Join us to participate in the upcoming 2019 City:One Challenge. 

Cumulous Cloudlet System

Mobile datacenter cloudlet integration with buses to improve commuter services and improve data collection around Pittsburgh.

Photo of Chase Klingensmith
1 0

Written by

As cities become smarter, they are going to need a communication platform to enable their evolution. A cloudlet, or miniature mobile datacenter that like a cloud, can provide powerful computing capabilities without falling victim to latency issues that might limit the scope of useful applications for such a device. By using cloudlets, The City of Pittsburgh can create new transit functionality as well as a communication platform between public transit and infrastructure. This opens up a new opportunity to expand existing services while introducing the opportunity for new applications that enhance the public sphere and make public transportation a streamlined and more attractive option for commuters and new visitors to the city alike. A network of cloudlets employed throughout a public transit system would allow for not just real-time location data that would give public transit users an improved ability to track their transportation but also allow the bus to serve as a mobile sensor, collecting traffic and environmental data throughout the city in real-time for planners. A cloudlet would also be able to process data from a mobile payment platform, creating a seamless streamlined experience people consider as a cost-effective and timely alternative to driving via car.

Cloudlets could help create a framework to address the needs of a public transit system while providing useful data-collection in real-time without additional effort. In his book Walkable City, Jeff Speck emphasizes the growth of multi-modality in successful public transit systems and as a direction that inter-city travel is going as a whole. However, in order for people to have confidence in public transportation, a city must provide responsive location data. While the PAT tracker that is currently in place attempts this, it does not always work thanks to latency issues and a lack of historical data to provide context for a normal wait on a route during a specific time. Since a cloudlet would have enhanced tracking data it would not only be able to provide this information but potentially an idea of how long a bus would take to reach that spot at that point in the day based on previously recorded trips.

Cloudlets would also extend what is already in place by enabling buses to become a mobile traffic sensor that monitors congestion, similar to Traffic21’s project to embed sensors in traffic lights throughout Pittsburgh’s Oakland neighborhood to monitor congestion and collect pollution or environmental stimuli, among other quality of life factors that would be crucial for city planning. If one bus along a route also has the pothole detecting app developed by student Kevin Christensen outfitted with it, a city can address them before they become potential liabilities while simultaneously not expending labor to places it isn’t needed. A cloudlet would effectively extend mobility options by creating new options for a city when addressing mobility since a bus is no longer just a vehicle but a mobile real-time urban environment monitor.

Cloudlets could also provide opportunity for development of a mobile payment platform similar to the Port Authority’s GoMobilePGH smart parking system that allows for personal parking management via a smartphone. An application with this could allow a smart phone to function as a bus pass the same way a smartphone functions of a wallet and automatically pays for shoppers at a physical Amazon store. If maps also list locations where FreeRide kiosks are located then this functionality would be enhanced that much further as citizens would find opportunities to mesh the modes of transportation they take in their day-to-day lives. Cloudlets would also provide an opportunity for Wi-Fi for public transit passengers, which would make public transit that much more welcoming for someone who would otherwise be gridlocked in traffic when they could easily be streaming on their way to work. In short, cloudlets would allow for natural extensions of things that already exist in ways that make busing a much more attractive option for passenger and city planner alike.

 In changing the paradigm for what is capable using a public transit system, cloudlets create a new digital infrastructure for not just the aforementioned data collection but also the opening of new services as other devices and businesses develop ways for their technology to communicate. Perhaps a transit cloudlet could house an app that interacts with a coffee shop’s cloudlet at a certain distance, allowing a commuter to place an order they know they would like on their way into work so that it is made and paid for as the commuter is leaving their bus. This would allow for time savings for both sides of the transaction, translating in a better user experience for the customer and improved efficiency for the business. Cloudlets would create a platform for such an application.

Describe who will use your solution (1,000 characters)

This solution would be useful for both daily commuters and visitors to the city. By enhancing the location data, payment processing and overall functionality daily users would see their overall user experience improve. Having a reliable transportation system where riders don’t have to pay with exact change in cash would make the ride enticing to new riders by removing the ease of use hurdles currently in place.

Describe your solution's stage of development

  • Initial Design - you are still exploring the idea and have not tested it with users

Tell us about your team or organization (500 characters) Carnegie Mellon's Living Edge Lab is led by Professor Mahadev Satyanarayanan and focuses on Mobile Edge Computing. Professor Satya leads the Open Edge Computing Initiative, which unites a diverse group of companies, researchers and academic programs in the promotion and development of Mobile Edge Computing projects.

Size of your team or organization

  • 2-10

Funding Request

  • $100,000

Describe how you would pilot your idea (1000 characters)

We would like to outfit two buses along routes in either the 61 or 71 bus corridors with cloudlets for a one month-long pilot where a cloudlet-equipped bus can operate as it normally would all the while collecting traffic and environmental data, such as pothole presence. In addition, a few individuals would be given access to a pay system as it is developed and refined. After its initial pilot, Cumulous could scale up to other routes in the area and integrate with the Port Authority system to improve functionality. Since data collection would be very useful for city-planning purposes cloudlets could potentially be an expenditure that would actually cut costs over time since planners could more efficiently use labor. Clients for this wouldn't be individuals so much as municipalities. Integration with a pay system, for instance, would pay for itself with a nominal fee, not unlike the minor surcharge included with the Pittsburgh PAT parking application.

Describe how you would measure the success of your pilot (1000 characters)

The traffic data and historical data collected from daily operation of the cloudlets will provide a clear indication of whether or not the pilot has been successful. Since the implementation will not have been in place yet, the data collected at the beginning of the pilot would serve as the control data set, with rider satisfaction measured via surveys throughout the pilot. Overall efficiency would be measured with the historical daily trip data as it is collected, while overall impact could be measured by traffic data collected by the buses.

1 comment

Join the conversation:

Photo of Diana Avart

Hi Chase,

Thanks so much for submitting this idea! It sounds like the Cloudlets would take advantage of already existing infrastructure to make information about public transit more accurate and easy to use. I also love the idea of using buses, which travel around the city all day, to collect transportation data. I am surprised that they don't do this already since it seems so logical!

- Diana, Facilitator