SpeedTuning: Speeding Up Policy Execution with Lightweight Reinforcement Learning
2026-08-10 • Robotics
RoboticsArtificial Intelligence
AI summaryⓘ
The authors study how to make robot actions faster without losing the ability to complete tasks. They point out that current robot programs learned by copying slow human actions can't just be sped up easily. To fix this, they create SpeedTuning, a method that learns the best speed for each action to make robots work quicker. Their tests show SpeedTuning can more than double speed while still doing tasks well, on things like pouring and throwing. This method improves robot performance without needing new training data.
robotic manipulationimitation learningreinforcement learningexecution speedpolicy optimizationdynamic tasksrobot controlspeed tuningtask successlinear interpolation
Authors
David D. Yuan, Tony Z. Zhao, Kaylee Burns, Chelsea Finn
Abstract
While learned robotic policies hold promise for advancing generalizable manipulation, their practical deployment is often hindered by suboptimal execution speeds. Imitation learning policies are inherently limited by hardware constraints and the speed of the operator during data collection. In addition, there are no established methods for accelerating policies learned via imitation, and the empirical relationship between execution speed and task success remains underexplored. To address these issues, we introduce SpeedTuning, a reinforcement learning framework specifically designed to enhance the speed of manipulation policies. SpeedTuning learns to predict the optimal execution speed for actions, thereby complementing a base policy without necessitating additional data collection. We provide empirical evidence that SpeedTuning achieves substantial improvements in execution speed, exceeding 2.4x speed-up, while preserving an adequate success rate compared to both the original task policy and straightforward speed-up methods such as linear interpolation at a fixed speed. We evaluate our approach across a diverse set of dynamic and precise tasks, including pouring, throwing, and picking, demonstrating its effectiveness and robustness in enhancing real-world robotic manipulation.