About

Roshan Pradhan is an accomplished Robotics Autonomy Engineer with a Master's degree in Mechanical Engineering - Robotics from Carnegie Mellon University and a Bachelor's degree in Mechanical Engineering from Birla Institute of Technology and Science, Pilani. He is currently serving as the Flight Controls Lead for the DARPA Subterranean Challenge at Carnegie Mellon University. In this role, he is responsible for leading controls and local planning for the quadcopter fleet to fly aggressively in unseen subterranean environments with degraded state estimation. He has developed a guidance controller, search-based local planner, MPC for reference tracking, and Kalman Filter disturbance estimation. He has also written efficient implementations in ROS/C++, and deployed them on drones for competition. Roshan has a strong research background and has previously worked as a Graduate Research Assistant at the Biorobotics Lab at Carnegie Mellon University. In this role, he worked on making optimal control and MPC feasible for large, modular (’graph-like’) dynamical systems (eg: Graph Neural Networks) by adapting sparse graph optimization algorithms from robot SLAM. He also developed a nonlinear trajectory optimization solver in C++/GTSAM leveraging inference on a factor graph. Additionally, Roshan has worked as a Research Intern at Nanyang Technological University, where he implemented evolutionary algorithms for combined task and motion planning in automated construction. He has a diverse set of skills, including research, search, and neural networks. Roshan is an ambitious and driven individual with 2.19 years of relevant experience. He is passionate about robotics and is always looking for new challenges to overcome. He is a quick learner and has a strong work ethic, which makes him an asset to any team. He is currently joining SkySpecs as an Incoming Robotics Autonomy Engineer.