NVIDIA Isaac ROS

NVIDIA Isaac ROS

NVIDIA Isaac™ ROS (Robot Operating System) is a collection of NVIDIA® CUDA®-accelerated computing packages and AI models designed to streamline and expedite the development of advanced AI robotics applications.
NVIDIA Isaac™ ROS(机器人操作系统)是NVIDIA® CUDA®加速计算包和AI模型的集合,旨在简化和加快高级AI机器人应用程序的开发。

Download NowDocumentation
立即下载文档
Forums  论坛


How NVIDIA Isaac ROS Works
NVIDIA Isaac ROS的工作原理

Isaac ROS gives you a powerful toolkit for building robotic applications. It offers ready-to-use packages for common tasks like navigation and perception, uses NVIDIA frameworks for optimal performance, and can be deployed on both workstations and embedded systems like NVIDIA Jetson™.
Isaac ROS为您提供了构建机器人应用程序的强大工具包。它为导航和感知等常见任务提供了即用型软件包,使用NVIDIA框架实现最佳性能,并且可以部署在工作站和NVIDIA Jetson™等嵌入式系统上。

A diagram showing  how NVIDIA Isaac ROS works

Quick Start Guide  快速入门指南

Learn what you need to get started and how to set up using the Isaac ROS suite to tap into the power of NVIDIA acceleration on NVIDIA Jetson.
了解入门所需的内容以及如何使用Isaac ROS套件进行设置,以充分利用NVIDIA Jetson上的NVIDIA加速功能。

Introductory Talk  介绍性演讲

Isaac ROS offers modular packages for robotic perception and easy integration into existing ROS 2-based applications. This talk covers Isaac ROS GEMs and how to use multiple GEMs in your robotics pipeline.
Isaac ROS提供了用于机器人感知的模块化软件包,并可轻松集成到现有的基于ROS 2的应用程序中。本讲座涵盖了Isaac ROS GEM以及如何在机器人管道中使用多个GEM。

Introductory Webinars  在线研讨会

Check out a series of Isaac ROS webinars covering various topics, from running your own ROS 2 benchmarks to harnessing the power of NVIDIA NITROS.
查看一系列Isaac ROS网络研讨会,涵盖各种主题,从运行您自己的ROS 2基准测试到利用NVIDIA NITROS的强大功能。

What’s New—Isaac ROS 3.2  新功能-Isaac ROS 3.2

Check out the latest Isaac ROS update to boost your robot’s capabilities with advanced AI-based perception and manipulation
查看最新的Isaac ROS更新,通过先进的基于AI的感知和操纵来提升机器人的能力


Key Features  关键特征

Open Robotics ROS logo

ROS is a trademark of Open Robotics
ROS是Open Robotics的商标

Open Ecosystem  开放生态

Built on ROS  基于ROS构建

NVIDIA Isaac ROS is built on the open-source ROS 2™ software framework. This means the millions of developers in the ROS community can easily take advantage of NVIDIA-accelerated libraries and AI models to fast track their AI robot development and deployment workflows.
NVIDIA Isaac ROS基于开源ROS 2™软件框架构建。这意味着ROS社区中的数百万开发人员可以轻松利用NVIDIA加速库和AI模型来快速跟踪他们的AI机器人开发和部署工作流程。

NVIDIA Isaac Transport for ROS (NITROS) processing pipelines

Hardware Acceleration  硬件加速

NVIDIA Isaac Transport for ROS

The NVIDIA implementation of type adaption and negotiation is called NITROS, which are ROS processing pipelines made up of Isaac ROS hardware-accelerated modules (a.k.a. GEMs). NITROS lets ROS 2 applications take full advantage of GPU hardware acceleration, potentially achieving higher performance and more efficient use of computing resources across the entire ROS 2 graph.
NVIDIA实现的类型适配和协商被称为NITROS,这是由Isaac ROS硬件加速模块(也称为Ros)组成的ROS处理管道。GEM)。NITROS让ROS 2应用程序充分利用GPU硬件加速,从而在整个ROS 2图形中实现更高的性能和更有效的计算资源使用。

