Physical AI Radar · 2026-10-07

Language: ZH · EN · JA

A daily radar on the Physical AI frontier: eight lanes, trilingual digests, traceable evidence.

• Generated: 2026-10-07 07:19 UTC
• Window: papers 2026-10-03 → 2026-10-07 · posts 2026-09-07 → 2026-10-07 · no repeats within 30 days (UTC)
• Evidence tags: O = first-party official · R = paper/preprint · M = media/secondary

Today’s picks

1. Demo: Closed-Loop Sionna-Isaac Sim Co-Simulation Framework for Wireless-Aware Robot Navigation over ROS 2

• Lane: Simulation & evaluation (sim2real / closed-loop) ｜ Evidence: [R] ｜ Source: arXiv 2610.08618 ｜ 2026-10-06
• Key numbers: 16ms (“is refreshed in ~ 16ms (~60 Hz)”) · 60 Hz (“refreshed in ~16ms (~ 60 Hz )”) · 7.4% (“outage at only + 7.4% traversal time over”)
• Signals: quantitative results · closed-loop success reported · latency or control-rate numbers
• Why it matters: Preprint, not yet peer-reviewed; the text gives no real-robot results. Watch how well the simulated result predicts real-robot performance.

A robot that offloads its control loop to the network carries the receiver with it, so link quality is decided by where it goes.

2. StairVLA: Stage-Aware Hierarchical Action Generation for Vision-Language-Action Models

• Lane: Simulation & evaluation (sim2real / closed-loop) ｜ Evidence: [R] ｜ Source: arXiv 2610.07756 ｜ 2026-10-06
• Key numbers: 96.5% (“improves average success from 96.5% to 97.8% while”) · 97.8% (“success from 96.5% to 97.8% while reducing amortized”) · 115.0 ms (“amortized inference latency from 115.0 ms to 44.2 ms”) · 44.2 ms (“from 115.0 ms to 44.2 ms per action chunk”)
• Signals: quantitative results · closed-loop success reported · latency or control-rate numbers
• Why it matters: Preprint, not yet peer-reviewed; the text gives no real-robot results. Watch how well the simulated result predicts real-robot performance.

Vision-language-action (VLA) models increasingly rely on diffusion- or flow-matching-based action heads to generate continuous robot actions.

3. ESP: Energy-Score Policy for One-Step Multimodal Action Generation

• Lane: Foundation models (VLA / WAM) ｜ Evidence: [R] ｜ Source: arXiv 2610.07696 ｜ 2026-10-06
• Signals: closed-loop success reported · real-robot experiments · latency or control-rate numbers
• Why it matters: Preprint, not yet peer-reviewed; the text gives no figures. Watch whether it cuts the data a new task or embodiment needs.

Generative action models based on diffusion and flow matching have been increasingly adopted in vision-language-action (VLA) policies for their ability to capture diverse behaviors, including multiple valid action sequences under the same…

4. Adapting Vision-Language-Action Models to Unknown Visual Disruptions During Execution

• Lane: Foundation models (VLA / WAM) ｜ Evidence: [R] ｜ Source: arXiv 2610.07946 ｜ 2026-10-06
• Key numbers: 43.9% (“persistent visual corruptions from 43.9% to 53.2% with”) · 53.2% (“corruptions from 43.9% to 53.2% with SmolVLA and”) · 58.7% (“with SmolVLA and from 58.7% to 66.0% with”) · 66.0% (“and from 58.7% to 66.0% with GR00T N1.7”)
• Signals: quantitative results · real-robot experiments
• Why it matters: Preprint, not yet peer-reviewed; the text gives no closed-loop results. Watch whether it cuts the data a new task or embodiment needs.

Visual disruptions can arise while a robot is executing a task, leaving a vision-language-action (VLA) policy to respond without knowing the disruption type or timing.

5. LightVLN: Efficient Aerial Vision-and-Language Navigation with Compact Memory and History-Guided Local Aggregation

• Lane: Edge & real-time (on-device inference / control rate) ｜ Evidence: [R] ｜ Source: arXiv 2610.05024 ｜ 2026-10-04
• Key numbers: 50.93% (“LightVLN achieves 50.93% Test-Seen and 36.14%”) · 36.14% (“achieves 50.93% Test-Seen and 36.14% Test-Unseen success rates”) · 25.83% (“It also achieves 25.83% SR on AerialVLN-S”) · 16 GB (“external Jetson Orin NX 16 GB for closed-loop onboard-compute”)
• Signals: quantitative results · closed-loop success reported · latency or control-rate numbers
• Why it matters: Preprint, not yet peer-reviewed; the text gives no real-robot results. Watch whether the speed-up holds on the target hardware at control rate.

Aerial vision-and-language navigation (VLN) enables unmanned aerial vehicles to execute long-horizon natural-language instructions from visual observations in complex three-dimensional environments.

6. ActTune: Action-Aware Precision and GPU Operating-Point Adaptation for Energy-Efficient Vision-Language-Action Inference

• Lane: Edge & real-time (on-device inference / control rate) ｜ Evidence: [R] ｜ Source: arXiv 2610.08444 ｜ 2026-10-06
• Signals: closed-loop success reported · latency or control-rate numbers
• Why it matters: Preprint, not yet peer-reviewed; the text gives no real-robot results or figures. Watch whether the speed-up holds on the target hardware at control rate.

