seuwanglab/Intelligent-Microrobotic-Swarm-in-Unknown-Environments: Intelligent Microrobotic Swarm in Unknown Environments

This work presents a reinforcement learning-based approach for intelligent navigation of microrobotic swarms in unknown environments with dynamic obstacles. The system utilizes Proximal Policy Optimization (PPO) with a Transformer architecture to train agents capable of autonomous exploration and navigation in complex, dynamic environments. The training framework supports both static and dynamic obstacle scenarios, making it suitable for real-world applications in microrobotics.