发表机构
Jilin University; State Key Laboratory of Special Vehicle Design and Manufacturing Integration Technology; Changchun University of Science and Technology（吉林大学; 特种车辆设计制造集成技术全国重点实验室; 长春理工大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
RecMorph提出拓扑引导的空间递归架构，通过深度优先遍历和双向转换实现高效跨肢体通信，在UNIMAL任务和物理四足机器人上取得最优性能，并降低速度误差43.5%。
AI 中文摘要
通用形态控制要求单一策略能够跨不同物理角色的肢体转换信息，协调全身运动，并在身体规模增长时保持高效。现有的通信机制仅部分满足这些要求。我们提出了RecMorph，一种拓扑引导的空间递归架构，利用递归序列计算联合执行跨肢体通信和表示转换。深度优先遍历将运动学树转换为形态衍生序列，沿该序列共享的双向转换在动作解码前逐步转换肢体信息。残差保留、RMS归一化和输入依赖的通道调制稳定了这种重复的空间转换，在固定模型宽度和深度下实现了线性令牌复杂度。在五个UNIMAL任务中，RecMorph在评估的通用形态控制器中取得了最强的平均最终训练性能，并在FT上实现了最高的测量推理吞吐量，同时泛化到未见变体和多达30个肢体的身体。我们进一步将代表性通用控制器从UNIMAL基准迁移到四平台四足设置。RecMorph在标称和高摩擦下取得了最佳宏平均性能，相对于专家MLP将标称速度RMSE降低了43.5%，并且一个共享策略完成了40次物理Go1/Go2试验而无跌倒。这些结果表明，拓扑引导的递归转换为通用形态控制提供了一种有效且高效的通信机制，并且在从程序化身体迁移到物理机器人平台时仍然有效。代码和实验资源可在此https URL公开获取。
英文摘要
Generalized morphology control requires a single policy to transform information across limbs with different physical roles, coordinate whole-body motion, and remain efficient as body size grows. Existing communication mechanisms address these requirements only partially. We introduce RecMorph, a topology-guided spatial recurrent architecture that uses recurrent sequence computation to jointly perform cross-limb communication and representation transformation. A depth-first traversal converts the kinematic tree into a morphology-derived sequence, along which shared bidirectional transitions progressively transform limb information before action decoding. Residual preservation, RMS normalization, and input-dependent channel modulation stabilize this repeated spatial transformation, yielding linear token complexity at fixed model width and depth. Across five UNIMAL tasks, RecMorph achieves the strongest mean final training performance among the evaluated generalized morphology controllers and the highest measured inference throughput on FT, while generalizing to unseen variations and bodies with up to 30 limbs. We further migrate representative generalized controllers from UNIMAL benchmarks to a four-platform quadruped setting. RecMorph achieves the best macro-averaged performance under nominal and high friction, reduces nominal velocity RMSE by 43.5% relative to specialist MLPs, and one shared policy completes 40 physical Go1/Go2 trials without falls. These results show that topology-guided recurrent transformation provides an effective and efficient communication mechanism for Generalized Morphology Control and remains effective when transferred from procedural bodies to physical robot platforms. Code and experimental resources are publicly available at https://github.com/quanruirao/RecMorph.
Comments26 pages. Code and experimental resources are available at https://github.com/quanruirao/RecMorph
