提出传感器优先的三重智能架构，提升机器人视觉系统精度与效率
[Emerging Ideas] Artificial Tripartite Intelligence: A Bio-Inspired, Sensor-First Architecture for Physical AI
![论文配图：[Emerging Ideas] Artificial Tripartite Intelligence: A Bio-Inspired, Sensor-First Architecture for Physical AI](https://arxiv.org/html/2604.13959v1/figure1.png)
- 仿生设计三层次结构：基础层控安全，小脑层校准传感器，大脑层执行推理
- 动态光照下准确率从53.8%提升至88%，远程推理调用减少43.3%
- 适合嵌入式视觉、机器人感知等对延迟和能耗敏感的物理AI场景
随着AI从数据中心走向机器人与可穿戴设备，单纯扩大模型规模已无法满足需求。物理AI面临严苛的延迟、功耗、隐私与可靠性限制，其性能不仅取决于模型容量，更依赖于在动态环境中通过可控传感器获取信号的质量。本文提出人工三重智能（ATI），一种生物启发的传感器优先型物理AI架构。ATI在系统层面分为三部分：基础层（L1）实现反射式安全与信号完整性控制，小脑层（L2）持续进行传感器校准，皮层推理子系统（跨L3/L4）支持常规技能选择与执行、协调及深度推理。该模块化设计使传感器控制、自适应传感、边缘-云协同执行与基础模型推理可在闭环中共同演进，关键感知与控制保留在本地设备，仅在必要时调用高层推理。我们在移动摄像头原型上验证了ATI在动态光照与运动条件下的性能。在分层推理评估中，相比默认自动曝光设置，ATI（L1/L2自适应传感）将端到端准确率从53.8%提升至88%，同时远程L4调用减少43.3%。结果表明，传感与推理的联合设计对具身AI具有显著价值。
原文摘要 · Abstract (English)
As AI moves from data centers to robots and wearables, scaling ever-larger models becomes insufficient. Physical AI operates under tight latency, energy, privacy, and reliability constraints, and its performance depends not only on model capacity but also on how signals are acquired through controllable sensors in dynamic environments. We present Artificial Tripartite Intelligence (ATI), a bio-inspired, sensor-first architectural contract for physical AI. ATI is tripartite at the systems level: a Brainstem (L1) provides reflexive safety and signal-integrity control, a Cerebellum (L2) performs continuous sensor calibration, and a Cerebral Inference Subsystem spanning L3/L4 supports routine skill selection and execution, coordination, and deep reasoning. This modular organization allows sensor control, adaptive sensing, edge-cloud execution, and foundation model reasoning to co-evolve within one closed-loop architecture, while keeping time-critical sensing and control on device and invoking higher-level inference only when needed. We instantiate ATI in a mobile camera prototype under dynamic lighting and motion. In our routed evaluation (L3-L4 split inference), compared to the default auto-exposure setting, ATI (L1/L2 adaptive sensing) improves end-to-end accuracy from 53.8% to 88% while reducing remote L4 invocations by 43.3%. These results show the value of co-designing sensing and inference for embodied AI.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。