Enables 10x faster robot trajectory generation by distilling diffusion models into movement primitives.
March 27, 2026
Original Paper
FODMP: Fast One-Step Diffusion of Movement Primitives Generation for Time-Dependent Robot Actions
arXiv · 2603.24806
The Takeaway
By distilling multi-step diffusion processes into a single-step movement primitive decoder, FODMP enables real-time, closed-loop control of complex robot behaviors. This solves the latency bottleneck that previously prevented diffusion-based policies from being used in high-speed manipulation.
From the abstract
Diffusion models are increasingly used for robot learning, but current designs face a clear trade-off. Action-chunking diffusion policies like ManiCM are fast to run, yet they only predict short segments of motion. This makes them reactive, but unable to capture time-dependent motion primitives, such as following a spring-damper-like behavior with built-in dynamic profiles of acceleration and deceleration. Recently, Movement Primitive Diffusion (MPD) partially addresses this limitation by parame