ABPolicy: Asynchronous B-Spline Flow Policy for Real-Time and Smooth Robotic Manipulation

2026-02-27

AI Summary

This work addresses the limitations of existing robotic manipulation methods, which often suffer from trajectory jitter, discontinuous segments, and execution pauses due to synchronous inference, thereby compromising motion smoothness and environmental responsiveness. To overcome these issues, the authors propose an asynchronous flow-matching strategy that generates trajectories in the B-spline control-point action space. By integrating bidirectional action prediction with refitting-based optimization, the approach ensures both intra-segment and inter-segment motion continuity while enabling truly real-time responsiveness. Combining B-spline representation, asynchronous inference, and a flow-matching framework, the method significantly reduces trajectory jerk across seven static and dynamic tasks, markedly improving motion smoothness and task success rates.

Technology Category

• Intelligent Robots: Manipulation
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Robotic manipulation requires policies that are smooth and responsive to evolving observations. However, synchronous inference in the raw action space introduces several challenges, including intra-chunk jitter, inter-chunk discontinuities, and stop-and-go execution. These issues undermine a policy's smoothness and its responsiveness to environmental changes. We propose ABPolicy, an asynchronous flow-matching policy that operates in a B-spline control-point action space. First, the B-spline representation ensures intra-chunk smoothness. Second, we introduce bidirectional action prediction coupled with refitting optimization to enforce inter-chunk continuity. Finally, by leveraging asynchronous inference, ABPolicy delivers real-time, continuous updates. We evaluate ABPolicy across seven tasks encompassing both static settings and dynamic settings with moving objects. Empirical results indicate that ABPolicy reduces trajectory jerk, leading to smoother motion and improved performance. Project website: https://teee000.github.io/ABPolicy/.

Problem

• robotic manipulation
• smoothness
• synchronous inference
• action discontinuity
• real-time responsiveness

Innovation

• B-spline
• asynchronous inference
• flow matching
• trajectory smoothness
• robotic manipulation