Half-Day ACC 2023 Workshop
The workshop slides can be downloaded here:
Loading...
Reload document 
Open in new tab Time: 8:30 am – 12:30 pm, May 30, 2023.
Location: American Control Conference (ACC) 2023, San Diego, CA, USA
Room: Sapphire 411B
Workshop registration: https://acc2023.a2c2.org/registration/
Organizers: Guangyi Liu and Nader Motee, Lehigh University
Welcome to our workshop on “Principles of Risk Quantification in Networked Control Systems.” In today’s world, we are facing numerous crises in various sectors due to weaknesses in their network structures. For example, the air traffic congestion, power outages, financial crisis of 2008, Fukushima, H1N1, COVID-19 pandemic, and many more are engineering, economic, and natural networks vulnerable to exogenous stochastic disturbance inputs.
This workshop aims to introduce participants to the mathematical principles of risk analysis in control and dynamical systems. It will provide an understanding of various notions of risk that appear in various real-world applications in control and dynamical systems. The workshop will also present novel methodologies to assess risk in networked control systems.
Our goal is to equip participants with the necessary tools to analyze and manage risks effectively. We will discuss recent advances in risk analysis, which suggest new notions of risk in terms of surrogacy for efficiency and robustness. These new approaches hold promise for a broad range of applications where traditional methods for optimal performance and uncertainty have been subject to criticism. The workshop will feature interactive tutorial sessions and case studies to help participants to understand and apply these concepts in practice. Whether you are an engineer, scientist, or researcher, this workshop will provide valuable insights and tools to analyze and manage risks in networked systems effectively. Join us to learn how to apply the latest mathematical principles of risk analysis in control and dynamical systems to solve real-world problems!
Abstract:
The recent crisis in various sectors show specific vulnerabilities of networks due to weaknesses in their structures, for example, air traffic congestion, power outages, the financial crisis of 2008, and other major disruptions such as Fukushima, H1N1, COVID-19 pandemic. These are all examples of engineering, economic, and natural networks in that the effect of exogenous stochastic disturbance inputs cannot be effectively captured using traditional approaches. Furthermore, recent advances suggest that new notions of risk in terms of surrogate for efficiency and robustness hold promise for a broad range of applications in which the traditional approaches to optimal performance and uncertainty have been subject to criticism. The overarching goal of this workshop is to introduce mathematical principles of risk analysis in control and dynamical systems, provide an understanding of various notions of risk appearing in various real-world applications in control and dynamical systems, and present novel methodologies to assess risk in networked control systems.
Tentative list of speakers, talk/activity titles, and abstracts:
8:35am-9:15am Title: Introduction to Risk Measures
Abstract: We will review some concepts in Probability theory and introduce the Value-at-Risk (VaR) measure, the Conditional Value-at-Risk (CVaR) measure, notions of Cascading Failures, and the class of coherent risk measures.
9:15am-10:00am Title: Risk Quantification in Networked Control System (Part I)
Abstract: We will present a systematic method to quantify various risk measures in control and dynamic systems. The focus of this presentation will be on quantifying risk in the following application areas: Multi-agent Rendezvous.
Abstract: We will present a systematic method to quantify various risk measures in control and dynamic systems. The focus of this presentation will be on quantifying risk in the following application areas: Platoon of Autonomous Vehicles.
11:25am-12:00pm Title: Risk Quantification in Networked Control System (Part III)
Abstract: We will present a systematic method to quantify various risk measures in control and dynamic systems. The focus of this presentation will be on quantifying risk in the following application areas: Synchronous Power Networks.
12:00pm-12:20pm Title: Risk Quantification in Robotic Perception
Abstract: We will present a novel framework to assess the risk of misperception in traffic sign classification tasks in the presence of exogenous noise.
12:20pm-12:30pm Concluding Remarks
Diversity, Equity, and Inclusion (DEI): We will advertise flyers of this workshop in various venues to attract diverse participants.
