Papers
3
Total Citations
155
H-Index
2
About
Ben Moran is a leading researcher at the intersection of robotics and artificial intelligence, specializing in deep reinforcement learning (deep RL) for agile, full-body robot control. His work focuses on enabling low-cost, miniature humanoid robots to master complex, dynamic tasks through end-to-end learning from egocentric vision. Moran’s most impactful contribution is demonstrating that deep RL can synthesize sophisticated, safe movement skills for bipedal robots, allowing them to play simplified one-versus-one soccer—a feat requiring active perception, agile locomotion, and long-horizon planning. His seminal 2024 paper on this topic has already garnered 147 citations, underscoring its significance in the field. By training robots with 20 actuated joints using only onboard computation and RGB cameras, Moran has pushed the boundaries of real-world robotics, tackling challenges like active perception and agile control. His work not only advances humanoid robotics but also provides a scalable framework for learning complex behaviors in dynamic environments, making him a pivotal figure in the future of autonomous, physically capable robots.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3