Rishabh Jangir

Robotics ML Engineer, Nimble Robotics.

I develop end-to-end Deep Learning systems for vision based control of autonomous agents. Through my extensive research experience and publications at robotics conferences (ICRA, ICLR, RA-L, etc.) I have demonstrated the skills necessary to develop complex vision models and achieve sim2real transfer of learned policies trained with RL. Furthermore, my experience at Nimble Robotics (a warehouse automation startup) has geared me with industry standard programming skills and practices. I mostly deal in Python and C++.

I am very grateful to have been granted the opportunity to embark upon a PhD in Robotics journey at the University of Texas at Austin in 2022. After deliberate consideration and talking to leaders in both academia and industry, I decided to find a balance between research and it's application. My curiosity led me towards working on building robotics systems at Nimble Robotics and improving my software engineering skills.

Prior to this I graduated with a MS degree specializing in Intelligent Systems, Robotics and Control at the Electrical and Computer Engineering Department, at University of Californai San Diego (UCSD), advised by Prof. Xiaolong Wang. Even Prior to that I spent 2.5 years in the amazing city of Barcelona working as a research assistant at the Perception and Manipulation group, Institut de Robòtica i Informàtica Industrial. There I worked on robot reinforcement learning with Prof.Carme Torras and Dr. Guillem Alenyà at the Perception and Manipulation Group.

I received my Bachelor's degree in Engineering Physics from the Indian Institute of Technology Guwahati. I have also spent time at International Institute of Information Technology (IIIT Hyderabad) where I worked with Prof.K Madhava Krishna who introduced me to robotics research.

Research interest

I am broadly interested in research on robot learning which stems from my interest in looking for parallels between human learning and artificial learning. I believe human beings are an excellent product of evolution and they could be used as a blue-print for building AI systems. I want to develop algorithms for deploying in real world robots that can learn from human demonstrations, and evolve through self-supervised learning/curiosity. I want to push it towards a stage where robots can learn to carry out a task just by observing a human once. In doing so, I want to explore generality, safety and adaptability. I work with robotics, machine learning, and computer vision.

Publications and preprints

Look Closer: Bridging Egocentric and Third-Person Views with Transformers for Robotic Manipulation
Rishabh Jangir*, Nicklas Hansen*, Sambaran Ghosal, Mohit Jain, Xiaolong Wang
Robotics and Automation Letters (RA-L + ICRA ), 2022
project page / arXiv / code / bibtex
Graph Inverse Reinforcement Learning from Diverse Videos
Sateesh Kumar, Jonathan Zamora*, Nicklas Hansen*, Rishabh Jangir, Xiaolong Wang
Conference on Robot Learning (CoRL ), 2022
project page / arXiv
Self-Supervised Policy Adaptation during Deployment
Nicklas Hansen, Rishabh Jangir, Yu Sun, Guillem Alenyà, Pieter Abbeel, Alexei A. Efros, Lerrel Pinto, Xiaolong Wang
International Conference on Learning Representations (ICLR), 2021 (Spotlight)
project page / arXiv / blog / code / bibtex
Dynamic Cloth Manipulation with Deep Reinforcement Learning
Rishabh Jangir, Guillem Alenya, Carme Torras
International Conference on Robotics and Automation (ICRA), 2020
arXiv / presentation / bibtex
Data Driven Strategies for Active Monocular SLAM using Inverse Reinforcement Learning
Vignesh Prasad, Rishabh Jangir, R. Balaraman, K.M. Krishna
International Conference on Autonomous Agents and Multiagent Systems (AAMAS), 2017 (Extended Abstract)
proceedings / bibtex

Teaching

ECE 276A Sensing and Estimation in Robotics (Winter 2022) by Prof. Nikolay Atanasov
University of California San Diego
Teaching Assistant
Designed and graded assignments for the course, mentored projects and conducted weekly discussions.
website
ECE 285 Introduction to Visual Learning (Spring 2021) by Prof. Xiaolong Wang
University of California San Diego
Teaching Assistant
Designed assignments for first-ever offering of the course, mentored projects and conducted weekly discussions.
website
ECE 176 Introduction to Deep Learning and Applications (Winter 2021) by Prof. Xiaolong Wang
University of California San Diego
Teaching Assistant
website

Selected open-source projects and blogs

Overcoming Exploration from Demonstrations
Rishabh Jangir, Guillem Alenya, Carme Torras
March 2018
Blog / code

Contact

You are very welcome to contact me regarding my research. I can be contacted directly at jangirrishabh [at] gmail .com