Max Robotics — Global Service Robot Directory + US Certification

About

Guangxi Wan is a rising researcher at the forefront of intelligent robotic systems for agile manufacturing, with a focus on deep reinforcement learning, continual learning, and sim-to-real transfer. His work addresses critical bottlenecks in industrial automation, particularly how robots can learn and adapt in unstructured, dynamic environments without catastrophic forgetting. In his most cited work (16 citations), Wan designed and implemented an agent-based robotic system for the ARIAC 2021 competition, demonstrating practical solutions for flexible manufacturing. He further advanced the field with a Kalman Filter-based one-shot sim-to-real transfer learning approach (9 citations), enabling deep reinforcement learning policies to transfer from simulation to physical robots with minimal real-world data—a breakthrough for equipment safety and lifespan constraints. Wan also proposed a guided policy search method enhanced with memory-aware synapses (7 citations) to mitigate catastrophic forgetting in continual learning scenarios, allowing robots to sequentially master multiple tasks. His work on optimizing robotic task sequencing and trajectory planning (7 citations) revealed the synergistic potential between these traditionally separate problems. Through his contributions to the ARIAC competitions and novel learning frameworks, Wan is shaping the next generation of autonomous, adaptable industrial robots.