OpenSensingCity aims at facilitating uses and usage of open data, whether static or dynamic, in smart urban areas. The results of the project will help application developers that exploit open data to better manage, search, process open data, including real time data.
The main goal of the project is decomposed into the following scientific objectives: <br />O1. Develop ontologies for smart cities. Having well formalised ontologies will allow the description of sensors, data and services that are deployed, produced or used in the context of smart cities. This will improve the understandability of the available data and thereby, their reuse. <br />O2. Formalism for querying and aggregating both static and dynamic data taking into account spatial and temporal dimensions. The formalism should allow to define virtual sensors (continuous queries) feeding the platform with new open data from the sensed ones. <br />O3. Define strategies and technology appropriations that support and build the foundation of an open data ecosystem. <br />In complement to these scientific objectives, the project aims at deploying an open data platform and demonstrate its utility on an intelligent transport application. <br />To achieve these objectives, OpenSensingCity addresses two scientific and technological challenges (C1 & C2) and one societal challenge (C3). <br />C1. Providing the necessary and sufficient vocabularies. Regarding O1, the difficulty lies in defining what information is relevant to the data and metadata for sensors as well as to smart cities (buildings, roads, traffic, energy management, etc.) and making this available in an interoperable way. In general, ontology development is often challenging, and in this case it is further complexified as the domains of knowledge are very heterogeneous and the coverage has to be defined. <br />C2. Combining ontological, spatial and temporal reasoning. Objective O2 poses a theoretical challenge in being able to easily select and aggregate data available within a spatial region, a time frame and/or a type of sensor. <br />C3. Defining socio-technical appropriation through analyses of users’ point of view, in order to understand their needs and the conditions of feasibility. This analysis will help define the storage and players strategies.
The project takes the assumption that open data requires the use of open standards, especially W3C Semantic Web standards and accompanying Linked Open Data best practices. Additionally, to achieve the goal mentioned above, the project will bring together the following scientific expertise that define the methods and technologies used:
E1. Representation, reasoning and querying contextualised knowledge: in smart cities, data and knowledge extracted from sensors have to be qualified according to various dimensions of context such as a time frame,
spatial area, provenance, reliability.
E2. Ontology engineering: the design of ontologies will allow a better management of the data and the knowledge extracted from the data.
E3. Usages and practices from a communicational point of view to ensure that the developed solutions are not only answers to scientific challenges but will also be acceptable, usable and useful to the concerned actors.
Since the domain of smart cities is currently very active, the project aims at delivering focused results centred on lowering the costs of accessing and reusing data from sensors in a smart city environment. The core results are two important theoretical contribution from each academic partner:
R1. Smart city query language for the access and production of sensor data. Most notably, this language will be used to define virtual sensors that consist in delivering continuously the result of a query to the smart city data.
R2. Analysis of practices and expectations at two levels: the end user level, often confused with the figure of the citizen; and the intermediate user level, commonly ignored, who is the producer of contents and/or services. In addition to these research results, the project will produce concrete results with the help of its industrial partners.
R3. Software platform for the deployment of semantic sensors (real and virtual ones) in a smart city. This platform is targeted to the application developers and will contribute to the dissemination of the results. This platform will support a prototype Intelligent Transport System (ITS) dedicated to the management of parking spots. This transportation resource is characterised by dynamic spatio-temporal data as well as static data such as its status (e.g., private, reserved for disabled, public). This application exemplifies two functionalities of the platform: 1) The simplified access to dynamic open data; 2) The integration of new dynamic open data (coming from social networks in this application).
R4. Ontologies for smart city. By meeting challenge C1, the project will design computerised ontologies using Web standards and also provide human readable terminologies for the stakeholders.
Opening sensor data brings similar advantages as opening other types of data: the burden of implementing services on top of the data is pushed towards third parties, and the data publisher makes a decisive move towards transparency. However, the benefits only become significant when this project gives to the developers a set of tools that allow them to easily access and manipulate the data.
The project is also helping all the stakeholders of the data ecosystem to shift to an open data governance model, even for stream data. Few projects propose both a technical solution to a hard scientific problem, and guiding analysis of best actor strategies. This duality is likely to ensure better suitability of the output of the project to our current society.
