Skip to content

论文arXiv 2611.233643 天前

Rat v2仿真框架发布:生物神经先验实现无模仿步态学习与机制解析 ​

本文由 AI 根据公开内容整理,可能有误,请以原文为准:Neural Circuit Architectural Priors for Rat Locomotion(arXiv 2611.23364)

核心内容 ​

针对具身仿真中复杂躯体与神经回路协同建模难题,研究团队发布Rat v2高效生物力学躯体模型、基于哺乳动物脊髓回路的Rat NCAP神经架构及配套GPU加速训练设施。该系统无需模仿学习即可自主学习不同速度 locomotion 并完成步态转换,其生成的运动学统计量、步态特征及细胞类型特异性神经活动均与生物学实验数据高度一致,为中等抽象层级的具身智能研究提供了新基准。

技术方案核心在于引入细胞类型特异性连接与新型屈伸肌神经机械接口(flexor-extensor neuromechanical interface),将生物归纳偏置嵌入人工神经网络。分析表明该接口显著影响模型可训练性与步态质量;通过硅基细胞消融实验,系统成功复现已知步态表型并预测分级扰动下的步态变化。这种融合解剖学约束的架构设计,证明了在具身仿真中整合生物先验可同时实现复杂行为生成与神经机制解析。

关键看点 ​

  • 无模仿步态学习:Rat NCAP架构仅凭生物结构先验与强化信号,自主习得多速度运动与步态转换,摆脱对专家演示数据的依赖
  • 神经机械接口:创新屈伸肌接口设计被证实是提升模型收敛效率与步态自然度的关键变量,直接影响仿真训练可行性
  • 细胞级机制验证:硅基消融实验精准复现生物步态表型并预测扰动响应,确立了仿真模型作为神经回路因果推断工具的有效性
  • GPU加速基础设施:构建专用并行仿真与训练管线,支撑高自由度生物力学躯体与细粒度神经动力学模型的联合优化

工程与产业观察 ​

该工作确立了“生物结构先验+具身仿真”的技术范式,证明在算法设计中嵌入解剖学与神经科学约束,可在缺乏大规模示教数据时引导策略涌现。其神经机械接口设计与细胞级验证方法,为四足机器人步态控制算法开发及具身大模型训练中的inductive bias构建提供了可迁移的工程参考与理论依据。

英文摘要 (Abstract) ​

Animals generate behavior through continuous interactions between their neural circuits, bodies, and environment. Embodied simulations study these interactions by testing neural circuit models in biologically realistic settings. Recent efforts have simulated diverse animals, and emerging work on biologically grounded neural architectures has begun to integrate circuit connectivity and single-unit dynamics from experimental data into artificial neural networks, introducing stronger inductive biases and closer mechanistic alignment with biology. How can we develop embodied simulations that enable the study of increasingly complex bodies and neural circuits? In this work, we address the challenge through coordinated advances in the simulated body, neural architecture, and training infrastructure. We introduce Rat v2, an efficient and flexible biomechanical body for simulating rat behavior. To control this body, we develop Rat NCAP, a biologically grounded neural architecture for locomotion based on mammalian spinal circuits that incorporates cell-type-specific connectivity and a novel flexor-extensor neuromechanical interface. To optimize circuit parameters, we develop GPU-accelerated simulation and training infrastructure. The model learns to locomote at different speeds and transition between gaits without imitation learning, producing locomotion statistics, kinematics, and cell-type-specific neural activity consistent with experimental data. Analyses reveal that the neuromechanical interface substantially affects trainability and gait quality, and in silico cell-type ablations reproduce known gait phenotypes and predict gait changes under graded perturbations. Together, these findings support the view that embodied simulations at an intermediate level of abstraction can simultaneously reproduce complex behavior and yield mechanistic insights into the neural circuits that generate it.

作者:Nikhil X. Bhattasali, Jaron Cui, Lerrel Pinto, Grace W. Lindsay

基于本页技术方向关联的整机参数、核心零部件、实测报告与同赛道选型横评。

外部来源

本页事实对应的公开页面,以原文为准。(点击将离开本站)

  1. arXiv 2611.23364:Neural Circuit Architectural Priors for Rat Locomotion ↗访问 2026-11-19