I found that EasyEda has a lot of user contributed components for many of the arduino modules and also the arduino and raspberry pi shield boards. So it's pretty easy to take what you would have on a breadboard or strip boards and make it into a pcb. Hope that helps. I enjoy your content greatly
This is AMAZING! you can literally make a commercial product. Can you make more in-detail videos on this? I would love to pay you to teach me some more of this. Can you please send me your email brother
I'm working on version 2 now, should be easier to build. I'll have the video up in a month or so I expect. I'm not really making this commercial, but hopefully that one is simple enough anyone can try it out.
@@nathanbuildsdiy as we speak, I am building it right now. I am doing it with a fixed-wing aircraft. I just have to hook up a raspberry pi now and do training with the Google's tensor
hello, @NathanBuildsDIY i hope you can help me for our final year project as a electrical engineering strudent. wanted to be part of innovation in agriculture as part of sustainability. we want to create a drone that carries a payload of 2 kg minimim or less. but the main point of our idea to innovate a drone that automated flying and spraying, the automated was focus on soil nutrient sensor, we will make a sensor then the data of the sensor will be put in data server or data cloud then, we want the drone to get the data and proceed to the crops that need be sprayed based on the data given, and then after spraying we want the drone to automated back to its original stationary. is this kind of idea attainable/ achievable? if yes i wish you could help us. as part of our final year in college.
That does sound doable, and you wouldn't need any of the AI I added to this. Instead there is a sprayer function built into mission planner. Basically mission planner can include (in the waypoints file) when to turn on the sprayer. So build the route to head over the areas that need to be sprayed (based on sensor data) and turn on the sprayer at those points. I think this is a good start point: ardupilot.org/copter/docs/sprayer.html
I'm working on version 2 now, should be easier to build. I'm not really making this commercial, but hopefully that one is simple enough anyone can try it out.
Not directly. I suppose you could collect that though. If you flew at a constant altitude by barometric pressure (that's the default mode) then took lidar measurements facing down you'd get a rough contour map.
Thank You! Thank you for this video! Following your ideas, I took my R/O line into the basement under the kitchen sink, ran it into a keg in a fridge and back upstairs to be served in a dual spigot next to tap R/O water. I am overjoyed to never have to carry kegs filled at the kitchen sink into the garage and then run back out to the garage 10 times a day to fill up my glass. I live near Corny Keg, and the owner was very nice in encouragement and ideas for getting my system up and running. Rough cost for new system when all was said and done: $500. Having cold fizzy water that requires practically NO further work: priceless! I was a little on the obsessive side when making sure the fridge in the basement was up high so I would not have to waste a lot of Carb water waiting for it to get cold. As it turns out, I used something like 1/4" tubing from Corny Keg, and it takes no time at all to get to cold. I could have gotten by with not having the top of the fridge at ceiling level. It was a little difficult to work with the 3 tubes coming out of the fridge so close to the ceiling. My tubing goes through the top of the fridge. I was a little wary of drilling holes in the side due to possibly hitting a coolant line.
Does your R/O tap disturb the water too much? I just got one of these taps but I think mine is taking a lot of the carbonation out of the water, it comes out cloudy just like the one in this video. I'll still do some tests but previously when I used a picnic tap the water wasn't disturbed and came out super fizzy.
You can make a very effective apple crusher from a sink garbage disposer. Just mount to the bottom of a board with an appropriately sized hole, put together a simple drain line aimed at a bucket, and away you go. Apples go in the top, pulp (I think the term is pomace?) lands in the bucket. Then press and process as you like.
This is great content. I think its time time you moved on from jumper wired modules to PCBs. I have just started to use Kicad and Easyeda and created PCB's to hook stuff together, its so much easier. I'm sure if you asked PCBWay or JLCPCB they would jump at the chance to provide free boards for you and using an EDA make is much easier to keep track of connections even if you don't create an actual PCB. Love your work, keep making video's!
Yellow or any color really should work fine. This thing is slow and doesn't take hard corners or anything like that. A fast flying drone might care about which cube is in there, but this one doesn't!
It could! That's why I talked about warnings - Don't run it where there is dry fuel load. On a dirt bed (there just isn't fuel to get a fire going) or after a rain is where I run it.
