RoboCompiler：闭链机器人的图原生编译，实现一致的建模、控制与仿真
RoboCompiler: Graph-Native Compilation of Closed-Chain Robots for Consistent Modeling, Control, and Simulation
- Tampere University（坦佩雷大学）
- Korea University（高丽大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
RoboCompiler通过图原生编译闭链机器人，构建统一机械接口，实现一致建模、控制与仿真，显著提升计算效率。
AI中文摘要：
具有运动学闭环、耦合执行器和变化接触的机器人，需要配置、运动、力和动力学的统一模型。然而，这些接口在控制和仿真中常常被分别重建，导致闭合和驱动一致性难以维护。本文提出RoboCompiler，一个图原生框架，将规范机构图编译为共享的机械接口。从物体、关节、坐标系、惯量和执行器端口出发，它构建闭合路径和分析残差雅可比，然后通过秩检查的延拓和校正组装可行配置。切提升将独立速度映射到完整机器人和任务运动，而成对的执行器端口映射保持虚功。约束曲率校正将降阶扩展到加速度和投影刚体动力学，包括浮动基座和支撑模式。循环局部评估、生成的雅可比和依赖感知重用，使得在闭合输入变化时能够进行局部更新。我们在小松挖掘机、Unitree Go2、Franka Panda、Kangaroo和六-UPS Stewart平台上评估了物理闭环和任务诱导约束。高精度约束动力学和独立的Pinocchio检查确认了机械一致性；MuJoCo和Isaac Sim/PhysX执行展示了在原生接触下的任务性能和模型重用。对于Kangaroo，编译将残差和雅可比评估时间减少了96.7%，闭环滚动墙钟时间减少了66.8%，同时保持动力学和控制固定。
英文摘要：
Robots with kinematic loops, coupled actuators, and changing contacts require consistent models of configuration, motion, force, and dynamics. Yet these interfaces are often reconstructed separately for control and simulation, making closure and actuation consistency difficult to maintain. This paper presents RoboCompiler, a graph-native framework that compiles a canonical mechanism graph into a shared mechanical interface. From bodies, joints, frames, inertias, and actuator ports, it constructs closure paths and analytic residual Jacobians, then assembles feasible configurations through rank-checked continuation and correction. A tangent lift maps independent velocities to full robot and task motion, while paired actuator-port maps preserve virtual work. A constraint-curvature correction extends the reduction to accelerations and projected rigid-body dynamics, including floating-base and support modes. Cycle-local evaluation, generated Jacobians, and dependency-aware reuse enable localized updates when closure inputs change. We evaluate physical loops and task-induced constraints on a industrial excavator, Unitree Go2, Franka Panda, Kangaroo, and a six-UPS Stewart platform. High-precision constrained-dynamics and independent Pinocchio checks confirm mechanical consistency; MuJoCo and Isaac Sim/PhysX executions demonstrate task performance and model reuse under native contact. For Kangaroo, compilation reduces residual-and-Jacobian evaluation time by 96.7% and closed-loop rollout wall time by 66.8%, with dynamics and control held fixed.
