Proposes Physical Imitation Learning (PIL) to offload up to 87% of a control policy's mechanical power to passive robotic joints.
April 2, 2026
Original Paper
A Physical Imitation Learning Pipeline for Energy-Efficient Quadruped Locomotion Assisted by Parallel Elastic Joint
arXiv · 2604.00611
The Takeaway
It provides a computationally efficient way to achieve brain-body co-design by distilling RL policies into physical joint responses. This allows robots to exploit intrinsic dynamics for energy efficiency without the massive search space of traditional co-optimization.
From the abstract
Due to brain-body co-evolution, animals' intrinsic body dynamics play a crucial role in energy-efficient locomotion, which shares control effort between active muscles and passive body dynamics -- a principle known as Embodied Physical Intelligence. In contrast, robot bodies are often designed with one centralised controller that typically suppress the intrinsic body dynamics instead of exploiting it. We introduce Physical Imitation Learning (PIL), which distils a Reinforcement Learning (RL) con