Despite the rise of “autonomous” cars and drones, a fully autonomous robot that can operate reliably outside a carefully structured factory floor is extremely rare. The main reason is uncertainty: An autonomous robot must make good decisions despite various types of errors and disturbances affecting its actuators and sensors, and despite the lack of information and understanding about itself and its environment. Unfortunately, robotics technology for making good decisions in the presence of uncertainty remains an open problem. The problem itself is not new. In fact, mathematically principled concepts --called Partially Observable Markov Decision Processes (POMDPs) and their extensions-- have been developed since the '60s, exactly to address the aforementioned problem. However, such concepts are notorious for its computational complexity, that they have been considered impractical and often abandoned at the expense of reliability and robustness. In this talk, I will present some of our effort in addressing the computational complexity issues of POMDPs, and demonstrate that this decision making concept has now become practical (to some extent) for solving various problems in robotics. Towards the end, I will also present an application of this new capability in decision making to model building.
Dr Hanna Kurniawati is a lecturer at the School of Information Technology and Electrical Engineering, the University of Queensland (UQ). Her current research focuses on algorithms to enable decision theory become practical software tools in robotics. Along with colleagues and students, she won a best paper award at the International Conference on Automated Planning and Scheduling (ICAPS) 2015 and was a finalist for the best paper award at the IEEE International Conference on Robotics and Automation (ICRA) 2015. Before joining UQ, she was a Research Scientist at the Singapore-MIT Alliance for Research and Technology, MIT. She received a Ph.D. and BSc. in Computer Science from National University of Singapore and the University of Indonesia, respectively.