Capstone II Spring 2024 - Short Version

Kate Morgan Author

06/11/2024 Added

25 Plays

Description

Learn about the research students on Scott Campus did in the spring semester 2024. Short version.

Searchable Transcript

Search:

[00:00:00.375](upbeat music)
[00:00:06.690]So, this is Object Detection Drone.
[00:00:08.340]So, what it is, (crowd bustling)
[00:00:09.720]we're controlling a drone
[00:00:11.520]through our boards that we designed here.
[00:00:15.510]We're going to survey a field,
[00:00:17.640]detect whatever object you want it to detect,
[00:00:19.890]but really, the idea is
[00:00:22.320]we wanna be able to identify weeds with it,
[00:00:24.990]so that way a farmer can use it for weed management,
[00:00:28.560]without having to take a plane out
[00:00:30.630]and crop dust the entire field.
[00:00:32.490]So the idea is,
[00:00:33.323]if we can identify where the weeds are using AI,
[00:00:37.080]then we can reduce the over-spray that would be involved
[00:00:41.100]with just applying herbicide to the entire field.
[00:00:44.490]So yeah, our device,
[00:00:46.290]it flies over this field,
[00:00:50.291]it'll take pictures,
[00:00:51.150]use that AI to identify what's there,
[00:00:53.520]and then we can attribute a GPS coordinate
[00:00:56.460]to each detection,
[00:00:58.770]so that way we can go back out and do the spray later.
[00:01:02.310]So, we use the golf balls because of the season
[00:01:04.950]that we are working on this project in,
[00:01:06.720]we're working on it over the winter,
[00:01:07.890]so we're not gonna have weeds that we can detect.
[00:01:10.170]However, the nice thing is,
[00:01:12.930]with AI object detection models,
[00:01:15.630]whatever sample set you give it
[00:01:17.940]is what it can detect.
[00:01:20.010]So, if we wanted to refit our model for weeds,
[00:01:25.650]all that we would need to do
[00:01:26.790]is get a bunch of pictures of weeds, retrain it.
[00:01:28.890]So it would be the same AI model,
[00:01:31.650]just with a different training set,
[00:01:33.387]and it would work the same way with our system.
[00:01:38.610]So, it's like a Life Alert,
[00:01:40.950]when somebody falls down in a small room,
[00:01:43.170]it detects the person falling down,
[00:01:44.910]and then you can contact anybody for them,
[00:01:48.360]they can set up their phone number
[00:01:49.920]for who it should contact,
[00:01:51.420]and instead of them actually pressing a button
[00:01:54.570]to say that they'd fallen down,
[00:01:56.760]this system is actually built on radar,
[00:02:00.420]so it actually scans the room all the time
[00:02:03.437]when somebody is in,
[00:02:04.420]it always checks if the person has fallen down.
[00:02:07.493]Then if the person did fall down, it triggers,
[00:02:09.750]and then asks the person we detected had fallen,
[00:02:13.057]"Would you like us to call for help?"
[00:02:14.717]And if you say yes, or you say no,
[00:02:17.760]it's actually gonna call the person,
[00:02:19.128]or like sending text to the person.
[00:02:22.920]We really wanted to do something as a proof of concept,
[00:02:25.560]show that it can be done.
[00:02:27.060]And then when people have more time,
[00:02:29.070]and more investment into it,
[00:02:30.750]more than just a capstone project for us,
[00:02:33.180]we're trying to get a degree,
[00:02:34.530]hopefully someone that is really, really, really
[00:02:36.240]invested in this can take it somewhere,
[00:02:37.073]and market it, and try to save lives.
[00:02:42.360]This is project Quick Grade, we're team 124,
[00:02:45.060]and our main purpose is to have a base station air rover
[00:02:48.510]to aid in construction grading on job sites,
[00:02:51.270]to get accurate longitude, latitude, elevation data,
[00:02:55.170]to minimize human error when trying to find the correct spot
[00:02:58.200]to dig or filter on a construction site.
[00:03:01.380]And here we have our base station,
[00:03:03.960]and then we have our rover set up over here,
[00:03:06.930]the antenna,
[00:03:08.670]and then we have our app demonstration over here
[00:03:11.526]that connects to our rover, and displays PDFs,
[00:03:14.970]and then the measurements taken.
[00:03:18.093]Yep, so it'll tell you who's taking your measurement,
[00:03:20.370]and then the height of the rod that the rover's sitting on.
[00:03:23.910]And then the blueprint will come up,
[00:03:26.430]everyone can see what the benchmark elevation's
[00:03:28.500]supposed to be.
[00:03:29.490]And you click measure,
[00:03:30.780]and then it'll tell you longitude, latitude, and elevation,
[00:03:34.350]and then who took the measurement.
[00:03:36.000]The data is transmitted between the Gen Assess model
[00:03:40.200]and the satellite,
[00:03:41.033]and I sort of have to find a way to intercept that message,
[00:03:45.780]and then find the data that I'm looking for,
[00:03:50.100]look at specific messages,
[00:03:51.750]also have to make sure that the data is specific enough,