Vision-language-action (VLA) policies repeatedly invoke inference to control robots, making graphics processing unit (GPU) energy a recurring cost of task execution.

7. ルネサスが低耐圧GaNデバイス市場に参入、業界トップクラスの性能指数を実現

• Lane: Embodiment & supply chain (hardware / cost) ｜ Evidence: [M] ｜ Source: MONOist (ITmedia) ｜ 2026-10-07
• Why it matters: Secondary report, so check the primary source before citing; the text gives no figures. Watch price, availability and who is shipping it in volume.

ルネサス エレクトロニクスは、同社初となる100V耐圧のEモードGaNパワートランジスタ製品群を発売した。AIデータセンターやヒューマノイドロボットなど向けに、超高速スイッチングと優れた放熱性能を実現する。

8. Boston Dynamics appoints former Amazon exec Rohit Prasad as new CEO

• Lane: Embodiment & supply chain (hardware / cost) ｜ Evidence: [M] ｜ Source: The Robot Report ｜ 2026-10-06
• Why it matters: Secondary report, so check the primary source before citing; the text gives no figures. Watch price, availability and who is shipping it in volume.

Before Boston Dynamics, Prasad served as senior vice president and head scientist for Alexa and artificial general intelligence at Amazon.

Frontier baseline (curated)

1. EgoScale: scaling dexterous manipulation with diverse egocentric human data

• Lane: Data engine (collection / curation / scaling) ｜ Evidence: [R] ｜ Source: arXiv 2602.16710 ｜ 2026
• Key numbers: 44k hours of egocentric video · +54% average success vs no-pretraining on a 22-DoF hand · transfers to lower-DoF hands
• Why it matters: Large-scale human motion is shown to be a reusable, embodiment-agnostic motor prior, so dexterous hands do not start from zero.

2. Fine-tuning π0 for robotic manipulation on Amazon SageMaker HyperPod EKS

• Lane: Training & self-improvement (RL / post-training) ｜ Evidence: [O] ｜ Source: AWS Physical AI Blog ｜ 2026-09-10
• Key numbers: MSE -89.7% (DROID) · MSE -88.9% (LIBERO) · 8xH100, 20k steps, 6h15m · 50-step chunk <200 ms at 1 ODE step · checkpoint ~45 GB
• Signals: quantitative results
• Why it matters: An earlier version of the same work reported ~96% because eval episodes overlapped training; enforcing the split moved it to ~89%. The best evaluation-discipline case of 2026.

3. RoboTTT: context scaling for robot policies via test-time training

• Lane: Edge & real-time (on-device inference / control rate) ｜ Evidence: [R] ｜ Source: arXiv 2607.15275 (NVIDIA) ｜ 2026-07-16
• Key numbers: 8K timestep visuomotor context · 3 orders of magnitude beyond prior policies · no inference-latency growth
• Signals: latency or control-rate numbers
• Why it matters: Long context enters the policy without added latency, so one-shot in-context imitation and on-the-fly improvement can live in the same inference pass.

4. Regulation (EU) 2023/1230 becomes fully applicable

• Lane: Safety, permissions & compliance ｜ Evidence: [O] ｜ Source: EUR-Lex / TUV SUD ｜ 2027-01-20
• Key numbers: fully applicable 2027-01-20 · replaces Directive 2006/42/EC
• Why it matters: A hard date you can copy into a project plan: after it, non-compliant machinery cannot be placed on the EU market.

5. ISO 25785-1 (under development): safety for dynamically stable industrial mobile robots

• Lane: Safety, permissions & compliance ｜ Evidence: [O] ｜ Source: ISO ｜ 2026
• Key numbers: committee stage as of 2026-09 · covers legged / self-balancing robots
• Why it matters: Humanoids break the fixed-base premise of ISO 10218; until the dedicated standard lands, safety rests on architecture and runtime guarantees.

6. SHIELD: safety on humanoids via control barrier functions in expectation on learned dynamics

• Lane: Safety, permissions & compliance ｜ Evidence: [R] ｜ Source: arXiv 2505.11494 ｜ 2026
• Key numbers: runtime constraint specification · no retraining when constraints change
• Why it matters: Makes ‘change the constraint without retraining’ possible - the academic counterpart of a local authority gate.

Lane distribution (picks, last 7 days)

Recent cadence

• Source mix (picks, last 7 days): [O] 1 · [R] 40 · [M] 15
• Picks per recent run: 10-01 8 · 10-02 8 · 10-03 8 · 10-04 8 · 10-05 8 · 10-06 8 · 10-07 8

Standards & compliance watch

• ISO 25785-1 — Safety requirements for dynamically stable industrial mobile robots — Under development; committee-stage as of 2026-09.
• Regulation (EU) 2023/1230 (Machinery Regulation) — Fully applicable 2027-01-20.
• EU AI Act official timeline (Council of the EU) — High-risk deadlines 2027-12-02 / 2028-08-02 per Digital Omnibus.

This repository is technical observation. It is not investment advice, a product commitment, or a safety certification conclusion. Paper claims are author-reported.

Methodology · Weekly roundup · Archive