I am figuring about the AI ROBOT platform with RPi 3 out.

And control the Alexa Go Go wirelessly.

i use arduino nano controlling the wheels with servo
and dc-motor controlling by the remoter(part_4).

maybe there are someone wondering about the IR remoter

that is not smart enough. so that i make with an

Alexa to drive it from upon the cloud.

means, to control the moving of the robot using voice command .

Step 1: Material

RPi 3 * 1

MIC * 1

USB sound card * 1

dc-motor gear * 1

micro servo motor * 1

screw * 9

BBQ stick * 1

RPi 3 * 1

AAA battery * 4

AAA battery holder * 2

3.7 v Li battery(18650) * 1

3.7 v battery (18650)holder * 1

IR rcvr (arduino module) * 1

IR remoter(arduino module ) * 1

RY1 : Solid State Relay * 2

RY 2, 3 : relay 5 v * 2

TR 1, 2 : 2N 3904 * 2

D 1, 2 : IN 4148 * 2

C 1 : 100 uF * 1

C 2 : 0.01 uF * 1 (for bypass noise)

R1 : 10 K OHM * 1

R 4,5 : 1 K OHM * 2

PCB * 1

mini breadboard * 1(prototype use)

Step 2: Make Alexa's Legs

Step 3: Install RPi 3 to Set Up Alexa

Install RPi 3 to Desktop

1. RPI 3 * 1 (as Alexa's body)

2. HDTV * 1(FOR setup Alexa & as her mouth)

3. USB MIC * 1 (as Alexa's ear)

4. wireless keyboard & mouse * 1(FOR setup Alexa)

reference link :

https://www.instructables.com/id/How-to-Make-Alexa...

Step 4: Set Up the Ngrok to Interchange to the Cloud

5. you need to build a security tunnel from the cloud to your RPi 3.

Visit https://ngrok.com/download and get the latest Linux ARM release as a zip

and unzip inside the home directory:

unzip /home/pi/ngrok-stable-linux-arm.zipNext,

run : sudo ./ngrok http 5000 (to run in memory)

Your screen should look like the image above. Note the 'Forwarding' URL that starts with https, it will be used later.

Note: Unfortunately the ngrok URL changes every time the service is started so it is not a permanent solution if you are trying to run this full time. I'd recommend a service like Yaler or Page Kite if you need a more permanent URL to use with your new Skill.

6. Python Script Open a new terminal session and create a new python file named gpio_control.py:

edit : nano gpio_control.py

example file:gpio_control.py

run : sudo python gpio_control.py (to run in memory)