Working on a circuit with 2 motor drivers and an esp32 for my rc car
In this session, I set up the Raspberry Pi OS and enabled SSH for remote code edits.
Unfortunately, my Pi Camera doesn’t work anymore, so I’ll have to figure out something else.
Next Steps:
- I may step away from a Raspberry Pi as the complexity and time isn’t worth it…
In this session, I completed the 2 servo prototype and tuned PID values for accurate and smooth movement.
I found out that with my current system, the camera and the distance to/from the yellow object causes parallax since the camera is stationary.
Next Steps:
In the next session, I want to step away from my laptops CPU and transfer everything to a Raspberry Pi 3B paired with a Pi Camera V2.1 for more portability.
In this session, I designed and 3D printed my first prototype to test my X and Y servo movement! One of my two prints had incorrect dimensions, so it will have to be done again.
What I want to do next:
Complete programming my Y-axis movement
Reprint the second part
Assemble the prototype and test!
VisionTrack Turret — Devlog. I started building an auto-tracking turret using Python, OpenCV, Arduino, and a servo. I first set up the camera and used HSV color detection to identify yellow objects. I then used OpenCV contours to find the object’s center and send its X position to the Arduino through serial communication. The servo was mapped across the camera’s full 0–640 pixel range to control 0–180°.After that, I added PID control to make the servo movement smoother. I also added position smoothing, a dead zone, and a maximum movement step to reduce jitter. I widened the HSV range so the yellow object can be detected under different lighting conditions. Next Steps: Improve PID tuning and reduce remaining jitter. Add a second servo for the Y-axis. Build the 2-axis turret mechanism. Improve target detection and trackingAdd a target-lock system
Completed PID Controller
Built around half of my new PID Positioning Controller.
Devlog #2 — Connecting Computer Vision to a Servo Today I continued working on the software for my automatic tracking turret. After getting OpenCV to detect yellow objects and find their position, I started working on using that information to control a servo. I set up serial communication between Python and my Arduino so that Python could send the detected object’s X coordinate to the Arduino. The Arduino then uses this value to determine whether the object is to the left or right of the camera’s center. I added a small center area where the servo stays still when the object is already close to the middle. This helps prevent the servo from constantly moving back and forth because of small changes in the detected position. The basic tracking system is now working: OpenCV detects the yellow object, finds its X position, and sends that information to the Arduino to control the servo. Next, I want to improve the movement so the servo moves faster when the object is far from the center and slower when it gets closer.
VisionTrack Turret - Day 1: Getting Eyes on the Project
Today I started building my VisionTrack Turret!
I began by connecting a webcam and using Python + OpenCV to detect objects. I had some issues getting the camera window to stay open, but I learned that testing each part separately makes debugging much easier.
I got OpenCV to detect yellow objects, draw a box around them, and calculate their X and Y coordinates.
These coordinates will eventually tell the turret which direction to move.
Next, I want to connect the vision system to the turret and get it moving based on what the camera sees!
