Chirag Modi

@cmodi3006
0 Followers
1 Following
27 Posts
Automated driving, F1, tech... that's where my mind is.

Hot take: the "LiDAR vs camera" debate isn't really about sensors. It's about how much you trust an AI to infer depth correctly at 120 km/h.

#selfdrivingcars #adas #autonomousvehicles #sensors

I am very happy to announce our talk titled "EasyNav: An open-source framework for navigating everywhere" has been accepted for #ROSCon2026 🚀🥳

With Juan S. Cely and all the #EasyNav Team at @IntellRobotLabs

We obsess over #LiDAR sensor resolution in #self-driving cars. But a "perfect" point cloud captured from a moving vehicle is already wrong before you process a single point. Sometimes the upgrade isn't hardware — it's correcting for time and motion.

Looking back at my already published blog on Medium, I thought of sharing a general idea of #KalmanFilter to help people understand what it is actually doing. Check it out:

https://open.substack.com/pub/cmodi306/p/the-simplest-kalman-filter-youll?utm_source=share&utm_medium=android&r=2zfweu

Most Engineers Use the Kalman Filter. Only Few Understand What It's Actually Doing.

A few weeks ago, I published a blog where I showed a very simple and basic C++ implementation of a Kalman filter along with two plots, and a confession that I ended up calling Python from C++ anyway.

CModi's Substack

Hi #ROS2 devs, introducing ROSMAN (ROS2 Manager) - an interface that allows you to control your nodes via a UI that runs on your localhost. It uses #FastAPI on the backend and a simple web frontend, with #Python in the backend.

GitHub: https://github.com/cmodi306/rosman_app

I've kept it free and open source, so feel free to customize it, use it and do share your feedback. :)

GitHub - cmodi306/rosman_app

Contribute to cmodi306/rosman_app development by creating an account on GitHub.

GitHub
AI - a tool for the wise; a weapon of self-destruction for the naive.

The concept you'll find in almost all of the #selfdrivingcars - Kalman Filter.
I just published The Simplest Kalman Filter You’ll Ever See (And It’s in #C++)

https://medium.com/@cmodi306/the-simplest-kalman-filter-youll-ever-see-and-it-s-in-c-b1d2652ec36c

The Simplest Kalman Filter You’ll Ever See (And It’s in C++)

If you’ve worked with sensors, robotics, or anything that moves and needs to know its position, someone has thrown the words “Kalman…

Medium
@TheBadPlace (2/2) So when the raw data from the sensors contains invalid points, any system would have problems working on it. So the author needs to provide more information regarding the system and the test evaluation. Just my 2 cents :)
@TheBadPlace Interesting read. Would like to know more about this AI that imitates human. However, one thing I would want to note here is in the part where they test the system, they do not mention any test in rains. That's a crucial scenario to test in because lidar could suffer from "hallucinations" in rain (due to light rays getting refracted due to water drops) and camera will have problem in vision too. (1/2).
Now that I have Prism running also in my raspberry pi, time to step up. Finally got the stone of my home server rolling! #homelab #homeserver #selfhosting