Multiple Openings for Autonomous Navigation in Robotics
Prof. Mohanty's Autonomous Navigation for Robotics Lab is seeking passionate students to work on cutting-edge projects related to increasing the deployment of autonomous vehicles in land, aerial and space terrains.
The lab has a wide variety of ongoing projects in the areas of sensor fusion, machine learning, state estimation, such as:
- Creating a distributed simulation platform for testing the use of ultra-wide band sensors for cooperative navigation in lunar terrains and validating initial results using real-world rovers such as Turtlebots.
- Improving GPS navigation in dense urban areas by learning context from vision language models.
- Creating a framework for optimizing 3D reconstruction (neural radiance fields) from 2D images from multiple drones and validating the framework with real-world experiments.
- Creating synthetic datasets for increasing reliability of AI algorithms for Mars rover by using diffusion models and game engine, such as Unreal Engine.
As a mentor, I believe in creating a collaborative and supportive environment for my students. I meet with each student 1-2 times a week to help address concern and set achievable milestones. Students are expected to maintain weekly meeting notes to track their progress. Beyond individual meetings, all students are expected to contribute to lab brainstorming sessions and be proactive problem solvers. My goal is to equip my students with a set of hard/soft skills that makes them best prepared for future careers and/or graduate school in broad areas of machine learning/robotics/aerospace systems.
By joining this lab, you will have the opportunity to:
- Develop a deep understanding of autonomous systems, sensor fusion, and machine learning.
- Increase your expertise in commonly used hardware platforms, software frameworks, simulation platforms in robotics.
- Gain an in-depth understanding of setting up and managing physical testing facilities.
- Structuring real-world and simulated experiments for autonomous systems.
- Exposure to machine learning techniques, including neural networks, diffusion models, and Neural Radiance Fields.
- Collaborate on publications for leading conferences and journals.