HCPG-Flow:Hierarchical Contact-Progress Guidance for Flow-Policy Robot Manipulation
2026-07-20 • Robotics
Robotics
AI summaryⓘ
The authors improved a robot control method that picks actions from many possibilities by introducing HCPG-Flow. This new approach helps the robot focus on making progress when touching objects, rather than just moving its hand closer. It scores possible actions based on how much they reduce the distance to completing a task and then selects the best one to perform. They tested it in simulations and real tasks, finding it led to higher success rates and fewer steps needed to finish tasks.
robot manipulationflow policiesSAC-Flowactor-critic methodscontact progressaction selectionmultimodal action distributiontask completionrobot controlsimulated robotics
Authors
Guanghu Xie, Mingxu Li, Shuo Zhang, Yonglong Zhang, Yifan Yang, Yang Liu, Zongwu Xie, Baoshi Cao
Abstract
Flow policies can represent multimodal action distributions for robot manipulation, yet a robot must execute one action at each control step. When several proposals are sampled, critic-based ranking makes data collection depend on value estimates over candidate actions that may be weakly represented in replay. We introduce HCPG-Flow, an analytic rollout-time selector that augments SAC-Flow with hierarchical, object-centric contact-progress guidance while preserving its actor and critic objectives. HCPG switches from end-effector approach to task progress after contact, scores each proposal by the first-order reduction of a task-relevant distance, standardizes scores within the candidate set, and executes a temperature-controlled action embedding. Across ten simulated tasks, HCPG improves mean success over SAC-Flow on both benchmarks, including a 9.5 percentage-point gain on Maniskill. Four physical tasks further show high success with a 17.4% reduction in successful completion steps.Project page: https://hitxraz.github.io/HCPG-Flow/