So do you have a 24v system being charged by 12v , or is it a 2 batteries in (i belive its ) parallel for longer range. I just built mine and im haveing a hard time to figure out the charging system. Thanks and i love the project.
yeah - 12V system, you can use 2 12V batteries in parallel. I had to go with that so I could use a 12V alternator. I don't know of any 24V alternators. Honestly I think I can cut back to 1 12V battery, because the alternator has been keeping it charged
Why Rear L wheel is also on Pin 9? isn't that gnd? And Dir is also pin 25 which is also GND. Or are these not the pin numbers but the GPIO numbers? Sorry I'm just a little confused
I think it's confusion about board pin numbers vs. GPIO pin numbers. www.raspberrypi.com/documentation/computers/images/GPIO-Pinout-Diagram-2.png?hash=df7d7847c57a1ca6d5b2617695de6d46 The little circle there with 9 (left side 5 circles down) in it is connected to GND, but that's the board pin number. I was talking about GPIO 9 which is connected to the little circle 21 (left side 11 circles down). In the python script with gpiozero library it uses the GPIO pin numbers, so when you say "9" in the python script it actually thinks you mean gpio 9 which is on the left side, 11 pins from the top. Make sense?
The flow for creating a custom image classification model and running it has been updated quite a bit. I've nearly got version 2 of this thing complete where that'll be explained. Basically there is a google colab that greatly simplifies creating a custom image classification model from a few hundred sample images and you can use a raspberry pi zero 2 w and global shutter camera to do classification on board, no need for a video link.
Looking to do something similar with a project, may I ask why you went with an AC alternator to converter instead of just a DC alternator? Also have you noticed a drop in HP after adding it?
Sensor suggestion: touch lamp on/off technology or measuring millivolts or continuity! Grass is conductive. Use A few (4?) copper wires stretched across front should give you a general idea of grass height, and left or right side. Plus it’s super cheap.
You do a really good job explaining everything. I find it very difficult to make a video like this. I build projects like this all the time but I'm terrible at making videos. So they never see the light of day. I think that your 3D printed touch sensors probably just need a slight redesign. I think those were pretty good idea. Also I would definitely add the kill switch. If you take the cover off the motor you'll find a magneto with a black wire coming off. Grounding that wire out kills the engine. You may be able to control that Killswitch wire with a transistor.
Any chance you've been able to do that parts list? i'm trying to copy what you've done, but also fit it in a kitchen cabinet under the counter. Hard to tell how big things are, how tall the fridge needs to be, etc. Would really appreciate a very specific parts list so I don't end up buying the wrong things. Thanks!
For grass height. What about angle measurements, attach some rollers with a hinge, measure the angle depends how long the grass is. Would need to be light weight as not to squish the grass to much.. great projects can't believe why you don't have more subs, your projects are amazing.
Next upgrade could be precise GPS, how does that work? The magician's staff with the gadget on top that makes the GPS signal cm precise. RPI lidar V2 is also pretty cheap comparatively.
Ultrasonic sound is not heard by the human ear. Nor is it created by the loud environment. Super high frequencies, like ultrasonic sound only travel in straight lines like a focused light beam.! The word sound in the term is a bit of a misnomer..
@@beckettrj Sorry this is not true. Just because you can't hear it, it doesn't proof that the mower don't produce ultrasonic sound. In fact, moving air (like around the mowing blades) usually generates white noise ... so basically all freaquencies. This is not focused at all but jammes every sensor near by. For utrasonic sensors in vehicles I know for sure that the noise of releasing truck brakes (released air pressure) is a big problem.
