Smart tech provider VIMOC Technologies has released a new SDK for its Landscape Computing API. The SDK aims to make it easier for developers to acquire and use sensor-collected data across the pilot sites of Palo Alto and Los Gatos, California, and Newcastle, Australia.
VIMOC Technologies sees its full potential as being a sensor data enabler that is hardware-agnostic, but in the meantime — while that vision has yet to come to fruition — the team has commenced work with four cities by installing a range of open hardware sensors that can monitor car parking spaces and measure pedestrian flow through public spaces.
“In the IoT world, if you don’t have sensors and deployment, you don’t have value,” says CEO Tarik Hammadou. “So that’s exactly what we are doing — we are using open hardware but trying to move to being hardware-agnostic.” Hammadou says an initial goal is to collect data from 12,000 sensors to enable developers to create meaningful apps.
So far, VIMOC has signed up four cities: Palo Alto, Los Gatos and Redwood City, all in California's Silicon Valley basin, and Newcastle on Australia’s east coast, above Sydney.
Each city is called a "site" within the Landscape Computing API (a query can return all sites covered and define their boundaries via latitude and longitude). Then each site is made up of a series of zones, with each zone having between six and 27 sensors. Each zone has individual sensorIDs for each of the sensors monitoring, for example, an individual car parking space. Each parking sensor can return data on turnover rates, average length of stay, vacancy rates and occupancy counts.
One difficulty is that rates are calculated depending on the length of time the parking space is considered in official operation. For example, if it is a maximum two-hour space, vacancy rates and average length of stay are calculated against a two-hour limit. This may create some confusion in comparing rates when other parking spaces in the same zone have a 24-hour parking limit. There does not appear to be an API call available that would let you know the maximum amount of time a car can be parked or when the parking space is operating as a metered space and when it is classified as free parking. Some improvements also need to be made to more clearly articulate the data being returned. The API documentation is clear in some places where the units of measure are described (for example, that average duration is shown in minutes), but that is not always the case, so developers will need to consult the documentation each time they need to remember what unit of measure is being returned.
API calls can be returned in JSON or XML. API documentation sandbox features allow interactive views of the exact location of each parking sensor that can be shown as a map or returned as a lat/long definition.
The new SDK has been written in React.js and is available online with the API description and via GitHub. The SDK includes sample forms and views. A sample application that shows how to create a car parking app that can guide you in real time to a vacant space in downtown Palo Alto has also been made available.
At present, the API sign-up is not self-serve, and some of the materials are still in PDF format. A PDF tutorial walks potential developers through how the sensor network operates across a city and proposes the value that can be created from using the APIs. An example in the PDF tutorial shows a potential dashboard measuring pedestrian walking activity against temperature, light and humidity measures. Such a tool could be useful in helping cities plan nighttime safety for pedestrians or to manage the climate effects of urban heat islands in densely populated cities.
Enough of a minimum viable product (MVP) is available through the the current API and materials, but developers need to accept that this is still a work in progress. And despite this infancy and MVP nature, VIMOC Technologies' approach to opening up sensor data to developers is one of the most advanced available for anyone considering how to start making use of real-time city sensor data. Hammadou confirms that the VIMOC Technologies team is working on the API and SDK on a weekly basis, so continued progress is expected to happen fairly quickly to provide other sensor data being collected (such as the pedestrian flow counters).
By working with three cities in the U.S. Silicon Valley area that have a high level of technical expertise among the residential working population, Hammadou is hopeful that developers will begin to experiment with the sensor data on offer and that this will inspire developers to start creating applications.
Other sensors being brought on board in the near future aim to support climate adaptation capabilities of cities, including upcoming water management sensors.
While Hammadou says the API and new SDK can help create smart cities applications, for now it is limited to applications that can identify available parking spaces. This in itself could be useful, especially to existing applications that may want to integrate this real-time information into their events or transport route applications.
And there is no doubt that this type of sensor data will be useful for other city planning applications. Developers may not yet be able to build commercially viable applications off the data flow being generated by these sensor data APIs, but those interested in pursuing this area of civic tech may want to familiarize themselves with the landscape computing sensor mesh network approach. Where VIMOC Technologies is heading may enable developers to build real-time interactive applications for city residents and visitors and to create valuable city monitoring dashboards that look at characteristics like walkability, retail space engagement, climate adaptation, civic participation and traffic flow. It is also a fascinating set of data points for any city planners or urban researchers who want to begin thinking through the sort of analysis that smart city sensor data will make available.
Developers can see the full API and SDK documentation and register for access at the VIMOC Technologies website.