Find Connor Malone on
Bachelor of Engineering (Mechtronics)(Hons), Doctor of Philosophy (PhD)(Robotics)
Connor is a Research Fellow who recently received his PhD titled “Beyond Place Matching for Visual Place Recognition on Robots and Autonomous Vehicles”. He has a background in Mechatronics Engineering, with his studies at the Queensland University of Technology (QUT) focusing on robotics and computer vision. His studies throughout his Bachelor and Doctorate focused on traditional, machine learning, and deep neural network approaches to ground plane estimation, semantic scene understanding, and localization. Primarily, his post-graduate and post-doctorate research have been based in autonomous vehicles and Visual Place Recognition (VPR). So far, Connor has been fortunate to work on projects for Amazon, Ford, and the Centre for Advanced Defence Research, investigating how to make use of semantically similar images to improve scene understanding in challenging conditions (night, rain, snow, etc.); automatically choose the best VPR algorithms for any given environment; augment VPR methods using spatial priors; and improve the resilience of VPR methods’ to adversity and adversarial attacks. Connor is now pursuing further research in this direction as well as investigating how to improve the safety of autonomous vehicles in the Australian operating domain, specifically focusing on the detection and analysis of water on roads.
Research Interests:
- Computer Vision
- Deep Learning
- Autonomous Vehicles – Localization and Scene Understanding
- Deployable Technologies
- Improving the robustness of feature-based techniques in challenging conditions (night, rain, snow, etc.)
Publications:
C. Malone, S. Garg, M. Xu, T. Peynot and M. Milford. “Improving Road Segmentation in Challenging Domains Using Similar Place Priors,” IEEE Robotics and Automation Letters, 2022.
C. Malone, S. Hausler, T. Fischer and M. Milford. “Boosting Performance of a Baseline Visual Place Recognition Technique by Predicting the Maximally Complementary Technique,” IEEE International Conference on Robotics and Automation, 2023.
C. Malone, A. Vora, T. Peynot and M. Milford, “Dynamically Modulating Visual Place Recognition Sequence Length For Minimum Acceptable Performance Scenarios,” IEEE International Conference on Intelligent Robots and Systems, 2024.
C. Malone, S. Garg, T. Peynot and M. Milford. “Improving Semantic Segmentation with Calibrated Whole Image and Patch-Based Similar Place Priors,” Australasian Conference on Robotics and Automation, 2021.
O. Claxton, C. Malone, H. Carson, J. Ford, G. Bolton, I. Shames, and M. Milford}. “Improving Visual Place Recognition Based Robot Navigation Through Verification of Localization Estimates,” IEEE Robotics and Automation Letters, 2024.
Projects: