Embedded Files
A new transcription, change point localization, and mesh refinement scheme is presented for direct optimization-based solutions and for uniform approximation of optimal control trajectories associated with a class of nonlinear constrained optimal control problems. Our method proves to be more superior in terms of quality of solution, particularly for singular optimal control problems, when compared to existing state of art methods.
Reinforcement learning assisted Model Predicitve control
Reinforcement learning assisted Model Predicitve control
Pure Reinforcment learning (RL) strategies often remain impractical in process control due to long and expensive training and absence of safety guarantees. We present a practical RL supervised MPC approach that uses a closed-loop re-identification procedure to correct for plant-model mismatch, with dynamic tuning of MPC parameters using RL updates.
Control for Quadcopter trajectory tracking
Control for Quadcopter trajectory tracking
Quadrotor maneuvering in three dimensions in a tightly constrained setting is a compelling research problem. We employ time optimal minimum snap polynomial spline trajectories subject to obstacle avoidance and actuator constraints, that are then tracked using robust (upto some degree) reference tracking methods. We also develop pipelines and test algorithms on Pixhawk Gazebo MAVROS and hardware. (GitHub)
Terrain relative optical navigation
Terrain relative optical navigation
Existing lunar navigation technology suffer from poor performance in pinpoint lunar landing missions. This work employs a feature detection strategy to develop a high-accuracy visual-based positioning algorithm for inertial state estimation during lunar descent. (Github)
Page updated
Google Sites
Report abuse