Close
0%
0%

miESlate - a peer to peer learning platform

A low cost peer-to-peer educational platform to make elementary education engaging and fun. Specially for schools in under-developed areas.

Similar projects worth following
What is the problem this platform is trying to address ?
The barrier to learning in most remote/rural areas in India and around the world are :
1. Most of the times a single teacher, teaches a group of around 30 to 60 students. So paying individual attention to students is difficult.
2. Because of point #1 above, the emphasis is on repetition and rote learning rather than making learning fun and interesting.
3. Although Low cost laptops and tablets have been made available to many rural schools they are expensive and fragile so kids aren't allowed to use them all the time.

My Solution is to build a low cost peer to peer device with a display, speaker and user input.
Each device can be a master or a slave. The master node publishes questions (audio / images / text) along with their answers to the slave nodes. The nodes have computational power only to process the user input and compare it with the answer.

The multimedia and p2p interaction makes the learning fun & engaging

A Brief Video Description of my project.

  • 1 × Pinoccio Dev Board This board is being used in the prototype to provide the P2P functionality and power.
  • 1 × MP3 decoder chip To playback audio samples
  • 1 × 8 ohm speaker Audio output
  • 1 × Nokia 5110 LCD module To provide visual feedback to the user.
  • 1 × 20 x 20 Capacitive Touch Pad To provide user input, this allows user to enter custom alphabets in languages other than english.

View all 6 components

  • Recognising the free form text written on the cap-touch pad

    kunal.t208/21/2014 at 13:17 0 comments

    The idea of replacing the expensive touch screens using cheap capacitive touch pads, for free-form text input is quite interesting. However the challenge lies in identifying what the user has written on the touch pad.

    To solve this problem I plan to use an implementation of M.Turk and Pentland's classic paper titled "Eigen Faces for Recognition"

  • User Input

    kunal.t208/21/2014 at 13:08 0 comments

    I plan to use a matrix of 20 x 20 capacitive touch pads which would be scanned sequentially. To get a 20 x 20 matrix of on/off data.

    Something like the illustration shown below on a 5 x 10 matrix

    0010000100

    0010000100

    001 1 1 1 1 100

    0010000100

    0010000100

    If you observe carefully, I have written a capital "H" in the bit field above.

    The idea is to draw inspiration from the slate and provide the user with a writing area where he/she can write anything. Numbers / English Alphabets / Hindi Alphabets / Kanji ( Chinese ) Alphabets etc. What ever the user writes is scaled and displayed on the Nokia LCD. This is the Unique Selling Point of this device. ( To replace expensive touch screens with a cheap cap-touch matrix )

    Along with the above, I also plan to have a "Submit" and "Clear" Push buttons to allow the user to "Save" the bit field, or clear the bit field respectively.

  • Power Consumption

    kunal.t208/21/2014 at 13:00 0 comments

    The pinoccio dev board consumes around 17mA when running at full blast. ( From the faq page of pinocc.io )

    The Nokia 5110 LCD module would take around 40mA with the back light LED on.

    Assuming I use an 8 Ohm ,  0.5W speaker. I would consume around 250mA

    Adding a buffer of 20mA for other components.

    Totally we have around 330mA.

    So, all the above would be easily powered using the 550mAh battery supplied with the pinocc.io dev board for atleast an hour.

    Given that the speaker would be used intermittently and the back light LED of the LCD module would be used only at night. We should still be able to get around 4 or 5 hours of usage from this device on a single charge.

  • Basic Idea

    kunal.t208/21/2014 at 06:18 0 comments

    So, the basic Idea of the build would be to get each component of the project working separately.

    I have a couple of pinocc.io dev boards and to get them to work in a P2P configuration would be trivial. 

    I have run the lcd display and the mp3 decoder + 8 Ohm speaker using pre-existing arduino libraries.

    The only part that I need to work on is the 20 x 20 touch pad. I want to check if the touch pad can be driven using the pinoccio itself. That way we can avoid the STM32F4 altogether.

View all 4 project logs

Enjoy this project?

Share

Discussions

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates