Posted on

Weekly Update 9/17 – 9/23

What we did this week as a group:

  • Placed orders for Raspberry Pi, ultrasonic sensors, camera, and wires
  • Set up project website
  • Started building ParkSmart Android App
  • Brainstormed on how to integrate the data from ultrasonic sensors and camera
    • Sketch
    • As shown in the sketch, we hope to draw an outline of obstacles around the vehicles based on the data from ultrasonic sensors, and then show the outline on the App so to give the driver a better visual understanding of the surroundings
    • We will then pass the data to our algorithm to provide instruction based on the spatial outline

Hubert Yu

  • Setting up android development environment
  • Upload android demo projects and introduce basic android development
  • Explore android & rpi video streaming options
    • H.264 https://raspberrypi.stackexchange.com/questions/7446/how-can-i-stream-h-264-video-from-the-raspberry-pi-camera-module-via-a-web-serve
    • MJPEG https://blog.miguelgrinberg.com/post/stream-video-from-the-raspberry-pi-camera-to-web-browsers-even-on-ios-and-android
    • Mp4 https://code.tutsplus.com/tutorials/streaming-video-in-android-apps–cms-19888
    • High level idea:
      • Camera stream -> MJPEG -> android app
      • Camera stream -> H.264 -> android app
      • Camera stream -> MP4 (compressed) -> android app
    • https://github.com/mbebenita/Broadway/issues/59
      • H264 streaming pear and open source app (looks very useful)

Zilei Gu

  • Created initial repository for android app
  • Setup android development environment & exploring android development panels
  • Android UI crash course
  • More readings on parallel parking (double touch: https://crucialminutiae.wordpress.com/2007/08/13/the-perfect-parallel-park/)
  • Readings on RPi wireless access point (https://github.com/SurferTim/documentation/blob/6bc583965254fa292a470990c40b145f553f6b34/configuration/wireless/access-point.md and (https://learn.adafruit.com/setting-up-a-raspberry-pi-as-a-wifi-access-point/overview)

Yuqing Ma

Posted on

What is Pakr?

Welcome to Carnegie Mellon University: ECE Capstone Projects – ParkSmart.

Pakr is an ECE capstone project done by a group of three: Zilei Gu, Hubert Yu, and Yuqing Ma.

Many new models of automobiles offer amazing parking assistance systems, distance detectors, back-up cameras, etc. While these features could greatly improve your driving/parking experience, many vehicles of older models don’t have the console system that’s compatible with this upgrade. Installing a back-up camera and radar system would often mean replacing the entire console and incur a much higher cost. In addition, even though there exist parking assistance systems on the market, there is a lack of clear instructions on how to maneuver the vehicles based on the spatial feedback. That’s why we’re here to build an integrated parking assistance, Pakr, that provides useful and executable information to drivers through an Android app.

Our system will be able to detect distance through ultrasonic sensors and provide parking spot image through an Android App to the drivers. The distance detection system alerts the drivers when the vehicle gets too close to surrounding objects. The image on the Android App provides the driver with visual information on the relative location of the vehicle to the surrounding objects. In addition, Pakr provides real-time audio instructions on how to maneuverer the steering wheel. 

Want to know more about Pakr? Subscribe to our blog and read our weekly updates!