From Autonomy to Cognitive assistance in Emergency Operations
ACE-OPS is an EPSRC centre-to-centre collaboration across three universities - University of Oxford, University of Virginia and Queensland University of Technology. Other partners include the Defence Science and Tech Lab, the National Fire Chiefs Council, Oracle Corporation, Flir Systems AB and BB7 Fire Limited. The main objective of this project is to transform emergency operations globally by building on, and across, cutting-edge research in robotics and cyber physical systems to reduce risk and ultimately increase the number of lives saved. We are determined to address the following key objectives: O1) Address the lack of autonomous systems capable of handling multiple, and often conflicting, operational needs with limited resources; understand trade-offs between localising the team accurately and mapping the environment thoroughly. O2) Provide holistic awareness about the incident; moving away from siloed systems of location, welfare and environment sensing, and design an integrated situation awareness system that provides continually evolving insights for decision making in a dynamically changing context. O3) Explore how and when the output of the situation awareness system should be presented to the response teams; how to best interact with their protocols, presenting the right information at the right time, and providing triggers and suggestions that offer cognitive assistance in complex and often life-threatening situations. O4) Link the autonomy, situation awareness and cognitive assistance capabilities into an integrated system that provides valuable services to emergency teams.
For more information about the project, please see project page.
Selected Publications
-
Illumination−Aware Hallucination−Based Domain Adaptation for Thermal Pedestrian Detection
Qian Xie‚ Ta−Ying Cheng‚ Zhuangzhuang Dai‚ Vu Tran‚ Niki Trigoni and Andrew Markham
In IEEE Transactions on Intelligent Transportation. 2023.
Details about Illumination−Aware Hallucination−Based Domain Adaptation for Thermal Pedestrian Detection | BibTeX data for Illumination−Aware Hallucination−Based Domain Adaptation for Thermal Pedestrian Detection | Download (pdf) of Illumination−Aware Hallucination−Based Domain Adaptation for Thermal Pedestrian Detection
-
mmPoint: Dense Human Point Cloud Generation from mmWave
Qian Xie‚ Qianyi Deng‚ Ta−Ying Cheng‚ Peijun Zhao‚ Amir Patel‚ Niki Trigoni and Andrew Markham
In British Machine Vision Conference (BMVC). 2023.
Details about mmPoint: Dense Human Point Cloud Generation from mmWave | BibTeX data for mmPoint: Dense Human Point Cloud Generation from mmWave | Download (pdf) of mmPoint: Dense Human Point Cloud Generation from mmWave
-
Beyond Fusion: Modality Hallucination−based Multispectral Fusion for Pedestrian Detection
Qian Xie‚ Ta−Ying Cheng‚ Jia−Xing Zhong‚ Kaichen Zhou‚ Andrew Markham and Niki Trigoni
In IEEE/CVF Winter Conference on Applications of Computer Vision (WACV). 2023.
Details about Beyond Fusion: Modality Hallucination−based Multispectral Fusion for Pedestrian Detection | BibTeX data for Beyond Fusion: Modality Hallucination−based Multispectral Fusion for Pedestrian Detection | Download (pdf) of Beyond Fusion: Modality Hallucination−based Multispectral Fusion for Pedestrian Detection