Collaborative Research: CISE-MSI: RCBP-RF: CPS: Socially Informed Traffic Signal Control for Improving Near Roadway Air Quality

随着城市化的快速发展，空气污染已成为世界各地城市紧迫的环境和社会问题，特别是由于道路附近车辆直接排放的污染物。这一问题对各种弱势群体的健康和福祉构成重大威胁，包括儿童、老年人和已有疾病的个人，他们由于靠近主要道路和交通繁忙地区而面临更大的风险。因此，监测空气质量和管制靠近道路的交通是至关重要的，以有效地解决污染问题，促进受影响人群的福利。管理交通和缓解这些挑战的一个潜在解决方案是实施交通信号控制系统，它在减少拥堵，减少车辆排放，最终提高附近居民和通勤者的整体生活质量方面具有广阔的前景。受这一实际情况的启发，该项目的主要目标是开发一种创新的交通信号控制方法，以有效减轻道路附近空气污染的不利影响。该项目旨在通过将创新的基于机器学习的传感和控制技术与社会需求规范相结合，弥合网络物理系统与社会意识之间的关键差距。主要重点是为交通信号控制系统开发一个集成的、社会知情的框架，其中包括三个主要任务，包括：(1)开发一个低成本的移动空气质量传感系统，用于道路附近的空气质量传感和交通传感的时空图扩散学习模型；(2)设计一个基于强化学习的控制模型，结合社会需求规范和交通辅助监督；(3)实现基于真实数据和交通与空气质量相关性的模拟工具。该项目的成功具有重要的社会和技术意义，因为它有可能改善道路附近的空气质量，改善附近居民和旅客的生活质量。此外，全面的能力建设、教育和外联活动将向在计算机领域代表性不足的少数群体提供支助，并促进研究技能的发展和数据密集型领域的未来职业。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

With the rapid pace of urbanization, air pollution has emerged as a pressing environmental and societal concern in cities worldwide, particularly due to pollutants directly emitted from vehicles near roadways. This issue poses a significant threat to the health and well-being of various vulnerable groups, including children, the elderly, and individuals with pre-existing conditions, who face heightened risks due to their proximity to major roadways and high-traffic areas. Therefore, it is crucial to monitor air quality and regulate traffic in close proximity to roadways to effectively address pollution and promote the welfare of these affected individuals. One potential solution to manage traffic and alleviate these challenges is the implementation of traffic signal control systems, which offer promising prospects in reducing congestion, minimizing vehicle emissions, and ultimately enhancing the overall quality of life for residents and commuters in the vicinity. Inspired by this practical scenario, the primary objective of this project is to develop an innovative traffic signal control approach that can effectively mitigate the adverse impacts of air pollution near roadways. This project aims to bridge the critical gap between cyber-physical systems and social awareness by integrating innovative machine-learning-based sensing and control techniques with social requirement specifications. The primary focus is on developing an integrated, socially informed framework for traffic signal control systems, which encompasses three main tasks, including (1) developing a low-cost mobile air quality sensing system for near-road air quality sensing and a spatial-temporal graph diffusion learning model for traffic sensing, (2) designing a reinforcement learning-based control model incorporating social requirement specification and traffic-aided supervision, and (3) implementing a simulation tool based on real-world data and the correlation between traffic and air quality. The success of this project holds significant social and technological implications, as it has the potential to enhance air quality near roadways and improve the quality of life for nearby residents and travelers. Furthermore, comprehensive capacity-building, education, and outreach activities will provide support to underrepresented minorities in computing and foster the development of research skills and future careers in data intensive fields.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.