Dual system model improves robot planning speed and success
DualWAM: Dual-System World Action Models for Asynchronous Global Planning and Local Refinement
Robotics
Summary
Planning robot actions by predicting future visuals is slow, which makes robots less responsive. The authors introduce DualWAM, a method that separates long-term global planning from short-term local adjustments. This allows robots to plan broadly while quickly updating actions based on recent feedback. Their approach worked better and faster than previous methods in robot manipulation tasks.
What this means in practice
- •For robot developers: Enable robots to perform complex manipulation tasks with faster and more responsive planning using DualWAM’s dual global and local update system.
- •For edge computing engineers: Deploy robot planning systems that reduce communication overhead by asynchronously splitting global plans and local refinements between cloud and edge devices.
Authors
Yixin Zheng, Jiangran Lyu, Yuntian Deng, Kai Liu, Yizhou Zhou, Yizhou Wang, Xiaoguang Zhao, He Wang, Zhizheng Zhang
Abstract
World Action Models (WAMs) jointly generate robot actions and predict future world states, transferring priors from video pretraining to robot control. However, future visual prediction is computationally expensive, so existing WAMs often rely on long action chunks to amortize inference cost across control steps, at the cost of closed-loop responsiveness. We present \method, a dual-system WAM that preserves broader-horizon world-action generation while enabling high-frequency closed-loop action updates by decoupling global planning and local refinement. \systwo periodically performs high-noise bidirectional denoising over a broader world-action chunk to establish a global plan, while wrist-only \sysone extracts a temporally aligned short window from the intermediate denoising state and completes low-noise refinement using the latest wrist observations, which provide action-aligned cues about local geometry, motion, and contact during interaction. The two systems operate asynchronously along a shared denoising trajectory: each global plan is reused across multiple local updates, while \sysone repeatedly incorporates fresh interaction feedback. Across zero-shot manipulation tasks on Franka and Galbot, \method improves success over the strongest evaluated baseline by 4.5 percentage points on average, while achieving a 16.6$\times$ critical-path speedup. Further studies show that role-matched egocentric and UMI data improve success by 14 percentage points, and that the decoupled design naturally supports edge--cloud deployment with substantially lower communication overhead than the baseline.