Positioning in GPS-denied Environments
Whereas GPS is a de facto positioning solution for outdoor environments, it does not work well indoors or underground. We adopt two distinct approaches to solving the localisation problem in GPS-denied environments: infrastructure-based and infrastructure-free solutions. Our algorithms range from traditional geometry-based algorithms to deep learning approaches, and are applied to localising people, animals, robots and assets. We are particularly interested in positioning systems that combine multiple sensor modalities, including visual, inertial, radio and magnetic data.
DeepTIO: A Deep Thermal−Inertial Odometry with Visual Hallucination
M.R.U. Saputra P.P.B. de Gusmao C.X. Lu Y. Almalioglu S. Rosa C. Chen J. Wahlstrom W. Wang A. Markham and N. Trigoni
In IEEE Robotics and Automation Letters (RAL) + IEEE ICRA. 2020.
Learning Monocular Visual Odometry through Geometry−Aware Curriculum Learning
Muhamad Risqi U. Saputra Pedro P. B. de Gusmao Sen Wang Andrew Markham and Niki Trigoni
In IEEE International Conference on Robotics and Automation (ICRA). 2019.
Visual SLAM and Structure from Motion in Dynamic Environments: A Survey
M. R. U. Saputra; A. Markham; and N. Trigoni
In ACM Computing Surveys (CSUR) 51 (2)‚ 37. 2018.