Robotics

Augmented the vanilla RRT* path planning algorithm for dynamic environment by implementing replanning of paths through regrowth of nodes and reconnection to original tree. This method of dynamic path planning decreased time of convergence to optimal path by almost 15 times over vanilla RRT* implementation.

Implemented an Unscented Kalman Filter for orientation tracking of quadcopters. The data given was that of IMU (accelerometer and gyroscope). The results of the UKF was compared with the ground truth position data which was obtained from a motion capture system.