用 NVIDIA Holoscan 5 让机械臂实现失重感
原标题：Make a Robot Arm Feel Weightless with NVIDIA Holoscan 5
AI 摘要
NVIDIA 发布技术博客，介绍如何用 NVIDIA Holoscan 5 构建机器人手臂的重力补偿控制回路。教程基于 Seeed Studio reBot Arm B601-DM 机械臂，使用 Pinocchio 逆动力学求解器和 motorbridge 库，通过 Python API 将三个自定义 Holoscan 算子接入图结构，并在 Jetson Orin Nano、Jetson Thor 和 x86 Ubuntu 24.04 上测试。Holoscan 5 新增机器人控制能力，强调低延迟、低抖动运行时，并内置 Nsight Systems 注解用于检查延迟与抖动。
正文节选
Picture guiding a robot arm by hand while its motors supply just enough torque to support its weight and keep the robot upright. This simple interaction, a useful starting point for robots that learn by demonstration, manipulate objects, or work alongside people, depends on a gravity-compensation loop. Building such a loop is only possible by continuously calculating the inverse dynamics of the system based on precise models of all the links of the robot arm, their dimensions, masses, and moment