AI & ML
GRACE: Gradient-Free Robot Action Generation via Combined Diffusion-MPPI Posterior Mean Estimation
arXiv cs.RO0 views
arXiv:2607.21661v2 Announce Type: replace Abstract: Diffusion policies generate multimodal robot action sequences from demonstrations, but steering them toward deployment-time constraints typically relies on differentiable guidance costs. This excludes many practical safety constraints, such as binary collision checks, joint limits, and black-box rollout costs that are nondifferentiable. We propose Gradient-free Robot Action generation via Combined diffusion-MPPI posterior mean Estimation (GRACE), which guides a pretrained diffusion policy with Model Predictive Path Integral (MPPI) control using only forward cost evaluations. Building on the common score-ascent structure of diffusion and MPPI, GRACE constructs a cost-conditioned guidance posterior at each reverse step and estimates its mean with a single MPPI update centered at the diffusion reverse mean. For differentiable costs, GRACE recovers conventional gradient guidance under a first-order, matched-covariance approximation. GRACE attains higher success rates than diffusion-based and sampling-based baselines in simulation. On a real 7-DoF manipulator, GRACE achieves 43/45 successes without obstacle collisions, compared with 17/45 for the unguided prior. Code and experiment videos are available at https://rcilab.khu.ac.kr/grace/.
Source:https://arxiv.org/abs/2607.21661PreviousWarpMPC: Large-Batch MPC on GPU via ADMM with Unrolled $LDL^\top$ FactorizationNextFrom Uncertainty to Determinism: Coarse-to-Fine Visual Floorplan Localization without Ray Matching
Related posts
How a Small Team Turns Email and Lead Follow-up Into Scheduled Workflows — Without Giving Up Local Control
0dev.to
How AUMOVIO centralized AI access with Databricks Unity Gateway
0Databricks Blog
What's new for Genie in Microsoft: Copilot Cowork and Chat, Microsoft Teams, and Excel.
0Databricks Blog
Introducing Unity Gateway Insights
0Databricks Blog

Shadow AI Is Shipped to You, One Release Note at a Time
0Imperva Blog
Our Approach to Updating Our Commercial Terms of Service
0Meta Newsroom