Program : Dual Degree (Integrated Bachelors and Masters)
Advisor : Dr. Balaraman Ravindan
Areas of Interest : Path Planning, Robotics, Reinforcement Learning
Research Topic : Path planning in non-stationary environments using RRTs
Rapidly Exploring Random Tree(RRT) is a sampling based technique which has been successfully used in many path planning applications as they overcome the curse of dimensionality. However the performance depends on how closely the distance metric resembles the actual geodesic distance between the configurations. RRTPI was proposed which implements ideas of Reinforcement Learning to estimate the distance metric. The current work involves extending the idea to dynamic environment particularly non-stationary domain where the model of environment may change with time in a non-predetermined fashion.
Courses Completed: Machine Learning, Artificial Intelligence, Memory Based Reasoning in AI, Reinforcement Learning(currently enrolled)