Stochastic Control for Large-Scale Safety-Critical Systems

大规模安全关键系统的随机控制

• 批准号: RGPIN-2020-04892
• 负责人: Kamgarpour, Maryam
• 金额: $ 3.21万
• 依托单位: University of British Columbia
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2020
• 资助国家: 加拿大
• 起止时间: 2020-01-01 至 2021-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=717446
• 关键词: Stochastic; Control; Large; Scale; Safety

We began using robots to help us with simple and repetitive tasks. Now, we are introducing automation in ever increasing safety-critical and complex tasks: intelligent transportation networks, smart power grid, robotics search and rescue, personalized medicine. The goal of my research is to ensure these autonomous systems make our societies safer and improve our quality of life. There are two main control challenges in achieving this goal: 1) Autonomous systems needs to complete complex tasks in uncertain changing environments: consider, for instance, self-driving cars or search and rescue robots. 2) Interacting autonomous systems in a network need to coordinate their decision making: consider, for instance, intelligent transportation networks and the power grid. To address the above challenges, I develop provably safe and optimal control algorithms for large-scale and interconnected systems with uncertainty. In my approach, I combine physics based models and classical control with advanced optimization and learning to design control laws that learn from past data efficiently and adapt to changes in real-time. In the past, I have worked with air transportation researchers and NASA Aeronautics to improve safety and efficiency of the air traffic management, and with power system researchers to ensure safe and optimal integration of renewable energy into the power grid. As an incoming faculty member to Canada, I will additionally explore applications that are newer to me but are relevant to Canada. These include safe control algorithms for autonomous cars and safe decision making algorithms to assist humans in controlling wildfires. I will train graduate and undergraduate students on latest advances in control, optimization and learning theory. Having in mind societally relevant applications as the research motivation, my students will develop theory with practical use cases as well as an interdisciplinary problem solving skill leveraging advances in control, mathematics, computer science and robotics. I have been supporting women to pursue their talents in engineering through volunteer work with women groups at my home university and in the community. I will work with UBC Equity & Inclusion office to actively encourage talented women to join my research group and to support them in becoming leaders in academia, industry or government. The scientific impact of my research includes shedding light on fundamental understanding of challenges of decision making in uncertain and dynamic environments and an interdisciplinary solution approach to address these challenges. The potential practical impacts of my research include improving safety and efficiency of the autonomous systems and consequently, assisting humans in safety-critical complex tasks. My research has also the potential to lead to startup companies and collaborations with industry and government.

我们开始使用机器人来帮助我们完成简单和重复的任务。现在，我们正在将自动化引入到越来越多的安全关键和复杂的任务中：智能交通网络、智能电网、机器人搜索和救援、个性化医疗。我研究的目标是确保这些自主系统使我们的社会更安全，提高我们的生活质量。实现这一目标面临两个主要的控制挑战：