DyPSI: Dynamic Physics Sensing via Joint Field and Sensor-Trajectory Generation
Organizations: School of Mathematical Sciences, Ocean University of China · College of Information Science and Electronic Engineering, Zhejiang University · College of Energy Engineering, Zhejiang University
Abstract
Physics sensing in evolving environments requires reconstructing dense spatiotemporal fields from sparse observations while adapting where future observations should be collected. Existing reconstruction and sensor-placement methods have made substantial progress, but they typically rely on static sensing assumptions: they either use a single layout across all frames or optimize placements independently at each frame, which limits their ability to produce temporally coherent sensing trajectories. We introduce DyPSI, a dynamic physics sensing framework that formulates this problem as the joint generation of future fields and sensor trajectories conditioned on historical observations. DyPSI lifts discrete sensor coordinates into continuous Sensor Position Fields, encodes fields and sensing configurations with a shared Functional Tucker representation, and trains a joint diffusion model to generate future fields and sensing trajectories simultaneously in the resulting latent space. A Gaussian-process temporal kernel correlates the perturbation injected during diffusion training, biasing the denoiser toward temporally smooth recoveries, and we further provide an analysis linking the EDM training objective to a sequence-aggregated A-optimality criterion and a kernel-induced bound on discrete trajectory variation. Experiments on turbulent flow, GLORYS12 sea-surface temperature, and 3D car aerodynamics show that DyPSI consistently outperforms static and frame-wise placement strategies, with substantial error reduction under tight sensor budgets. The code is available at https://github.com/mengqi132/DyPSI.
Figures & tables
| Turbulent Flow | Sea Temperature | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Models | Placement | #200 | #100 | #50 | #30 | #100 | #50 | #25 | #15 |
| SSPOR | SSPOR | 0.3018 | 0.4429 | 0.5921 | 0.6637 | 0.0756 | 0.0752 | 0.0789 | 0.0719 |
| VoronoiCNN | Random | 0.3588 | 0.4870 | 0.6919 | 0.7992 | 0.1496 | 0.1537 | 0.2433 | 0.2817 |
| Senseiver | Random | 0.1842 | 0.2316 | 0.3740 | 0.5746 | 0.0715 | 0.0732 | 0.0769 | 0.0784 |
| S 3 GM | Random | 0.1856 | 0.2016 | 0.2681 | 0.6421 | 0.1115 | 0.1137 | 0.1581 | 0.1795 |
| DiffusionPDE | Random | 0.1993 | 0.3312 | 0.6604 | 0.9163 | 0.0664 | 0.0775 | 0.0859 | 0.1139 |
| Models | #200 | #100 | #50 | #30 | #15 |
|---|---|---|---|---|---|
| Senseiver | 0.1009 | 0.1012 | 0.1018 | 0.1022 | 0.1053 |
| DiffusionPDE | 0.0967 | 0.0966 | 0.0987 | 0.1055 | 0.2095 |
| PhySense | 0.0375 | 0.0382 | 0.0395 | 0.0416 | 0.0465 |
| PhySense-opt | 0.0369 | 0.0370 | 0.0370 | 0.0372 | 0.0386 |
| DyPSI | 0.0301 | 0.0301 | 0.0302 | 0.0307 | 0.0311 |
| Improvement | 18.43% | 18.65% | 18.39% | 17.47% | 19.43% |
| Turbulent Flow, 50 sensors | Sea Surface Temperature, 25 sensors | |||||||||
| Strategy | Future GT | Test-time Opt. | Rel. | Viol. | Smooth. | Runtime | Rel. | Viol. | Smooth. | Runtime |
| Static PhySense-opt | 0.1558 | 0 | 0 | 0.0437 | 0 | 0 | ||||
| Per-frame Oracle | 0.1007 | 89% | 142.5 | 887 | 0.0259 | 75% | 165.3 | 628 | ||
| Sequence Teacher | 0.1206 | 5% | 3.5 | 1683 | 0.0355 | 2% | 4.5 | 1429 | ||