[00:03:56.250]because since it's measuring longitude,
[00:03:59.970]usually it's not accurate all the way to an inch,
[00:04:03.420]so I have to set it into a mode
[00:04:05.430]where it can measure to the inch.
[00:04:11.010]We created the Sustainable Small House
[00:04:14.340]Indoor Air Quality Monitoring System.
[00:04:16.230]It's for Dr. Chen's sustainable small house projects
[00:04:19.980]that they're building here, just south of campus.
[00:04:22.710]Indoor air quality is important for the occupants' health,
[00:04:25.560]so we needed a system to monitor such a thing
[00:04:28.560]as particulate matter levels, carbon monoxide levels,
[00:04:32.370]sulfur dioxide and ozone levels,
[00:04:34.710]and all those things are used
[00:04:35.730]to create an air quality index,
[00:04:38.700]similar based on the one for from the EPA.
[00:04:42.450]So like, if you ever see on the news,
[00:04:43.830]when they give a weather forecast,
[00:04:45.060]they have an air quality index up there,
[00:04:47.610]that's the same type of thing that we're calculating,
[00:04:49.620]but we're doing an active monitoring system.
[00:04:52.080]So, we created three modules,
[00:04:54.450]that will be in the three main rooms of the house:
[00:04:56.760]the basement, the utility room, and the main living room.
[00:05:00.060]And it will constantly collect data,
[00:05:02.610]and report it back to the HomeSeer automation system,
[00:05:05.820]which, if someone will eventually
[00:05:08.760]completely automate the house,
[00:05:10.260]and we'll make decisions to where,
[00:05:11.790]when there's certain bad air quality inside,
[00:05:15.000]and it's worse than the one that's outside,
[00:05:18.000]it'll be able to turn on a ventilation unit
[00:05:19.560]that'll swap the air.
[00:05:23.880]So, our project is the Localized Home Assistant.
[00:05:26.370]It is essentially,
[00:05:27.390]the closest type of product would be
[00:05:29.040]sort of an Amazon Alexa for Google Home.
[00:05:31.530]Our device has no connection to an ISP,
[00:05:33.480]no internet access at all,
[00:05:34.680]it hosts its own local area network with a webpage on it.
[00:05:37.313]You can actually see here,
[00:05:38.520]it's the webpage that it's hosting right now, live.
[00:05:41.670]You can add things like flight schedules to it,
[00:05:44.400]add new smart devices that it'll be able to control,
[00:05:47.130]control the timers on it.
[00:05:49.140]Traditionally, it has things like voice control,
[00:05:52.440]as well as, like we already said,
[00:05:54.690]the ability to control smart lights.
[00:05:56.790]The type we're using in our design right now are Sengled,
[00:05:58.874]but it should be able to use anything.
[00:06:01.620]That's sort of the basis for our design.
[00:06:08.520]So, this is P.L.A.N.E.T,
[00:06:09.600]which stands for Plant-based Long-term Autonomous
[00:06:11.670]Nurturing Eco-friendly Technology.
[00:06:13.440]A bit of a mouthful.
[00:06:14.273]But how it works is,
[00:06:15.300]it uses four different sensors in the soil,
[00:06:18.030]and the two that we use for the healthcare functions
[00:06:19.680]are soil, moisture, and pH.
[00:06:21.300]Based on those, it'll automatically water and fertilize it.
[00:06:24.030]So, it'll use those sensors twice a day,
[00:06:25.650]and then it'll just take care of the plant for you.
[00:06:27.300]And we also have different settings
[00:06:28.980]within our home screen down here,
[00:06:31.050]and we can select between four different plant types,
[00:06:33.720]in the home screen right here.
[00:06:36.240]And depending on the plant that you select,
[00:06:37.830]it'll take care of the plant differently,
[00:06:39.000]because a lot of plants are different,
[00:06:40.380]so you have to be able to water it less frequently,
[00:06:42.570]or more frequently, depending on the plant.
[00:06:44.430]Then we also have a light attachment.
[00:06:45.930]Plants need the visible spectrum light
[00:06:48.990]more than they need UV light.
[00:06:50.580]And so that's why we chose these light colors,
[00:06:53.670]as I believe we explained,
[00:06:55.230]the colors shift throughout the day.
[00:06:56.910]Another big thing that we wanted to focus on
[00:06:59.160]was sustainability, so we used a lot of recycled materials,
[00:07:02.910]and we have a very low energy consumption.
[00:07:05.430]At the end of the day,
[00:07:06.263]we're using like five watt hours per hour,
[00:07:08.400]so it's very low compared to like, a light bulb,
[00:07:11.430]and most of our packaging is recycled,
[00:07:13.650]so we're making less of an impact on the environment
[00:07:16.710]while trying to grow plants.
[00:07:18.990]That was our mission.

The screen size you are trying to search captions on is too small!

You can always jump over to MediaHub and check it out there.

Comments

0 Comments

Capstone II Spring 2024 - Short Version

Description

Searchable Transcript

Comments icon comment

Related Channels

Comments