I have an immense interest in autonomy.  I started off working with autonomous vehicles. For the past couple of years I have been working in bringing automation into the field of Histopathology, a field that hasn't seen innovation in nearly 100 years.

Most of the work that I have done has derived from personal experience. Like for eg. During the testing of our vehicle during college, I realized the importance of error detection and logging in autonomous systems(video below). First of all, they can be fatal as the following video shows. 

Secondly, in some ways, error prevents the system from being autonomous. Logging the errors allow us to understand these challenges and pave the way towards complete long-term autonomy. 

Currently, I am working in bringing automation to the medical field of Pathology, particularly cancer pathology. The diagnosis of cancer is a long process often requiring human tissues depending on the kind of diagnosis. The sample is processed in many stages for almost all parts manually. I am working as an Automation Developer at Inveox GmbH to automate this process with a vision of diagnosing cancer in one hour.

Previously, I have worked in Research & Development of Autonomous Indoor Vehicles for industrial applications at Hi-Tech Robotic Systemz. One of the applications that I have worked on solving is highly accurate localization for safety-sensitive operations. I developed a landmark-based localization to augment a global landmark-less(natural) localization for such operations.

Take a look at the robots I work with, in the video below.

I have also made a lightweight simulator in ROS to simulate localisation algorithms. Its an ongoing thing that helps me at times. It takes map input as an image and generate necessary tfs and laserscans to simulate localisation.This project is open source and can be accessed from Github

A more detailed overview of the work I have done can be found in my CV.