Download NVIDIA Isaac Transport  下载NVIDIA Isaac Transportfor ROS (Github)  关于ROS(Github)
High-Throughput Perception
高吞吐量感知

Isaac ROS delivers a rich collection of individual ROS packages (GEMs) and complete pipelines (NITROS) optimized for NVIDIA GPUs and NVIDIA Jetson™ platforms. This helps you achieve more with reduced development times.
Isaac ROS提供了丰富的单个ROS包(GEM)和完整的管道(NITROS),针对NVIDIA GPU和NVIDIA Jetson™平台进行了优化。这有助于您在缩短开发时间的同时实现更多目标。

Modular, Flexible Packages
模块化、灵活的封装

Plug and play with a selection of packages—for computer vision, image processing, robust object detection, collision detection, and trajectory optimization—and easily go to production.
一系列软件包即插即用-用于计算机视觉、图像处理、稳健的物体检测、碰撞检测和轨迹优化-并轻松投入生产。

The Power of NVIDIA AI
NVIDIA AI的力量

Isaac ROS is compatible with all ROS 2 nodes, making it easier to integrate into existing applications. Develop robotic applications using NVIDIA AI and pretrained models from robotics-specific datasets for faster development.
Isaac ROS与所有ROS 2节点兼容,使其更容易集成到现有应用程序中。使用NVIDIA AI和来自机器人特定数据集的预训练模型开发机器人应用程序,以加快开发速度。


Getting Started on NVIDIA Isaac ROS
NVIDIA Isaac ROS入门

A 101 decorative image representing system setup

System Setup  系统设置

Tap into NVIDIA-accelerated libraries and AI models to speed up your AI robot workflows. Check your system requirements and set up your system.
利用NVIDIA加速库和AI模型来加快您的AI机器人工作流程。检查系统要求并设置系统。

 

Get Started  开始

A decorative image representing plug-and-pay ROS packages

Plug-and-Play ROS Packages
即插即用ROS包

Read through the Isaac ROS concepts and easily move to production with a selection of advanced packages.
阅读Isaac ROS概念,并通过一系列高级软件包轻松进入生产环境。

 

Isaac ROS Packages  Isaac ROS包

Deploy at the edge with NVIDIA Jetson

Deploy at the Edge  在边缘部署

NVIDIA Jetson provides hardware acceleration, optimized AI software, a robust ecosystem, and energy efficiency, making it an ideal platform to deploy your Isaac ROS applications. Nova Carter and the Nova Orin™ developer platforms also help you accelerate AMR development.
NVIDIA Jetson提供硬件加速、优化的AI软件、强大的生态系统和高能效,使其成为部署Isaac ROS应用程序的理想平台。Nova Carter和Nova Orin™开发平台还可帮助您加快AMR开发。

 

Buy Nova Orin Developer Kit  购买Nova Orin Developer Kit(Leopard Imaging)  (Leopard Imaging) Buy Nova Orin Developer Kit  购买Nova Orin Developer Kit(Segway Robotics)  (赛格威机器人公司)
1.00
NVIDIA Isaac Manipulator  NVIDIA Isaac机械手

Isaac Manipulator, built on Isaac ROS, enables the development of AI-powered robotic arms that can seamlessly perceive, understand, and interact with their environments.
Isaac Manipulator基于Isaac ROS构建,能够开发出能够无缝感知、理解并与环境交互的AI驱动的机器人手臂。

 

Learn More About Isaac Manipulator  了解更多关于Isaac Manipulator
1.00
NVIDIA Isaac Perceptor

Isaac Perceptor, built on Isaac ROS, enables rapid development of advanced autonomous mobile robots (AMRs) that can perceive, localize, and operate in unstructured environments like warehouses or factories.
Isaac Perceptor基于Isaac ROS构建,能够快速开发先进的自主移动的机器人(AMR),这些机器人可以在仓库或工厂等非结构化环境中进行感知、定位和操作。

 

Learn More About Isaac Perceptor  了解有关Isaac Perceptor的更多信息

High-Performance Perception With NITROS Pipelines
使用NITROS Pipelines实现高性能感知

ROS 2 graphs using NITROS-based, NVIDIA-accelerated Isaac ROS packages can significantly increase performance.
使用基于NITROS、NVIDIA加速的Isaac ROS软件包的ROS 2图形可以显著提高性能。

You can find a complete performance summary here.
您可以在这里找到完整的性能摘要。

Node  节点
Input Size  输入大小
AGX Orin  AGX奥林
Orin NX
x86_64 w/ RTX 4090  x86_64,带RTX 4090
AprilTag Node  AprilTag节点
720p
249 fps  249帧/秒
4.5ms @ 30Hz  4.5ms@30Hz
116 fps
9.3ms @ 30Hz  9.3ms@30Hz
596 fps
0.97ms @ 30Hz  0.97ms@30Hz
Freespace Segmentation Node
自由空间分段节点
576p
2120 fps
1.7ms @ 30Hz  1.7ms@30Hz
2490 fps
1.6ms @ 30Hz  1.6ms@30Hz
3500 fps
0.52ms @ 30Hz  0.52ms@30Hz
Depth Segmentation Node  深度分割节点
576p
45.8 fps
79 ms @ 30Hz  79 ms@30 Hz
28.2 fps
99ms @ 30Hz  99ms@30Hz
105 fps
25ms @ 30Hz  25ms@30Hz
TensorRT Node  TensorRT节点
PeopleSemSegNet
544p
460 fps
4.1ms @ 30Hz  4.1ms@30Hz
348 fps
6.1ms @ 30Hz  6.1ms@30Hz
-
Triton Node  Triton节点
PeopleSemSegNet
544p
304 fps
4.8ms @ 30Hz  4.8ms@30Hz
206 fps
6.5ms @ 30Hz  6.5ms@30Hz
-
DNN Stereo Disparity Node
DNN立体视差节点
Full
576p  576 P
103 fps  103 fps的
12ms @ 30Hz  12 ms @ 30 Hz的
42.1 fps
26ms @ 30Hz  26 ms @ 30 Hz(26 ms @ 30 Hz)
350 fps  350 fps的
2.3ms @ 30Hz  2.3ms @ 30Hz(2.3ms @ 30Hz)
H.264 Decoder Node  H.264 Decoder Node协议
1080p  1080 P
197 fps  197 fps的
8.2ms @ 30Hz  8.2ms @ 30Hz的
-
596 fps  596 fps的
4.2ms @ 30Hz  4.2ms @ 30Hz(30 Hz)
H.264 Encoder Node  H.264 Encoder Node编码
I-frame Support  I—Frame支持
1080p  1080 P
402 fps  402 fps的
13ms @ 30Hz  13 ms @ 30 Hz的
-
409 fps  409 fps的
3.4ms @ 30Hz  3.4ms @ 30Hz的
H.264 Encoder Node  H.264 Encoder Node编码
P-frame Support  P型支架
1080p
473 fps
11ms @ 30Hz  11ms@30Hz
-
596 fps
2.1ms @ 30Hz  2.1ms@30Hz
Nvblox Node  Nvblox节点
-
4.87 fps
35.9ms
4.95 fps
-1.43ms
4.95 fps
195ms  195毫秒

Starter Kits  初学者工具包

Start developing your robotics and AI application with Isaac ROS with these forumsrelease notes, and comprehensive documentation.
开始使用Isaac ROS开发您的机器人和人工智能应用程序,包括这些论坛发行说明全面的文档

Localization and Mapping  定位与地图构建

Isaac ROS Visual SLAM provides a high-performance, best-in-class ROS 2 package for VSLAM (visual simultaneous localization and mapping).
Isaac ROS Visual SLAM为VSLAM(视觉同步定位和地图绘制)提供了一个高性能、同类最佳的ROS 2包。

 

3D Scene Reconstruction  3D场景重建

Isaac ROS nvBlox uses RGB-D data to create a dense 3D map, including unforeseen obstacles, to generate a temporal costmap for navigation.
Isaac ROS nvBlox使用RGB-D数据创建密集的3D地图,包括不可预见的障碍物,以生成用于导航的时间成本图。

 

Pose Estimation and Tracking
姿态估计和跟踪

NVIDIA’s FoundationPose is a state-of-the-art foundation model for 6D pose estimation and tracking of novel objects.
NVIDIA的FoundationPose是最先进的基础模型,用于6D姿态估计和跟踪新对象。

