機器人體驗營筆記（一）概要

全文内容來源于國外權威資料彙總整理，具體資訊請查閱文末參考文獻。

For specific information, please refer to the reference at the end of the article.

課程描述 Course Description

本課程探讨移動機器人中智能行為的實作，重點關注Anki的Cozmo機器人。預備基礎是中級程式設計技巧和Python基礎。如果具備機器人技術或人工智能的經驗更佳，但這不是必需的。

This course explores the implementation of intelligent behavior in mobile robots, focusing on the Cozmo robot by Anki.

The prerequisite for the course is intermediate-level programming skills and facility with Python. Prior experience in robotics or artificial intelligence is helpful but not required.

學習目标 Learning Objectives

學完本課程後，您将能夠：

使用Python在Cozmo機器人上程式設計智能行為。
掌握計算機視覺技術使用OpenCV識别标記的對象。
設計機器人環境，實作基于視覺地标的定位和導航。
使用語音識别實作機器人的語音控制。
評估對比三個Cozmo程式設計架構的優劣：Cozmo Python SDK，Code Lab和ROS。

After taking this course, you will be able to:

Program intelligent behaviors on the Cozmo robot using Python.
Employ computer vision techniques using OpenCV to recognize markers and objects.
Design robot environments that facilitate visual landmark-based localization and navigation.
Use speech recognition to provide voice control of a robot.
Assess the strengths and limitations of three Cozmo programming frameworks: the Cozmo Python SDK, Code Lab, and ROS.

學習資源 Learning Resources

課程沒有教科書。There is no textbook for the course.

更多資源查閱：Cozmo、ROS1、ROS2、OpenCV和OpenAI的官網維基。

More resources: Cozmo, ROS1, ROS2, OpenCV and OpenAI's official website wiki.

環境配置 Laboratory configuration

手機Phone：安裝Cozmo_3.4.0最新版應用。iOS直接在應用商店查找安裝，Android在網上下載下傳xapk包安裝。Install the latest version of Cozmo_3.4.0. iOS finds the installation directly in the app store, and Android downloads the xapk package online.

電腦PC：下載下傳Cozmo_SDK、Cozmo_tools、ROS1驅動和ROS2驅動等，推薦使用Ubuntu 18.04 LTS桌面系統、VSCode等。Download Cozmo_SDK, Cozmo_tools, ROS1 drivers and ROS2 drivers, etc. It is recommended to use Ubuntu 18.04 LTS desktop system, VSCode and so on.

機器人Cozmo：連接配接無線，并依據提示更新系統。Connect WiFi and follow the prompts to update the robot system.

配置和使用虛拟環境如下：install and set up virtualenv:

sudo apt-get install python-pip pip install virtualenv virtualenv -p python3.6 ~/cozmo-env

激活虛拟環境：activate the virtualenv any time you use cozmo:

source ~/cozmo-env/bin/activate

但有時候會運作出錯，推薦用如下指令完成在Ubuntu 18.04或16.04的配置。

To install the SDK, type the following into the Terminal window:

pip3 install --user 'cozmo[camera]'

更新：

To upgrade the SDK from a previous install, enter this command:

pip3 install --user --upgrade cozmo

機器人學習對場地的要求，一般而言比較高，這是由于機器人本身體積較大。有些競賽項目甚至需要體育館大小空間。Learning robots have a relatively high requirement for space on the site, because robots generally have large dimensions. Some competitions even require as much space as a gym.

但是學習Cozmo，通常隻需一張A4紙大小的空間。But learning Cozmo usually requires only one A4 paper size space.

Cozmo機器人價格低于100美元/660元人民币，全部材料配齊也遠低于1000元人民币。Cozmo robots cost less than $100/660 yuan, and the price of all materials is below 1,000 yuan.

它是目前學習機器人學、機器人作業系統和人工智能最低成本的方案之一，但是功能非常強大！It is one of the lowest cost devices for learning robotics, ROS and AI, but it is very powerful!

硬體架構、軟體設計和課程生态都是非常豐富多彩的。The hardware architecture, software design, and curriculum ecology are all outstanding.

它由300多個部件組成。It consists of more than 300 parts.

它内部有四個電機，分别驅動左輪、右輪、頭部和起重臂。It has four motors inside that drive the left wheel, the right wheel, the head and the lift.

三塊小電路闆。Three small circuit boards.

一個攝像頭傳感器，可以輸出320x240的灰階或160x240的彩色圖像。A camera sensor that can output 320x240 grayscale or 160x240 color images.

面部為LED點陣顯示并配備一個揚聲器。Its head has an LED dot matrix display and is equipped with a speaker.

皮克斯設計的動畫表情包。Pixar Designer “Animations”

Vector機器人資料，請自主查找。更多功能，更多内容。

持續更新完善，時間标簽：2019-07-25。

參考文獻references：

Anki文檔：http://cozmosdk.anki.com/docs/
cozmopedia：https://github.com/touretzkyds/cozmopedia/wiki
認知機器人學：https://blog.csdn.net/ZhangRelay/article/details/86736743