Hanna Kurniawati
  CENSAM IRG
  Singapore-MIT Alliance for Research and Technology
  Building S16, Level 6, Room 20
  Phone : +65-6516-5592

I am currently a research scientist at the Singapore-MIT Alliance for Research and Technology, working with Nicholas M. Patrikalakis. Previously, I worked with David Hsu as a Ph.D. student and then as a research fellow in the Department of Computer Science, National University of Singapore (NUS). Prior to my graduate studies, I studied computer science in Faculty of Computer Science, University of Indonesia and also worked there for almost a year.

My research interest includes motion planning, planning under uncertainty, robotics, computational geometry applications, machine learning, and randomized algorithm.

My goal is to develop robots that can use their dexterity strategically. This ability requires two major components. First, since many tasks require high dexterity and high dexterity requires many degrees of freedom (DOFs), we need efficient motion planners for robots with many DOFs. Second, we need motion planners that can efficiently generate strategic motion, despite environment uncertainty and dynamics, and errors in the robot's motion and sensing. Uncertainty and system errors are unavoidable and can not be ignored when robots operate in less structured environment, such as marine environment and our kitchen, or perform critical tasks, such as surgery.

During my Ph.D., I have focused on motion planning for robots with many degrees-of-freedom (DOFs). I have developed new probabilistic path planners, utilizing computational geometry methods and machine learning techniques. A planner I have developed (WCO) is able to solve a simulation of bridge inspection problem involving a hyper-redundant robot with 35 DOFs.

After my Ph.D., I expanded my work to motion planning under uncertainty. I have developed new global motion planners that are practical and take into account various sources of uncertainty in a systematic and mathematically principled manner. One of these planners (SARSOP) can be downloaded from here.

I am currently working with an interdisciplinary group of ocean engineers, marine scientists, and mechanical engineers, to develop a mobile sensor network for autonomous monitoring of Singapore marine environment, including to develop an autonomous system to scan and construct 3D models of marine structures. This work enables me to better understand the nature and effects of various sources of uncertainty to robot motion planning. Utilizing this new understanding, I have developed a motion planner under motion, sensing, and environment map uncertainty for marine robots navigating with active sensing mechanism in cluttered marine environment.

Last updated: 29 October 2011