When: 2nd of April, 1:00pm AEDT
Where: This seminar will be partially presented at the ACFR seminar area, J04 lvl 2 (Rose St Building) and partially online via Zoom. RSVP
Speaker: Dr. Jennifer Wakulicz
Title: Reduced latent belief spaces for active perception in robotics
Abstract:
Robotic systems are increasingly expected to operate in challenging and unstructured environments: suppressing bushfires, conducting search-and-rescue missions, and monitoring fragile ecosystems such as coral reefs. In these domains, failure is costly and conditions are uncertain. To function safely and effectively, robots must make rapid, autonomous decisions under uncertainty, requiring the ability to interrogate their environment for important information relevant to their task. Developing this capability is the concern of active perception research. A common approach is to design sensing policies that minimise the uncertainty in a robot’s belief, or probabilistic estimate, about a state. Planning in the space of all possible beliefs rather than physical space is then a prudent approach to active perception, as it places uncertainty minimisation at the heart of planning strategies. However, due to the size of belief spaces, it is generally infeasible to find optimal plans within the real-time requirements of robotics. In this talk I introduce a new conceptual approach to belief space planning: reduced latent belief space planning. Here, a partially observable latent variable is introduced that satisfies two properties: first, that the original state of interest can be inferred from it, and second, that its belief space is simpler than original. Then, planning in this reduced belief space to improve latent variable estimation is an efficient proxy for estimating the original state. We will talk through a suite of reduced latent belief spaces proposed for various active perception tasks in robotics applications and extensions of these ideas to future work.
Bio:
Jen is a Postdoctoral Research Fellow at the University of Sydney, working with CIs Dr. Donald Dansereau, Prof Stefan Williams and A/Prof Thierry Peynot (QUT) on the Introspective Robotics ARC Discovery project. She received the B.Sc. degree in Mathematics and Physics from the University of Sydney in 2018 and the Ph.D. degree in robotics from the University of Technology Sydney in 2025. Her research interests are in probabilistic modelling, information theory and planning, with the objective of developing robust robotic decision-making frameworks capable of operating reliably in unstructured environments.