Pranav Thakkar

Hi! I am a second-year PhD student at Cornell University, where I am advised by Prof. Tapomayukh Bhattacharjee in the EmPRISE Lab. My research focuses on using robots to dress people with mobility limitations, with everyday clothing. I am interested in haptic perception, deformable object manipulation, and safe human-robot interaction.

Previously, I graduated from the Indian Institute of Technology, Bombay where I worked with Prof. Leena Vachhani and Prof. Hemendra Arya. At IIT Bombay, I worked on robust state estimation techniques for ground robots with access to only bearing measurements.

I was also a research engineer at Honda R&D in Wako, Japan, where I worked on robust robot perception and mapping.

Email  /  CV (not up-to-date!)  /  Github  /  LinkedIn

We collect in-the-wild multimodal haptic data at scale to create visuo-haptic perception models that enable real-world robot manipulation.

We introduce a method to infer a users’ functional range of motion from occupational therapy assessments, to enable personalized robot-assisted caregiving.

We discuss the structure of unobservable spaces for a family of joint state & parameter localization problems, for a ground robot with access to only bearing measurements.

We develop an India-specific compartmental model that incorporates regional healthcare capacity and mobility to project COVID-19 spread and assess intervention strategies.

We show that a class of tunable nonlinear Kalman filters work well towards robust state estimation without having to re-estimate landmark locations.

A comparative analysis of heuristics-based algorithms that select a subset of landmarks for robot navigation with access to only bearing measurements.

A compilation of my work over two years at the ARMS Lab, ranging from observability analysis to robust state estimation for ground robots with only bearing measurements.

Website template borrowed from Jon Barron.