Sensing Light

This work is subdivided into three academic papers that together form a coherent exploration of the phenomena of intelligent street lights and their potential applications as a new type of digital urban infrastructure. In the first paper, I review existing cases of cities that are digitizing their public lighting infrastructure. I analyze their various approaches to smart lighting and then propose a framework by which we can maximize their potential uses. For the second paper, I perform an urban demonstration that pairs street lights with a prototype intelligent, networked digital imaging and computer vision platform, in order to monitor the utilization of curbside space, currently utilized for parking in cities, which serves as an example of how to develop interoperability between different urban infrastructure systems. For the third paper, I investigate the policy dimensions of implementing such a system, including the concerns raised by industry leaders and city officials, as street lights become multi-functional sources of urban data, and the dilemma this may pose for existing institutional arrangements and stakeholder's networks. Seeking to maximize social benefits I conclude by proposing a series of recommendations aimed at hybridizing functions of public lighting and real-time sensing of the built environment in cities, for the creation of a range of new urban experiences and civic benefits across a variety of use cases for cities.

https://dspace.mit.edu/handle/1721.1/127622

Urban infrastructures have traditionally been mono-functional: water, sewage, and electricity are notable examples. Embedded with digital technologies, urban infrastructures have the potential to communicate with one another and become multi-functional platforms that integrate data gathering and actuation cycles. In this paper, we focus on public lighting infrastructures. Despite the technological development of lights, including LED technology, streetlights have been primarily treated as a mono-functional infrastructure. Based on case studies, we discuss the potential of reimagining streetlight infrastructure, and advance some initial proposals that focus on sensing and actuation cycles, which could transform this pervasive infrastructure into a digital urban platform.

https://doi.org/10.1080/10630732.2017.1285084

Parking spaces are costly to build, parking payments are difficult to enforce, and drivers waste an excessive amount of time searching for empty lots. Accurate quantification would inform developers and municipalities in space allocation and design, while real-time measurements would provide drivers and parking enforcement with information that saves time and resources. In this paper, we propose an accurate and real-time video system for future Internet of Things (IoT) and smart cities applications. Using recent developments in deep convolutional neural networks (DCNNs) and a novel vehicle tracking filter, we combine information across multiple image frames in a video sequence to remove noise introduced by occlusions and detection failures. We demonstrate that our system achieves higher accuracy than pure image-based instance segmentation, and is comparable in performance to industry benchmark systems that utilize more expensive sensors such as radar. Furthermore, our system shows significant potential in its scalability to a city-wide scale and also in the richness of its output that goes beyond traditional binary occupancy statistics.

https://doi.org/10.48550/arXiv.1902.07401