I graduated with an MS degree from the University of Michigan in the Robotics Program in May 2020. I am passionate about AI, perception, machine learning to allow robots perceive this beautiful earth and be an essential part of it!

Have an interesting idea that we can work on, or want to discuss how we can make robots more human? Contact me...

LinkedIn                                 GitHub

      Email: kaustavc[at]usc[dot]edu

News: I am excited to announce that I have accepted a Ph.D. position in Electrical and Computer Engineering with an Annenberg Fellowship at the University of Southern California from Fall 2020

My current website is hosted here!  


"How to make robots more human?"

              ..........The main motive of my research is to find an answer to this question.  I am interested in the intersection of vision and robotics. How can we use vision as a stimulus and convert it into optimal actions that a robot should perform to reach its goal configuration? Can a robot "learn" to selectively "perceive" its environment and "control" its own actuation? 

I am working on object detection and pose estimation using deep learning on LiDARs under Prof. Jessy Grizzle at the Bipedal Robotics Lab at the University of Michigan. Additionally,  I am also a part of the UM Ford Center for Autonomous Vehicles, where I worked on extending the Pedestrian Prediction Project under Prof. Ram Vasudevan.

Apart from this, I have worked on mobile robotics, SLAM, computer vision, and a multitude of small robotic projects. 

I graduated with a BTech degree in Mechanical Engineering in 2018 from VIT University where I have worked on material science, mechatronics, and embedded robotics. I have been a member of the ASME-VIT student chapter and a department chair of  Team Anant, the official Human Powered Vehicle Team of VIT, Vellore operating under ASME-VIT.

CV (updated Feb 2020):

PROJECTS

Temporally Conditioned Deep Neural Network for classification and pose estimation of LiDARTag

Biped Robotics Lab(2019 - Present )

Stay tuned for more details.....

Deep Lagrangian Networks Applied to Underactuated Systems 

Machine Learning (2019)

Incorporating Lagrangian mechanics to Deep Neural Nets

[code]   [report]   [poster] 

JORB-SLAM: A Jointly optimized Multi-Robot Visual SLAM

Mobile Robotics(2019)

Multi-agent ORB-SLAM

[site]  [report]  [code]

Modified Style Transfer

Computer Vision(2019)

Add more than one Style to an image plus tweaks to make it more realtime!

[report]

LSTM-Pose-Machines

Biped Robotics Lab(2019)

My own implementation of LSTM-Pose Machines using Tensorflow

[code]

Perception: Self-Driving Cars

Self Driving Cars: Control and Perception(2018)

Implementation of  VGG based car class classifier in Matlab

[code]

Phone Finder 

Basic OpenCV implemenation of image segmentaion and thresholding to detect objects(2018)

[code]

Design and Analysis of Industrial Manipulator

Design and analysis of a mobile platform for a assembly plant(2017)

[poster]


Two-Wheeled Self Balancing Robot

PD-controlled Self-Balancing Robot(2017)

PUBLICATIONS

Effect of flake reinforcement on mechanical properties of AA 6061 nano composite with secondary nano platelet-Graphene processed through powder metallurgy

H.G.Prashantha Kumar M.Anthony Xavior Ashwath.P. Joel J. Kaustav Chakraborty

[Paper]

Laser Shock Peening on Microwave Sintered Aluminum Alloy Nanocompo-Sites

S. Prabhakaran, ; G., Prashantha Kumar H; Kalainathan, S; M., Anthony Xavior; Chakraborty, Kaustav

[Paper]

Synthesis and property evaluation of hot extruded AA2024 -MWCNT Nanocomposites

H.G.Prashantha Kumar M.Anthony Xavior P.Ashwath J.Joel Shivprasad Hosmani Prateek M.Shivalli Hrishikesh Singh Kaustav Chakraborty

[Paper]

Create your website for free! This website was made with Webnode. Create your own for free today! Get started