I asked Chat GPT if a camera could be coupled to a raspberry pi to control a grass mowing robot. It sounded like the answer meant it would be possible. This is what it told me, if I could paste the answer here. Since I don't know a dog gone thing about how to do something like this, it seems like it could be a feasible answer to get your robot to follow a cut path better to me. You have great ideas and I loved your video. This is not something I could do, I don't have the required skill, but somone like you might be able to get it to work well. +++++++ You said: Woiuld it be possible to program a camera to identify a line of cut grass on a lawn using code embedded in a raspberry pi computer board. ChatGPT said: ChatGPT Yes, it is absolutely possible to program a camera to identify a line of cut grass using a Raspberry Pi. This can be done by employing a camera module connected to the Raspberry Pi along with computer vision techniques. Here's an outline of how it could be done: Key Steps: Camera Module Setup: Connect a Raspberry Pi camera or compatible USB camera to capture real-time video or images of the lawn in front of the mower. Image Processing with OpenCV: OpenCV (Open Source Computer Vision) is a widely used library for image processing. You can use it to analyze the video feed, detecting differences in texture, color, or height between cut and uncut grass. Grass Line Detection: By processing the images, you can look for differences in pixel values that distinguish cut grass (usually more uniform and flatter) from uncut grass (likely darker and taller). This difference in texture or shading could allow the system to follow the boundary between the two areas. Edge or Contour Detection: OpenCV has built-in tools for edge and contour detection. You can use these tools to identify the transition line between cut and uncut grass. This line can be treated as a guide for the mower to follow. Path Following Algorithm: Once the line is identified, you can create a simple control algorithm that adjusts the mower's direction to stay aligned with the boundary of the cut grass. Obstacle Detection: OpenCV also allows integration with other computer vision techniques like object detection (using pre-trained models like YOLO or TensorFlow), which could help the mower detect and avoid obstacles. Additional Considerations: Processing Power: While a Raspberry Pi may be limited in processing power compared to more robust systems, it can handle basic image processing and camera feed with OpenCV. You may need to balance the resolution and frame rate of the camera feed to ensure real-time performance. Machine Learning (Optional): For a more advanced solution, you could even train a machine learning model to recognize cut vs. uncut grass. However, this would require collecting image data and training the model separately, which might demand more computational power. Hardware Requirements: Raspberry Pi (preferably a Raspberry Pi 4 for better processing power) Raspberry Pi Camera Module or USB Camera Python libraries (OpenCV, NumPy, etc.) Example Python Code (simplified): python Copy code import cv2 # Initialize camera cap = cv2.VideoCapture(0) while True: ret, frame = cap.read() # Convert image to grayscale gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY) # Detect edges using Canny edge detector (basic example) edges = cv2.Canny(gray, 50, 150) # Display the result cv2.imshow('Edges', edges) # Break the loop on key press 'q' if cv2.waitKey(1) & 0xFF == ord('q'): break # Release resources cap.release() cv2.destroyAllWindows() This code captures video, converts it to grayscale, and applies edge detection. You can build on this to detect the grass cut line. With this approach, you can definitely make your robotic lawnmower more accurate and efficient in following cut paths. +++++++++ Good Luck and I wish you great success with your project!
nice project. I made a very similar pair of robots a couple of years ago. One cleaned floors, one just roamed around and talked. Good job, this really is incredibly similar (mine had ESP32 cores and roamable http management). Well done
Hi, I'm looking into making my own mower too. It seems that all Ublox NEO-6 GPS can do RTK roving. And the Ublox NEO-6P is needed for the base station, that seems to make GPS RTK at slow speed quite affordable. I'm waiting for my GPS modules in the mail to check if I can really do slow RTK for cheap.
Would a Lidar help to map the terrain? There is some pretty reasonable ones these days Andreas Spiess put one up ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-ddxguAzzzJE.html
I like your work with ArduPilot I have a robotic tractor tractor I am making for small farms hear is a link to video I made a few years ago ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-pCdD5y2RGsc.html just wondering how much you would charge to get ArduRover working on it. I really like you AI Solar Weeder I would love to see that on the FarmHand tractor.
I've seen quite a few robot mower videos... some DIY and some retail and I'm always left wondering why these things don't simply use a 40-volt weed-eater. Very safe and effective. a 2ah battery would run the cutter for 30 minutes. I envision a robot mower that you slot in 2x 6ah 40v batteries, and it can mow for 90-120 minutes, with very little risk to life.
what about using two copper strips mounted parallel to the ground and when they come into contact with the grass you can measure the electrical resistance
Very cool. I don't think throttle control will help you much. You'd be better off using a relay and grounding out the magneto. That's usually how the kill switch on most mowers and 2 cycle engines work.
