Robotics Graduate for hire
ROBOTICS AND INTELLIGENT VEHICLE RESEARCH (RIVeR) LABORATORY , WPI
Master's thesis : Addressing task space constraints in C-space Manipulation Planning by projection sampling with heuristically guided Randomized A*
Manipulation for daily living tasks requires manipulating unknown objects and to address this, I'm developing a Perception framework for Robotic Manipulation/Grasping for objects without a priori knowledge from their surface geometry and then determine the 6D grasp configuration.My objective is to enable the Robot arm to perform autonomous manipulation tasks like Self feeding, assisting in indoor navigation and so on.
I am a Graduate Researcher at the RIVeR Laboratory of WPI working towards the realization of Human in the Loop Cyber Physical System (HiLCPS) that would serve people with disabilities. There are almost 50 million people in the US who have some degree of disability, and more than 6.5 million of them experience problems with self-care.
The aim of this research is to develop a control framework for shared human-robot autonomy for a wheelchair-manipulator system which will allow locked-in individuals, who are unable to interact with the physical world through movement and speech, to perform activities of daily living (ADL). Our current focus is on designing a modular, semi-autonomous robotic wheelchair platform with a 6-DOF robotic arm, controlled through a Body/Brain Computer Interface (BBCI). The key design requirements include safety, dependability, modularity, reliability and fault handling in the system. In our use-cased based design approach, we first identify relevant ADL-scenarios, such as handling objects, self-feeding, semi-autonomous movement in an indoor environment (opening doors, navigation through multi-level buildings, etc.) and then define the system requirements. At this stage of the research, we have developed a high-performance algorithm for simulating a tactile sensor system which will enable development of effective and safe control algorithms for grasping. Moreover, our current work focuses on a LIDAR-based navigation framework which will allow for safe and reliable indoor navigation.
WPI HUMANOID ROBOT LABORATOY (WHRL)
As a Graduate Student Researcher at WHRL, I'm working with a team that works towards footstep planning of the Humanoid Robot ATLAS. Footstep planners generate a sequence of steps for repeating a predefined walking motion of robot, which is used
for navigation of biped humanoids.
We are developing footstep a planning framework for Atlas Robot in cluttered environment by implementing the search based planner A* and its variants like Dynamic A*, Anytime Repair A*and Randomized A*. My primary focus is Randomized A* which is a local deterministic search with a randomized component, thereby trying to speed up the search by solving the whole planning problem with a series of short-range searches, each guided by the heuristic function towards a randomly chosen goal. This research focuses on minimizing the planning time, computational cost and path cost of Humanoid navigation by adaptively switching between the planners based on scene complexity and traversability.