| DyPSI | 0.1352 | 8% | 4.2 | 1.9 | 0.0394 | 4% | 9.2 | 2.1 | ||
| Senseiver | PhySense | DiffusionPDE | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Placement Strategy | #200 | #100 | #50 | #30 | #200 | #100 | #50 | #30 | #200 | #100 | #50 | #30 |
| Random Sample | 0.1842 | 0.2316 | 0.3740 | 0.5746 | 0.1233 | 0.1527 | 0.2586 | 0.5176 | 0.1993 | 0.3312 | 0.6604 | 0.9163 |
| Grid Sample | 0.1615 | 0.2273 | 0.3845 | 0.5869 | 0.1235 | 0.1521 | 0.2436 | 0.4930 | 0.1932 | 0.3319 | 0.6703 | 0.9079 |
| Min-Max Sample | 0.1549 | 0.2091 | 0.3795 | 0.5747 | 0.1196 | 0.1533 | 0.3157 | 0.5022 | 0.1889 | 0.3307 | 0.6574 | 0.9010 |
| PhySense-opt | 0.1388 | 0.1750 | 0.3025 | 0.4570 | 0.1106 | 0.1257 | 0.1558 | 0.2157 | 0.1672 | 0.2841 | 0.5319 | 0.7904 |
| DyPSI-opt | 0.1259 | 0.1670 | 0.2863 | 0.4527 | 0.0986 | 0.1101 | 0.1352 | 0.1826 | 0.1569 | 0.2640 | 0.5028 | 0.7116 |
| Noise type | Rel. | Smoothness |
|---|---|---|
| i.i.d. | 0.1515 0.0039 | 18.6 1.9 |
| RBF | 0.1489 0.0025 | 6.2 1.3 |
| Matérn-3/2 | 0.1352 0.0024 | 4.2 1.0 |
| Model | Rel. | Smoothness |
|---|---|---|
| Separate models | 0.1421 0.0045 | 9.1 1.1 |
| Joint model | 0.1352 0.0024 | 4.2 1.0 |
Appendix figures & tables5 assets
Supplementary material from the paper’s appendix.
Appendix
| Aspect | SDIFT [ 5 ] | DyPSI (Ours) |
|---|---|---|
| Task | Full-field generation from irregular sparse observations | Dynamic active sensing: coupled field reconstruction and adaptive path planning |
| Generated object | Physical fields only | Joint fields and sensor trajectories |
| Sensor role | Fixed, exogenous conditioning variable | A jointly generated modality |
| Sensor representation | None | Sensor Position Field: fixed-kernel encoding of discrete coordinates into a continuous surrogate |
| Latent structure | Functional Tucker over fields | Joint core with shared spatial bases, modality-specific channel ranks, and a concatenated unified latent |
| Training supervision | Observed fields | Offline-to-online amortization of a GT-informed sequence-level optimizer |
| Turbulent Flow | Sea Temperature | Car Aerodynamics | |
| Spatial dimension | 2 | 2 | 3 |
| Grid / mesh size | 95,428 vertices | ||
| Channels | 2 (velocity) | 1 (SST) | 1 (pressure) |
| Modality channel ranks | |||
| Spatial ranks | |||
| SPF bandwidth (in grid units) |
| Models | #200 | #100 | #50 | #30 | #15 |
|---|---|---|---|---|---|
| Senseiver | 0.1009 0.0030 | 0.1012 0.0031 | 0.1018 0.0037 | 0.1022 0.0030 | 0.1053 0.0032 |
| DiffusionPDE | 0.0967 0.0028 | 0.0966 0.0028 | 0.0987 0.0039 | 0.1055 0.0031 | 0.2095 0.0060 |
| PhySense | 0.0375 0.0013 | 0.0382 0.0013 | 0.0395 0.0017 | 0.0416 0.0015 | 0.0465 0.0023 |
| PhySense-opt | 0.0369 0.0012 | 0.0370 0.0012 | 0.0370 0.0016 | 0.0372 0.0015 | 0.0386 0.0021 |
| DyPSI | 0.0301 0.0013 | 0.0301 0.0013 | 0.0302 0.0011 | 0.0307 0.0014 | 0.0311 0.0014 |
| Improvement | 18.43% | 18.65% | 18.39% | 17.47% | 19.43% |
| Benchmark | #Sensors | Static (PhySense-opt) | Dynamic (Ours) | Relative Improvement |
|---|---|---|---|---|
| Turbulent Flow | 50 | 0.1558 | 0.1206 | 22.59% |
| Sea Temperature | 50 | 0.0430 | 0.0345 | 19.77% |
| Car Aerodynamics | 50 | 0.0370 | 0.0263 | 28.92% |
Virtual Sensing
Virtual sensing estimates unavailable or costly-to-measure quantities from other sensor data, physical models, and learned relationships.
See all 27 papers