MAster assignment
Interoperability of Smart Drone Systems
Type: Master CS
Period: Start date: as soon as possible
Student: Unassigned
If you are interested please contact:
Background:
In Smart Systems, machines and devices are connected and generate data during the execution of the processes, so the identified insights from these data can be used to understand what is actually happening in the organization and processes can be optimized faster. This creates new opportunities in the whole business value chain, once lead times can be shortened and work can be performed more efficiently. These integrated networks of automation devices (sensors and actuators), services, and enterprise systems bring interoperability challenges for the industry ecosystem. Interoperability is the “ability of two or more systems or components to exchange information and to use the information that has been exchanged” (IEEE, 1990). Therefore, interoperability defines the way of interconnection between sensors, devices, manufacturing systems, and people, including exchange of products and materials among facilities. In particular, semantic interoperability is the most challenging because it is about the “interpretation of shared data in an unambiguously way, ensuring that the understanding of the information is the same for senders and receivers” (Heiler, 1995). Establishing automatic semantic interoperability for seamless systems’ integration is an arduous task.
This assignment focuses on smart drone systems and represents a new step based on previous works on Live Mission Recording (LMR) with Consumer/Starter Drones. Within our lab (IoTCyberLab) we currently have available a Drone DJI Mini 2 kit, which has API support (mobile SDK) for developing smart applications, and we intend to acquire another consumer drone for research purposes. The first research cycle addressed the implementation of LMR with a consumer drone through a Bachelor research project (https://essay.utwente.nl/95814/).
The second iteration, done through a Bachelor design project, was inspired by a use case in the context of our collaboration with TNO (Mobility & Built Environment unit) and Rijkswaterstaat on inspection of damages in buildings and infrastructures. This second iteration addressed the design cycle of improving the LRM solution (specification and coding) and including an inspection capability on identifying concrete cracks in buildings, performed by configurable Image Processing algorithms, such as the ones based on Machine Learning approaches (<<link to document>>). The Figure below illustrates the solution architecture, and the code is available in our GitLab. A running instance of the solution is available for tests ('drone_username' and 'drone_password'): https://droneinc.stoplight.io/docs/dronelivemissionrecording/ibkorj5tmvfm6-drone-lmr-api
- Dashboard (Amplify): https://gitlab.utwente.nl/sidlab/DroneDashboard
- Backed + Database: https://gitlab.utwente.nl/sidlab/livemissionrecordingbackend
- Image Recognition: https://gitlab.utwente.nl/sidlab/DesignProjectImageProcessing2
- Application (Android): https://gitlab.utwente.nl/sidlab/dp-android-app
Master Project assignment:
In the third iteration of this research, we would like to focus on the investigation about technical and semantic interoperability of drone systems and autonomous inspection. For the research topics, we identified the following knowledge questions that require further research:
- What is the state of art on drone system (UAV) interoperability?
- Which are the main standards for data exchange?
- Are there ontologies and/or proprietary data models (e.g., API interfaces)? - What is the state of art on autonomous inspection?
- Which are the main standards for data exchange?
- Are there ontologies and/or proprietary data models (e.g., API interfaces)?
- Are there specific use cases for Buildings and Infrastructure?
As relevant topics for solution directions:
- For smart drone systems, how relevant is the Smart Applications REFerence (SAREF) ontology (standardized by ETSI)? How about Drone and other UAV ontologies?
- For autonomous inspection, how relevant are ontologies like: Damage Ontology, Bridge Ontology, Rijkswaterstaat Ontology, Kadaster Ontology?
- Which damage properties can be detected by drones, e.g., 'Kameleonhuid' brengt onzichtbare constructieve schade in beeld (cobouw.nl) ?
- How relevant are semantic-based APIs that use technologies such as JSON-LD?
- Which kind of non functional requirements , e.g., performance, scalability and security are required by drone systems?
- What is the role of IoT platforms for smart drone systems and how to address their interoperability?
- Open-source such as FIWARE with Orion-LD and Smart Data Models (e.g., Smart Aeronautics and Smart Sensing)
- Public cloud provider such as AWS with SNS and Neptune