Driving · 2026-11-02
ART of PIV: Agentic Real-Time Particle Image Velocimetry for Active Flow Control
Antonio Terpin, Francesco Banelli, Alan Bonomi, Raffaello D'Andrea
AI-assisted summary based on the paper. Consult the linked source for methods and limitations. Benchmark numbers appear after editor verification.
TL;DR
ART of PIV introduces an agentic, real-time PIV pipeline that selects a configured flow-estimation pipeline (from a library) for each image pair within a specified speed budget. Across cross-correlation, classical optical-flow, and learning-based estimators, the authors report 12.2–22.4% lower average end-point error (AEPE) versus the best single estimator at matched speed, with gains up to 1000–4000 image pair/s. They also show reduced closed-loop cost in simulation and physical flow tracking with a DreamerV3 controller.
Why it matters
When PIV is used for active flow control, measurements must be both accurate and fast, and the best estimation configuration can change with conditions. The authors propose selecting among preprocessing/estimator/postprocessing configurations on a per-image-pair basis under a speed constraint, aiming to reduce feedback errors and improve closed-loop performance. They also deploy a learning-based controller using DreamerV3 with physical flow feedback from ART of PIV.
Method
- Formulate PIV pipeline selection as a contextual bandit: for each image pair, an agent chooses one pipeline member from a speed-budget-constrained library to maximize an objective reward (e.g., negative AEPE).
- Train a small CNN agent to predict the preferred library member (classification with cross-entropy) using numerical scores or expert labels; support sequential or speculative execution to limit latency.
- Construct libraries by selecting estimator families/configurations (cross-correlation, optical flow, learning-based) and enforce a cap on library size; evaluate on accuracy and control tasks.
Abstract (from arXiv)
Particle image velocimetry (PIV) measures fluid motion from images of tracer particles, but its accuracy depends on choices throughout the estimation pipeline, including image preprocessing, flow estimation, and postprocessing. These choices are typically left to expert tuning, and a fixed configuration can become less accurate as conditions change. This matters especially when PIV supplies feedback for real-time control, where measurements must be both accurate and fast, and flow conditions vary. We introduce ART of PIV -- short for Agentic Real-Time and also a nod to the art of tuning a PIV pipeline -- which automates pipeline choice for each image pair. A learned agent selects a configuration according to a chosen measurement objective from a library assembled for a specified speed budget. Across cross-correlation, classical optical-flow, and learning-based estimators, ART of PIV reduces average end-point error (AEPE) by 12.2-22.4 % relative to the best single estimator at matched speed, with gains extending to 1000-4000 image pair/s. In a simulated local velocity regulation task, it reduces mean normalized closed-loop cost by 13.24-13.59 percentage points relative to a tuned fixed estimator, with the same classical controller in both cases. In a tabletop water channel, we illustrate two ways to tune ART of PIV for a fluidic pinball of three rotating cylinders: selecting between complementary optical-flow tunings for accuracy, and selecting regions of interest with coherent rotation or shear using a common flow estimator. For the first variant, we train a DreamerV3 reinforcement learning controller directly in the physical world and demonstrate tracking of a prescribed flow field in the wake of a fluidic pinball.
Related papers
- FlowVLA: Enabling World Dynamics Understanding through Pluggable 3D Flow Learning
- FATHOM: World-Model-Guided Diffusion Action-Trajectory Selection for Sparse-Marker AUV Tracking
- MotionCore: A Latent Neural Simulator for Video Generation with Rigid-Body Dynamics
- Readable Is Not Steerable: Pre-Registered Tests of Physical Structure in Learned Fluid Representations
- Audited Gradients, Anchored Policies: Model-Based Reinforcement Learning for Efficient Flow Control
Publication record
Bibliographic evidence describes where and when a record appeared. It does not validate scientific claims.
First public release: Date not supplied · Indexed by Pub-AI: 2026-11-02 UTC
ART of PIV: Agentic Real-Time Particle Image Velocimetry for Active Flow Control
Source date: Date not supplied · Original record ↗
- arxiv
- 2610.41304
Imported from the existing registry. No new publication milestone asserted.
Abstract · arxiv
Particle image velocimetry (PIV) measures fluid motion from images of tracer particles, but its accuracy depends on choices throughout the estimation pipeline, including image preprocessing, flow estimation, and postprocessing. These choices are typically left to expert tuning, and a fixed configuration can become less accurate as conditions change. This matters especially when PIV supplies feedback for real-time control, where measurements must be both accurate and fast, and flow conditions vary. We introduce ART of PIV -- short for Agentic Real-Time and also a nod to the art of tuning a PIV pipeline -- which automates pipeline choice for each image pair. A learned agent selects a configuration according to a chosen measurement objective from a library assembled for a specified speed budget. Across cross-correlation, classical optical-flow, and learning-based estimators, ART of PIV reduces average end-point error (AEPE) by 12.2-22.4 % relative to the best single estimator at matched speed, with gains extending to 1000-4000 image pair/s. In a simulated local velocity regulation task, it reduces mean normalized closed-loop cost by 13.24-13.59 percentage points relative to a tuned fixed estimator, with the same classical controller in both cases. In a tabletop water channel, we illustrate two ways to tune ART of PIV for a fluidic pinball of three rotating cylinders: selecting between complementary optical-flow tunings for accuracy, and selecting regions of interest with coherent rotation or shear using a common flow estimator. For the first variant, we train a DreamerV3 reinforcement learning controller directly in the physical world and demonstrate tracking of a prescribed flow field in the wake of a fluidic pinball.
