ENAC Seminar Series by Dr S. Xu
Event details
| Date | 26.02.2020 |
| Hour | 09:30 › 10:30 |
| Speaker | Dr Susu Xu |
| Location | |
| Category | Conferences - Seminars |
09:30 – 10:30 – Dr Susu Xu
Research Scientist, Qualcomm Technologies, San Diego, USA
Towards a Self-adaptive Smart City: Collaboratively Integrating Sensing, Learning and Actuation for Monitoring Urban Infrastructure Systems
With increasing populations and demand for high-quality urban services, there is an urgent need to build “self-adaptive” cities which can autonomously adapt their monitoring and management strategies for urban infrastructure systems under constantly changing urban dynamics. The recent rapid development of sensor networks and 5G technologies are enabling large-scale multi-source data and real-time multi-agent control. However, these large-scale and interdependent physical infrastructure systems pose challenges to data-driven monitoring and management strategies. For instance, how to design low-cost paradigms for large-scale and complex infrastructure sensing, how to capture and analyze the physical dynamic interplay between infrastructure systems from noisy and incomplete data, how to timely react to changes of urban dynamics, and more importantly, how to automate the process of sensing, learning and actuation to improve the quality of the urban services.
In this talk, Dr Xu will introduce a framework that collaboratively integrates resource-aware sensing, physics-informed learning and user-incentivizing mechanisms for monitoring large-scale urban infrastructure systems. First, she will talk about her work on embedding prior physical knowledge of infrastructures into adversarial transfer learning algorithms for infrastructure damage diagnosis. This framework enables knowledge transfer across different infrastructures without any labelled data on the target structure. This is especially important when the data is scarce, such as in post-disaster scenarios. Further, she will introduce the integration of indirect sensing methods, including “buildings as sensors” and “vehicles as sensors”, and physics-informed learning for large-scale infrastructure monitoring. Finally, Dr Xu will briefly mention her works on incentivizing vehicle mobilities and human activities to react to the detected changes in urban infrastructure systems, which improves the efficiency, reliability and sustainability of future cities.
Research Scientist, Qualcomm Technologies, San Diego, USA
Towards a Self-adaptive Smart City: Collaboratively Integrating Sensing, Learning and Actuation for Monitoring Urban Infrastructure Systems
With increasing populations and demand for high-quality urban services, there is an urgent need to build “self-adaptive” cities which can autonomously adapt their monitoring and management strategies for urban infrastructure systems under constantly changing urban dynamics. The recent rapid development of sensor networks and 5G technologies are enabling large-scale multi-source data and real-time multi-agent control. However, these large-scale and interdependent physical infrastructure systems pose challenges to data-driven monitoring and management strategies. For instance, how to design low-cost paradigms for large-scale and complex infrastructure sensing, how to capture and analyze the physical dynamic interplay between infrastructure systems from noisy and incomplete data, how to timely react to changes of urban dynamics, and more importantly, how to automate the process of sensing, learning and actuation to improve the quality of the urban services.
In this talk, Dr Xu will introduce a framework that collaboratively integrates resource-aware sensing, physics-informed learning and user-incentivizing mechanisms for monitoring large-scale urban infrastructure systems. First, she will talk about her work on embedding prior physical knowledge of infrastructures into adversarial transfer learning algorithms for infrastructure damage diagnosis. This framework enables knowledge transfer across different infrastructures without any labelled data on the target structure. This is especially important when the data is scarce, such as in post-disaster scenarios. Further, she will introduce the integration of indirect sensing methods, including “buildings as sensors” and “vehicles as sensors”, and physics-informed learning for large-scale infrastructure monitoring. Finally, Dr Xu will briefly mention her works on incentivizing vehicle mobilities and human activities to react to the detected changes in urban infrastructure systems, which improves the efficiency, reliability and sustainability of future cities.
Practical information
- General public
- Free
Organizer
- ENAC
Contact
- Cristina Perez