Hi all! I'm a data scientist (specialized in retail analytics) and I quite enjoy my job. However, recently, I've been having a lot of free time, and I'm looking for a fun (but challenging) long-term side project which is not 100% in my domain. I do not care about making money, and I would prefer it to have a "physical" dimension to them.<p>Some examples I'm thinking of: Building a face tracking drone, Using OpenBCI sensors, Adding functionalities to XGO Mini "robot dog".<p>I'd like to hear more ideas from you, especially if someone has some practical problems they think are worth solving. If anyone is already working on something like this, I'd love to join you.
I'd like to -- over a long period of time, and potentially with a few false starts -- gradually collect, analyze and turn public natural language recipes from the web into a more structured format.<p>In parallel, I'd like to identify the ingredients in those recipes, and build a knowledge graph of nutritional, price, availability and other metadata about them. That may be coarse-grained to begin with, but eventually it'd ideally support fine-grained (regionalized) supply chain information where available.<p>Finally I'd like to present and provide all of that information to consumers through a simple, understandable interface that lends itself just as well to collaborative, relaxed cooking as it does to economic, time-focused cooking, based on what the user needs.<p>I have few of the data analysis skills, and even fewer of the organizational skills, required to make this happen - and sometimes I think I need someone to provide a bit of direction and task-management for the project.<p>If that sounds at all appealing - and even if all you can offer in return is a bit of feedback or sharing the link with a friend or two: the project is currently available at <a href="https://www.reciperadar.com" rel="nofollow">https://www.reciperadar.com</a>
Some ideas:<p>Drone that can clean your gutters while avoiding obstacles like trees and chimneys. This might be very hard, but would be a challenge.<p>A series of smart heating/AC vents that open/close based on whether people are in the room. Make them solar powered and "sleep" to save power. I think this might be possible without a microcontroller, but maybe easiest with a very energy efficient one.<p>Window that opens when the outside temperature is closer to your target temp than the inside temp. E.g. if your home is set to cool but it's cooler outside, open the window. Need a rain sensor.<p>A very large Roomba. Traditional roombas are small, and not super effective as a result. Tolerating a bigger one might be work it. Get one of those battery powered vacuums that work off of power tool battery packs, perhaps.
Turn a multi-legged robot into a vacuum cleaner. Roomba style vacuum cleaners can not deal with obstructions like cables, cords, a dropped pen, toys on the floor, a sheet of paper on the floor, etc. A Spot-like robot could recognize and walk over all these and be able to get into odd nooks with a direct-able vacuum. The main difficulties would seem to be 2D path planning based on stereovision and 3D guidance of the vacuum nozzle, probably also based on stereovision. There are quite a few open source quadraped robots out there now, some you can just buy. Seems like a fairly well bounded problem and could make the world a cleaner place!
My home faces a busy street. I will give you camera feed. Do you want to analyze it for say, how many cars/trucks pass by it, what is written on the (large) trucks, how often? You can analyze how often a person(s) comes for walking etc. who is the most regular person etc. If interested, reply here with your contact details
My project does not have "physical" dimension, but it is a high impact social project. In case you want to join then the details of the project idea is documented here: <a href="https://loan-free-ed.neocities.org" rel="nofollow">https://loan-free-ed.neocities.org</a>