About

Gender

Male


Location

Kolkata, West Bengal


About Me:

I'm a 17 year old student from Kolkata , India. I'm currently a community developer for the PX4 Flight stack. I'm also a MAV consultant and systems integrator. I specialise in vision based estimation and control for UAVs. I'm interested in everything, from sustainable energy to artificial intelligence and neural network technology.


Please tell us a bit about your UAV interest

I specialise in vision based estimation and control for UAVs. This includes GPS denied navigation and sense and avoid. Visit us at www.uasys.io/research


Hometown:

Kolkata


Activity Feed

Kabir commented on Jerry Giant's photo
"Looks cute!"
Mar 12, 2017
Kabir commented on Bernardo Espinosa's blog post RTF Low Cost Professional Photogrammetry Drone, with GNSS RTK module, PPK, CORS and Ground RTK Support
"Nice design! Can you post some closer shots and more details of the gimbal?"
Mar 1, 2017
Kabir commented on Chris Anderson's blog post Airbus Cargo Drone using Dronecode/PX4 stack
"^ that."
Oct 15, 2016
Kabir commented on Zubax Robotics's blog post PX4 Sapog - advanced open source ESC from the PX4 team
"Sapog does not run on ESC32 hardware."
Aug 24, 2016
Kabir commented on Chris Anderson's blog post Intel announces new Aero drone development board based on Dronecode PX4 software
"Rob, I don't know when you last tried PX4 (maybe a couple of years back), but PX4 is definitely the premier flightstack for advance navigation and control integration. It's very much flyable.
About the open source / proprietary split, all features…"
Aug 17, 2016
Kabir commented on Nicholas Witham's blog post Pixracer (actually a counterfeit): Now available from Hobbyking!
"The Pixhawk hardware design was never 3DR's to open source."
Jun 15, 2016
Kabir replied to Forrest Haller's discussion What are the standard CAN protocols that drones use?
"They were spec'ed for different purposes.
CAN is a high speed onboard interconnect. MAVLink is an offboard interconnect, primarily designed to interface with a GCS or other MAVs."
May 28, 2016
Kabir replied to Forrest Haller's discussion What are the standard CAN protocols that drones use?
"The CAN specification has existed since the 90s. Not the UAVCAN protocol.
Mavlink is pre-2009, from the original Pixhawk student team.
UAVCAN came later."
May 28, 2016
Kabir replied to Forrest Haller's discussion What are the standard CAN protocols that drones use?
"Mavlink has existed long before UAVCAN."
May 28, 2016
Kabir commented on Chris's blog post Queen B Robotics introduces the Exo360! 5 integrated 4K cameras for the ultimate VR experience
"Hi Chris, as Gary and Rob asked, could you please clarify whether the footage you show in the video come from your actual vehicle?
It seems to me that the clips are not. Firstly the distinct lack of radial distortion from cameras so wide, and the…"
May 25, 2016
Kabir commented on Chris Anderson's blog post Intel gives sneak peak of drone development board
"The R200 reference design is actually a hybrid system. It's not just structured light like a Kinect. The dual cameras work like a normal stereo pair outdoors, albeit with a rather small baseline. The ASIC on the R200 does all the processing, so…"
Apr 19, 2016
Kabir replied to muavdrones's discussion Current architecture of drone autopilots is wrong , Drone hardware architecture needs to be completely re-engineered.
"Multi threaded FCs are nothing new. APM is sort of multi threaded in portions, and PX4 is fully modular, with each module (estimator, controller, sensor drivers, etc.) being a separate process which is connected by a very fast publish/subscribe bus.…"
Apr 13, 2016
Kabir replied to muavdrones's discussion Current architecture of drone autopilots is wrong , Drone hardware architecture needs to be completely re-engineered.
"And if it comes to better hardware with isolated processing, we're already doing that today. The Snapdragon Flight for example has a multicore Krait applications processor and a Hexagon DSP at 800mhz running a RTOS and the timing critical flight…"
Apr 12, 2016
Kabir replied to muavdrones's discussion Current architecture of drone autopilots is wrong , Drone hardware architecture needs to be completely re-engineered.
"Indeed. As someone also brought up sensor aliasing effects, if you consider the PX4 flight stack, sensor polling is at 2khz to mitigate effects of coning, etc. Running control loops at that rate is just a waste of processing though.

The only…"
Apr 12, 2016
Kabir commented on Juraj Kabzan's blog post Quadcopter able to track an object in video frame and follow it without permanent connection to GCS
"Is this actually actually accelerated with GP-GPU?
I've been able to achieve 30fps, limited by camera framerate on a ARM CPU only. No parallelisation, just fairly optimised algorithm."
Apr 8, 2016
Kabir replied to muavdrones's discussion Current architecture of drone autopilots is wrong , Drone hardware architecture needs to be completely re-engineered.
"Hz are the megapixels of drones.

A faster control loop does not translate to better physical control. Can you show real data which can prove otherwise?"
Apr 5, 2016
More…