Python platform for remote control of robotics systems and cameras.

Provides support for accessible robotics platforms, wireless camera streaming from onboard cameras mounted on robots, and external odometry taken from iOS devices.

It is assumed that an Arduino UNO board is being used, with continuous servos connected to pins 9 and 10. Note that this robot is fully tethered, so a USB extension may be required.

The iCreate robot is a Raspberry Pi running Raspbian Jessie with the iCreate connected to the Raspberry Pi through a USB serial interface. The Raspberry Pi should also have access to WiFi through a WiFi adapter.

Upload the file PythonRobot.ino to the Arduino. Check the serial port in the Arduino IDE. With the Arduino on and running, run

python arduino.py

in CLI. This will open a blank window. Click on the window. The WASD keys will control the robot.

If the script cannot connect, then change the serial port within the script to the serial port found within the Arduino IDE.

Start the script CreateServer.py on the Raspberry Pi. If an error is thrown, check the IP address with ifconfig and edit the script accordingly.

On the client computer, run

python CreateClient.py

in CLI. This will open a blank window. Click on the window. The WASD keys will control the robot.

The script

python cam.py

will read a video stream from a USB webcam.

For a camera module connected to the Raspberry Pi, refer to this tutorial to install mjpg-streamer on the Raspberry Pi. With the stream open, run

python PiCam.py

on the client computer, and the video stream should show up.

Start the script iPadServer.py on the iOS device through the app Pythonista, which can be found on the Apple App Store. On the client computer, run

python iPadClient.py

and a stream of accelerometer and gyroscope data should start.

If the connection is refused, check the IP address of the iPad in the WiFi settings. Ensure that the client computer is connected to the same WiFi and retry the connection.

Also included in this repository is an example script that uses OpenCV and an overhead camera to track a robot. The user can set a location with the mouse. To use this script, first ensure that Python 3.5+ and OpenCV 3.0.0 bindings are both installed. Next, secure an overhead camera. Finally, start either ArduinoTracker.py or CreateTracker.py and click and drag to select the robot. Click again to set the destination.

Harvey Mudd College Computer Science REU 2016

Ciante Jones: cjjones@hmc.edu

Chi-Yen Lee: johlee@hmc.edu

Andy Zhang: axzhang@hmc.edu

Professor Zach Dodds: zdodds@hmc.edu