Timothy L. Molloy

University of Melbourne, Australian National University, Australian Centre for Robotic Vision

Papers

Total Citations

H-Index

About

Timothy L. Molloy is a roboticist whose research lies at the intersection of perception, planning, and control under uncertainty. His work addresses fundamental challenges in enabling autonomous systems to operate reliably in complex, dynamic environments. A key contribution is in visual place recognition (VPR), where he developed an intelligent reference curation method using Bayesian selective fusion to help robots robustly recognize locations despite drastic changes in lighting, weather, or season—a problem central to long-term autonomy. In motion planning, Molloy tackled the minimum-time escape problem for a Dubins car, deriving optimal paths for a robot with constrained turning to escape a circular region, with direct applications in marine, aerial, and ground robotics. More recently, he has advanced perception for high-speed robotics with the Asynchronous Event Multi-Object Tracking (AEMOT) algorithm, which leverages event cameras’ microsecond-level temporal resolution to detect and track multiple objects in highly dynamic scenes. With his most cited work accumulating over a dozen citations, Molloy’s research is increasingly recognized for its practical impact on real-world robotic systems, bridging theoretical optimal control with cutting-edge sensor-driven perception.