I Built a Self-Navigating Robot From Scratch. No ROS. No Framework. 🤖⚡🧠
I built a robot that navigates my apartment autonomously.
No ROS. No framework. No shortcuts. A Raspberry Pi 4, a laser scanner, and Python written from scratch — every single line of logic custom built because I wanted to understand exactly what was happening at every level.
What you'll see in this video:
— YDLidar X4 scanning the environment 7000 times per second
— Custom sensor driver written from scratch in Python
— Real time obstacle detection — walls, table legs, chair legs
— Navigation engine making live decisions: forward, slow, turn, reverse
— Web dashboard for tuning behavior without touching the code
Why build it without ROS? Because frameworks hide the hard parts. When something breaks in ROS you debug the abstraction. When something breaks in raw Python you debug the physics. I wanted to know exactly why the robot turns when it turns and stops when it stops. The only way to know that is to write it yourself.
Every piece of this system is custom: the sensor driver, the navigation engine, the obstacle avoidance logic, the web interface. Nothing borrowed. Nothing assumed. Just code that either works or doesn't — and when it doesn't, you learn something real.
It's not perfect. But it moves on its own. And every bug I fix makes it smarter.
This is what 16 hours a week of side project work looks like.
🔗 Follow the full build:
𝕏 x.com/BuildSensor
TikTok tiktok.com/@sensor_build
LinkedIn linkedin.com/in/alessandro-prandini-iot
🔗 CONSILRED: consilred.com
#Robotics #AutonomousSystems #RaspberryPi #Python #Lidar #BuildInPublic #EmbeddedSystems #ROS #NavigationSystem #ObstacleAvoidance #YDLidar #ESP32 #DIYRobotics #MakerCommunity #IndieHardware #SideProject #Engineering #Innovation #IndustrialAutomation #SensorFusion #RealtimeSystems #PIF #CONSILRED #SensorBuild #MadeinItaly