boards. It has been deeply adapted for the hardware characteristics and partition structure of RDK X5.

Web Pilot (Browser OS): Run full Node.js+Python+Bash & AI agents purely in the browser with a 10MB WASM file. No cloud, zero deployment.