Biography
Gadd is a Postdoctoral Research Assistant (PDRA) at the University of Oxford and a Junior Research Fellow at Kellogg College. He works with Professor Paul Newman in the Mobile Robotics Group (MRG), part of the Oxford Robotics Institute (ORI). Gadd originally joined MRG as a student reading for a DPhil in Engineering Science at Keble College. Before arriving in the United Kingdom, he was an undergraduate student at the University of Cape Town, South Africa, where he studied Mechatronics Engineering.
Research Interests
- Robotics,
- Autonomous Vehicles,
- Computer Vision,
- Machine Learning
Research Groups
Current Projects
Sense Assess eXplain (SAX) - acceptably auditable action from autonomous vehicles in formidable environments is only going to be possible with robust sensors and sensing which are nevertheless introspective where unavoidably fallible
Related Academics
Recent Publications
What goes around: leveraging a constant-curvature motion constraint in radar odometry
Aldera R, Gadd M, De Martini D & Newman P (2022), IEEE Robotics and Automation Letters, 7(3), 7865-7872
BibTeX
@article{whatgoesaroundl-2022/6,
title={What goes around: leveraging a constant-curvature motion constraint in radar odometry},
author={Aldera R, Gadd M, De Martini D & Newman P},
journal={IEEE Robotics and Automation Letters},
volume={7},
pages={7865-7872},
publisher={IEEE},
year = "2022"
}
Contrastive learning for unsupervised radar place recognition
Gadd M, De Martini D & Newman P (2022), 2021 20th International Conference on Advanced Robotics (ICAR), 344-349
BibTeX
@inproceedings{contrastivelear-2022/1,
title={Contrastive learning for unsupervised radar place recognition},
author={Gadd M, De Martini D & Newman P},
booktitle={20th International Conference on Advanced Robotics (ICAR 2021)},
pages={344-349},
year = "2022"
}
The Oxford Road Boundaries Dataset
Suleymanov T, Gadd M, De Martini D & Newman P (2022)
BibTeX
@inproceedings{theoxfordroadbo-2022/1,
title={The Oxford Road Boundaries Dataset},
author={Suleymanov T, Gadd M, De Martini D & Newman P},
booktitle={32nd IEEE Intelligent Vehicles Symposium (IV21) -- Workshop on 3D-Deep Learning for Automated Driving (3D-DLAD)},
year = "2022"
}
Fool me once: robust selective segmentation via out-of-distribution detection with contrastive learning
Williams D, Gadd M, De Martini D & Newman P (2021), 2021 IEEE International Conference on Robotics and Automation (ICRA), 9536-9542
BibTeX
@inproceedings{foolmeoncerobus-2021/10,
title={Fool me once: robust selective segmentation via out-of-distribution detection with contrastive learning},
author={Williams D, Gadd M, De Martini D & Newman P},
booktitle={ICRA 2021},
pages={9536-9542},
year = "2021"
}
The hulk: design and development of a weather-proof vehicle for long-term autonomy in outdoor environments
Kyberd S, Attias J, Get P, Murcutt P, Prahacs C et al. (2021), International Conference on Field and Service Robotics (FSR), 16(2021), 101-114
BibTeX
@inproceedings{thehulkdesignan-2021/1,
title={The hulk: design and development of a weather-proof vehicle for long-term autonomy in outdoor environments},
author={Kyberd S, Attias J, Get P, Murcutt P, Prahacs C et al.},
pages={101-114},
year = "2021"
}
RSS-Net: weakly-supervised multi-class semantic segmentation with FMCW radar
Kaul P, De Martini D, Gadd M & Newman P (2021), Proceedings of the 2020 IEEE Intelligent Vehicles Symposium (IV2020)(2020), 431-436
BibTeX
@inproceedings{rssnetweaklysup-2021/1,
title={RSS-Net: weakly-supervised multi-class semantic segmentation with FMCW radar},
author={Kaul P, De Martini D, Gadd M & Newman P},
booktitle={IEEE Intelligent Vehicles Symposium (IV)},
pages={431-436},
year = "2021"
}
Sense-Assess-eXplain (SAX): building trust in autonomous vehicles in challenging real-world driving scenarios
Gadd M, De Martini D, Marchegiani M, Newman P & Kunze L (2021), Proceedings of the 2020 IEEE Intelligent Vehicles Symposium (IV), 150-155
BibTeX
@inproceedings{senseassessexpl-2021/1,
title={Sense-Assess-eXplain (SAX): building trust in autonomous vehicles in challenging real-world driving scenarios},
author={Gadd M, De Martini D, Marchegiani M, Newman P & Kunze L},
booktitle={IEEE Intelligent Vehicles Symposium (IV), Workshop on Ensuring and Validating Safety for Automated Vehicles (EVSAV)},
pages={150-155},
year = "2021"
}
Look Here: Learning Geometrically Consistent Refinement of Inverse-Depth Images for 3D Reconstruction
Sǎftescu A, Gadd M & Newman P (2021), International Journal of Pattern Recognition and Artificial Intelligence
BibTeX
@article{lookherelearnin-2021/1,
title={Look Here: Learning Geometrically Consistent Refinement of Inverse-Depth Images for 3D Reconstruction},
author={Sǎftescu A, Gadd M & Newman P},
journal={International Journal of Pattern Recognition and Artificial Intelligence},
year = "2021"
}
Keep off the grass: permissible driving routes from radar with weak audio supervision
Williams D, De Martini D, Gadd M, Marchegiani M & Newman P (2020), Proceedings of the 2020 IEEE Intelligent Transportation Systems Conference (ITSC), 1-6
BibTeX
@inproceedings{keepoffthegrass-2020/12,
title={Keep off the grass: permissible driving routes from radar with weak audio supervision},
author={Williams D, De Martini D, Gadd M, Marchegiani M & Newman P},
booktitle={IEEE Intelligent Transportation Systems Conference (ITSC)},
pages={1-6},
year = "2020"
}
On the road: route proposal from radar self-supervised by fuzzy LiDAR traversability
Broome M, Gadd M, De Martini D & Newman P (2020), AI, 1(4), 558-585
BibTeX
@article{ontheroadroutep-2020/12,
title={On the road: route proposal from radar self-supervised by fuzzy LiDAR traversability},
author={Broome M, Gadd M, De Martini D & Newman P},
journal={AI},
volume={1},
pages={558-585},
publisher={MDPI},
year = "2020"
}
Google Scholar
Awards and Prizes
2013 FirstRand Laurie Dippenaar Scholarship
