Adversity‐ and Adversary‐Robust Adaptive Positioning Systems with Integrity

For autonomous systems, providing guaranteed performance whilst dealing with both adversity and attack by adversaries, is a largely unsolved problem. One key challenge is that many autonomous system capabilities – such as human‐level scene understanding and action recognition – remain largely elusive. Any attempt to introduce adversity‐ and adversary robustness is often hamstrung by the poor base level performance of these systems, and lack of understanding of intricate mechanistic details.

One key capability pillar of autonomous systems that has reached maturity is positioning, or localization. Consequently, it provides a suitable foundation upon which to investigate these questions. This project seeks to use the positioning domain as a testbed in which to rapidly develop adaptive positioning system techniques that are adaptively robust to both adversity and adversaries. Expected outcomes include both more deployable positioning technologies for defence, and insights that can be adapted to other autonomous system capabilities.

This project is joint between QUT and ANU, and is funded by the Centre for Advanced Defence Research in Robotics and Autonomous Systems, headquarters at the University of Adelaide.

Media and Links