Max Robotics — Global Service Robot Directory + US Certification

About

Wenbo Gao is a roboticist pushing the boundaries of how machines learn and adapt in the real world. His research sits at the intersection of reinforcement learning, evolutionary algorithms, and neural architecture search, with a primary focus on enabling high-speed, dynamic locomotion and manipulation. Gao’s most impactful work, "Rapidly Adaptable Legged Robots via Evolutionary Meta-Learning" (60 citations), pioneered a method that allows robots to instantly adjust their gait to changing terrain or damage—a critical step toward truly autonomous machines. He is perhaps best known for his groundbreaking work on robotic table tennis, where his model-free reinforcement learning system (35 citations) learned to return balls at 100Hz, eventually achieving hundreds of consecutive rallies with human players. This system, detailed in a 2023 deep-dive (17 citations), showcases a fully integrated pipeline of perception, control, and learning. Gao also contributed "ES-ENAS" (9 citations), a scalable algorithm that combines evolution strategies with neural architecture search to automatically design efficient neural network policies. Through these contributions, Gao has demonstrated that combining evolutionary methods with deep learning can produce robots that are not only fast and precise, but also remarkably adaptable to the unpredictable human world.