Close
0%
0%

IoT with Swarm Biomimicry

I plan on using IoT framework to create swarm robots with biomimetic capabilities.

kirthikirthi
Following
Liked Like project

Just one more thing

To make the experience fit your profile, pick a username and tell us what interests you.

Pick an awesome username
hackaday.io/
Your profile's URL: hackaday.io/username. Max 25 alphanumeric characters.
Pick a few interests
Projects that share your interests
People that share your interests

We found and based on your interests.

Choose more interests.

OK, I'm done! Skip
Similar projects worth following
85 views
0 comments
2 followers
View Gallery
The first step of my project will be to create a basic robot with basic motion capability. Next, a duplicate robot would be created with a separate microprocessor (decentralized, localized systems). Next, the two robots should be programmed to autonomously navigate through an unknown (engineered) environment and separately communicate to a main hub. The robots should be programmed through a convolutional neural net to identify key features of the environment. The data and/or progress of the robot should be able to be stored in a central location, and displayed to a website (IoT).

Object Detection

My initial approach to object detection was using TensorFlow to recognize objects, and while I was able to do this via tutorials online [ref: https://towardsdatascience.com/real-time-object-tracking-with-tensorflow-raspberry-pi-and-pan-tilt-hat-2aeaef47e134, this was a pretty good resource]. However, I quickly realized that TensorFlow's robustness and preloaded object detection algorithms are too complex for the purpose of my project. At this point I had one of two options 1) continue with TensorFlow and use a process called transfer learning (change up the top few layers of the CNN), or 2) proceed with a more computer vision-focused approach, and start by using OpenCV. I decided to do the latter for now because that is what I have more experience in, and I spent a great deal of time downloading the necessary software. I used a tutorial online to do edge detection for images, as this is the first step to the object detection that is necessary for the project [ref: https://opencv-python-tutroals.readthedocs.io/en/latest/py_tutorials/py_imgproc/py_canny/py_canny.html]. I modified the parameters used in the Canny method to improve the edge detection precision. 


  • 1 × Raspberry Pi 3b+ And of course, all the accessories (SD Card, USB, etc.)
  • 1 × Raspberry Pi Camera
  • 1 × Pololu 3pi Robot And the accessories (AVR programmer)

  • Project Log

    kirthi • 01/23/2020 at 15:23 • 0 comments

    September 2-6, 2019

    Plan: I wanted to finalize a project idea and complete some background research.

    Procedures: I discussed ideas with various people (including my parents and other mentors). I did background research one one topic, although I later decided to do a different project. 

    Progress: I narrowed down my project and project area significantly. I plan to rewrite my project proposal and work on the pitch presentation over the weekend.

    Problems: I originally thought I would like to do one project, but I later decided that it was better for me to pick a different project. This is only a setback due to the time spent researching the previous project that could have been used to further develop the current project.

    September 9-13, 2019

    Plan: Research deeper into project idea and finalize actual project. 

    Procedures: I looked into various papers to find previous projects done in IoT with relation to swarms and biomimicry. I found previous framework that would be useful to my project. 

    Progress: I further narrowed down my project and created a timeline. I did the pre-proposal presentation and got feedback on my idea. I completed background research and made my website.

    Problems: I did not have that many problems this week. I was confused about some parts of the background research but I later got that clarified. Other then that this week I didn’t have any major problems. 

    September 16-20, 2019

    Plan: Research deeper into project idea and finalize actual project. 

    Procedures: I researched into various aspects of my project for background research. I focused my research into processing, communication, and building. 

    Progress: I completed my background research and made any necessary revisions.

    Problems: I did not have any problems this week. I had to redo some background research by adding more specific details (to solve this, I included diagrams).

    September 23-27, 2019

    Plan: Research deeper into project idea and finalize actual project. 

    Procedures: I researched into various aspects of my project for the purchase spreadsheet. I broke my spreadsheet down into two lists, one with arduino components and one with pi components. I will make a decision on how to do it next week. 

    Progress: I completed my purchase spreadsheet. 

    Problems: I did not have any problems this week. I got kind of overwhelmed by the different possibilities for parts. I think by narrowing down the project and by dividing up the spreadsheet into two columns I can make an informed decision. 

    September 30-Oct 4, 2019

    Plan: Research deeper into project idea to complete project proposal.

    Procedures: I researched into various aspects of my projects for the project proposal. 

    Progress: I completed my research proposal. 

    Problems: I had to really flesh out my project for it, but other than that there were no problems.

    Oct 7-11, 2019

    Plan: Research deeper into project idea to complete project proposal and project presentation.

    Procedures: I researched into various aspects of my projects for the project proposal and project presentation.

    Progress: I completed my research proposal and presentation.

    Problems: I had to really flesh out my project for it, but other than that there were no problems.

    Oct 14-18, 2019

    Plan: Research into the specific parts to ensure compatibility. Plan CAD.

    Procedures: I looked into each of the parts to make sure they would work together. I also considered what I would need to CAD.

    Progress: I put the necessary parts on the order sheet and I began CAD, which I plan on finishing over the weekend and on Monday.

    Problems: I realized that the WiFi chip may or may not be compatible with the Pi the way I want it to be. I plan on ordering the WiFi chip if necessary at a later date. I decided to only order one Pi and one camera for now, see how that works, and order more parts later. 

    Oct 21-25, 2019

    Plan: Finish CAD.

    Procedures: I tried to get my sketch...

    Read more »

View project log

Enjoy this project?

Share

Discussions

Similar Projects

This is going to be a long and hard project - and a really expensive one too. I'm happy to take you along for the ride!
Project Owner Contributor

(Exosuit?)/VR Controlled Bot

adam-jensenAdam Jensen

An open 3D printable optomechanics kit
Project Owner Contributor

μCube

mihails-delmansMihails Delmans

First proper attempt at building an underwater rover
Project Owner Contributor

Underwater ROV

adityaprakash23AdityaPrakash23

A raspberry pi robot with camera to make maps
Project Owner Contributor

Mapper Robot

bruno-caroBruno Caro

Does this project spark your interest?

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

Going up?

About Us Contact Hackaday.io Give Feedback Terms of Use Privacy Policy Hackaday API

© 2024 Hackaday

Yes, delete it Cancel

Report project as inappropriate

You are about to report the project "IoT with Swarm Biomimicry", please tell us the reason.

Send message

Your application has been submitted.

Remove Member

Are you sure you want to remove yourself as a member for this project?

Project owner will be notified upon removal.