ROS 笔记

ros的编程范式

- ros认为，linux平台下，机器人的软件由一个个小程序组成，这些小程序称为node，每个小程序负责一部分功能

- ros实现的框架就是，小程序的并发执行+相互通信，程序（进程）的并发由linux系统调度，相互通信是ros重点实现的message+service模型

- ros中的node可以都在同一个机器上，也可以运行在不同机器上并通过以太网等底层传输方式。ros的通信实现基于socket。

- ros的核心除了这个通信和并发机制外，是其生态社区，大量的工具和包可快速搭建机器人，深入学习。

Where Next?

Description: This tutorial discusses options for getting to know more about using ROS on real or simulated robots.

Tutorial Level: BEGINNER

目录

1. Launching a Simulator
2. Exploring RViz
3. Understanding TF
4. Going Deeper

At this point in the beginner's tutorials you should have an understanding of the core concepts of ROS.

Given a robot that runs ROS, you could use this understanding to list topics published and subscribed by the robot, to identify the messages consumed by these topics and then write your own nodes that process sensor data and act in the world.

The real attraction of ROS is not the publish/subscribe middle-ware itself but that ROS provides a standard mechanism for developers around the world to share their code. The best "feature" of ROS is its enormous community.

The number of packages available can be overwhelming. This tutorial attempts to give you an idea of what to explore next.

Launching a Simulator

Even if you have a real robot, it is good to get started using a simulator so that if something goes wrong you don't injure yourself or damage an expensive robot.

You can get started with the PR2 Simulator or the Turtlebot Simulator. Alternately, you might search for your robot and check whether it has a simulator of its own.

At this point, you might try to control the simulated robot using a 'teleop' package (e.g., turtlebot_teleop) or use your understanding of ROS to find a topic and write code that sends an appropriate message to drive your robot.

Exploring RViz

RViz is a powerful visualization tool that allows you to view the robot's sensors and internal state. The user guide will help you get started.

Understanding TF

The TF package transforms between different coordinate frames used by your robot and keeps track of these transforms over time. A good understanding of TF is essential when working with any real robot. It is worthwhile to work through the tutorials.

If you're building your own robot, you might at this point consider constructing a URDF model for your robot. If you're using a "standard" robot then one has probably already being built for you. Nevertheless, it may be worthwhile to briefly familiarize yourself with the URDF package.

Going Deeper

At this point, you're probably ready to start getting your robot to perform more sophisticated tasks. The following pages may help you:

1. actionlib - The actionlib package provides a standardized interface for interfacing with preemptible tasks. This is widely used by "higher-level" packages in ROS.

2. navigation - 2D navigation: map-building and path planning.

3. MoveIt - To control the arms of your robot

　　　　　　　　

------------------------------------------------------------------

Intermediate Level

More client API tutorials can be found in the relevant package (roscpp, rospy, roslisp)

1. Creating a ROS package by hand.

   This tutorial explains how to manually create a ROS package.

2. Managing System dependencies

   This explains how to use rosdep to install system dependencies.

3. Roslaunch tips for large projects

   This tutorial describes some tips for writing roslaunch files for large projects. The focus is on how to structure launch files so they may be reused as much as possible in different situations. We'll use the 2dnav_pr2 package as a case study.

4. Running ROS across multiple machines

   This tutorial explains how to start a ROS system using two machines. It explains the use of ROS_MASTER_URI to configure multiple machines to use a single master.

5. Defining Custom Messages

   This tutorial will show you how to define your own custom message data types using the ROS Message Description Language.

6. Using a C++ class in Python

   This tutorial illustrates a way to use a C++ class with ROS messages in Python.

7. Packaging your ROS project as a snap

   This tutorial covers how to package and deploy your ROS project as a snap.

8. How to Write a Tutorial

   This tutorial covers useful template and macros for writing tutorials, along with example tutorials that are available for guidance on ros.org