The project will have a direct impact on Grand Lyon as they will be the first beneficiaries of the results of the project. The analysis of uses and usages, the studies of expectations and practices will benefit the digital strategy of Grand Lyon, who will be able to make informed decisions regarding their open data platform where currently, little is known of the uses so far. Ultimately, agglomerations will be more likely to develop services in line with citizen needs. More generally, OpenSensingCity has the potential to improve city governance and increase city transparency.
Also, software or service developers, such as start-ups can benefit from the expectation studies to fine tune their business strategies. In the long term, we can envision that this will turn into a virtuous circle that, combined with the many open and smart city initiatives, will develop a successful service eco-system.
The facilitation of exploitation of real time data should enable data scientists and data journalists to obtain more up to date information for more reactive data analysis.
The project will also advance research in stream data processing and management that will encourage new value-added digital services in the open data ecosystem.
Larroche, V., Vila, M. (2015). « Urban Data et stratégies dans le secteur des services : Le cas de la métropole lyonnaise ». In : E. Broudoux, E., Chartron Gh. Big data - Open data: Quelles valeurs? Quels enjeux? Louvain-la-Neuve : De Boeck Supérieur, p. 183-197. En ligne : hal.archives-ouvertes.fr/hal-01184089.
Paquienséguy, F. (2016). « Les portails métropolitains Open Data : à qui profite le chiffre ? » CIST2016 - En quête de territoire(s) ?, Mar 2016, Grenoble, France. Proceedings du 3e colloque international du CIST, pp.351-356. En ligne : h/ttps://hal.archives-ouvertes.fr/hal-01353651/
Smart City Artifacts. Noorani Bakerally, Olivier Boissier, Antoine Zimmermann. In Proceedings of the Extended Semantic Web Conference, ESWC 2016, Anissaras, Greece. Springer 2016.
In the context of smart cities, the deployment of multiple sensors provides access to numerous data streams in real time. The open publication of sensor data brings innovation opportunities by combining the usual benefits of open data with those of real time updates. Indeed, open data ensures transparency and, in principle, allow anybody to develop services that were never envisioned by data providers. Real time updates allow one to consider the development of new services beyond the traditional use of open data, e.g., for logging evolution and conducting a posteriori analysis. We can therefore contemplate the creation of an ecosystem of smart open urban services. While publishing sensor data via an open data platform, such as Grand Lyon's "Smart Data" platform, is a first step towards our goal, it is now necessary to propose solutions to facilitate uses and usages of real time open data. As a matter of fact, these data are practically difficult to understand, find and eventually exploit. This is even more true when data is coming from raw stream of sensor data because constraints on processing and communication capabilities impose minimising transmitted information.
Consequently and in order to enable the development of a smart open city service ecosystem, we want to provide (1) technological solutions to help leverage open sensor data for city application developers, and (2) guidance to the actors of the open sensor data environment by analysing the stakeholder strategies, defining usage scenarios and terminologies. To this aim, we combine a social analysis of actor expectations, requirements and practices, with a technological and theoretical expertise in online data and knowledge processing and engineering. The social component of our proposal will ensure a better understanding of expectations and needs of various categories of real time open data users. The technological component is based on existing Semantic Web technologies as well as stream processing techniques. This will expectedly result in enriching and publishing stream data to an open platform, taking into account the new paradigm of linked data, stream data and reasoning. This major shift will be achieved by building new ontologies related to smart cities when necessary as well as a formalism for querying and combining streams in such a context. Additionally, by combing our work in the social and technological domains, we will define search and browsing capabilities that we will implement according to the identified expectations and needs. Finally, when those tools will be effectively developed, we will show their utility by providing a demo application that uses them and further study the usage scenarios. Our planned application will offer an intelligent transportation system that helps vehicle drivers to better find parking spots, with possible testing in the city of Lyon.
Monsieur Antoine Zimmermann (ARMINES)
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
ARMINES FAYOL ARMINES
ELICO Equipe de recherche de Lyon en Sciences de l'Information et de la Communication (Elico)
Help of the ANR 601,204 euros
Beginning and duration of the scientific project: September 2014 - 36 Months