 

Motion Planning  运动规划

Isaac ROS cuMotion is an NVIDIA CUDA-accelerated library for solving robot motion planning problems at scale by running multiple trajectory optimizations simultaneously to return the best solution.
Isaac ROS cuMotion是一个NVIDIA CUDA加速库,可通过同时运行多个轨迹优化来返回最佳解决方案,从而大规模解决机器人运动规划问题。

 

Testing and Validation in Simulation
仿真中的测试与验证

Virtually train, test, and validate robotics systems using NVIDIA Isaac Sim and NVIDIA Isaac Lab.
使用NVIDIA Isaac Sim和NVIDIA Isaac Lab虚拟训练、测试和验证机器人系统。

 


Isaac ROS Learning Library
Isaac ROS学习图书馆

Featured  特色
Documentation  文件

NVIDIA Isaac ROS Release Notes
NVIDIA Isaac ROS发行说明

Get the comprehensive updates on the latest features, improvements, and bug fixes for Isaac ROS.
获取有关Isaac ROS最新功能、改进和错误修复的全面更新。

Featured  特色
Tech Blog  科技博客

Advance Robot Learning, Perception, and Manipulation with the Latest NVIDIA Isaac Release
借助最新的NVIDIA Isaac版本,推进机器人学习、感知和操纵

Explore the latest Isaac release to enhance your robot learning, perception, manipulation, and environment mapping.
探索最新的Isaac版本,以增强您的机器人学习、感知、操纵和环境映射。

Featured  特色
Webinar  网络研讨

Isaac ROS Webinar Series  Isaac ROS网络研讨会系列

Check out a series of Isaac ROS webinars covering various topics, from running your own ROS two benchmarks to harnessing the power of NITROS.
查看一系列Isaac ROS网络研讨会,涵盖各种主题,从运行自己的ROS两个基准到利用NITROS的力量。


More Resources  更多资源

NVIDIA Developer Forums

Explore the Community  探索社区

NVIDIA Training and Certification

Get Training and Certification
获得培训和认证

NVIDIA Inception Program for Startups

Join the Program for Startups
加入创业计划

### NVIDIA Isaac Sim 机器人仿真开发平台使用教程 #### 下载与安装 NVIDIA Isaac Sim 是一个基于 Omniverse 的物理精确模拟环境,专为加速机器人的设计、测试和部署而构建[^2]。为了开始使用此工具,需先下载并安装软件。 访问[NVIDIA官网](https://developer.nvidia.com/isaac-sim),按照页面提示注册账号后可获取最新版本的 Isaac Sim 安装包。对于初次使用者来说,建议遵循官方文档中的指导完成设置流程[^3]。 #### 配置环境 成功安装之后,下一步就是配置工作空间以便能够顺利运行项目。这通常涉及到创建一个新的虚拟环境来管理依赖项以及确保所有必要的库都已正确加载。具体操作可以参照入门指南中关于如何准备开发环境的部分说明。 #### 导入自定义模型 当一切就绪时,就可以着手于将自己的 STL 文件转换成适合在 isaac sim 中使用的格式了。这一过程可以通过第三方建模软件实现,比如 Blender,在其中导出 USD (Universal Scene Description) 格式的文件再导入到 isaac sim 当中[^1]。 #### 编写控制逻辑 为了让车辆或其他类型的机器人动起来,则需要编写相应的程序代码去驱动它们的动作。这部分内容涉及到了ROS2框架的应用编程接口(APIs)。下面给出一段简单的 Python 脚本作为例子: ```python import rclpy from geometry_msgs.msg import Twist def main(args=None): rclpy.init(args=args) node = rclpy.create_node('move_robot') publisher = node.create_publisher(Twist, '/cmd_vel', 10) msg = Twist() while True: # 设置线速度和角速度 msg.linear.x = 0.5 msg.angular.z = 0 publisher.publish(msg) if __name__ == '__main__': main() ``` 这段脚本实现了让机器人向前移动的功能;实际应用中可能还需要加入更多复杂的运动规划算法以适应不同的任务需求。
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值