Raspberry Pi Voice Controlled Robot

Robot that is controlled through android app and voice processing

Comment

Inspiration

We were inspired with that hardware provided and the potential to make a robot that could interact with a mobile device and respond to voice control.

What it does

Our robot can read and respond to voice in put from a mobile app through a tcp connection as well as take commands from a keyboard to move around a room. The robot is built with ultrasonic distance sensors which light up green when there is an adequate distance between itself and an object, yellow when it is approaching an object, and red when it should stop. The robot also has function called explore which it will move around a room without guidance and avoid objects using it's sensors.

How we built it

We built our robot using a raspberry pi loaded with raspbian and wrote our code in python. We used the pins on the pi to interact with the motors, sensors, and leds. The app was written in android studio using java.

Challenges we ran into

We ran into challenges when we first started our project. We initial had wanted to accomplish this project with drones but quickly realized that hacking into radio signals would be extremely difficult to accomplish our project so we pivoted. While we were working on the robot we ran into problems with the tcp connection in which the pi connected to the phone app.

Accomplishments that we're proud of

We are proud of developing a functioning robot in under 24 hours and realizing where we should make a pivot. In addition we are proud of figuring out how to make a tcp connection between our mobile app and the raspberry pi.

What we learned

We learned how to make connections through tcp. In addition we gained experience with voice recognition libraries in android-studio. Finally we gained a lot of experience with hardware and sensors.

What's next for Raspberry Pi Voice Controlled Robot

We would like to add a function that would allow the robot to travel back to the starting location on the same path it left. We would do this by recording the track it took using a stack and then popping off those directions and reversing them.

Built With

Share this project:

Updates