This repository holds a prototype of the project I²oT, an inventory system that aims to monitor and track the patrimonial gods of an institution. The web system works like a social media where the smart objects are your friends and you can see information about them. We also created an Ontology called I²oTology, which is an extension of the IoT-Lite, a light ontology for IoT devices.
To understand more about the project, there are some articles and reports you can access:
Articles
-
I²oTegrator: a Service-Oriented IoT Middleware for Intelligent Object Management (English)
-
I²oTology - Tracking-Oriented Ontology (English)
Reports
- Java
- Eclipse
- Hibernate
- Tomcat Server
- Apache Jena
- Protégé
Currently I'm improving the documentation of this project to help others to understand what I did and how you can contribute with the community in this field. This table shows the goals and tasks I have reached and what are the ones I still need to complete:
| Goal | Description | Status | Date |
|---|---|---|---|
| Update Repository | Structure the project and update the GitHub repository. | ✔️ | 10/03/20 |
| Write Getting Started | Write a getting started readme with the configuration needed to run the project locally. | ✔️ | 10/03/20 |
| Write Architecture readme | Write about the architecture/layers I used in this project. | ✔️ | 10/04/20 |
| Write a table of tasks | Write about all tasks I have done and what must be done to continue improve the project documentation. | ✔️ | 10/04/20 |
| Write a project overview | Write about how the project should work, what are the pages, what is done until now. | ✔️ | 10/04/20 |
| Push the arduino code | Push the arduino code to the GitHub | ✔️ | 22/02/21 |
| Write about the hardware | Write about the hardware used in this prototype. | ✔️ | 23/02/21 |
| Push the hardware schematic | Push the hardware schematic to the GitHub | ✔️ | 20/04/21 |
| Improve the code documentation/javadoc | Improve the documentation of all the code. | ⏳ | |
| Write Ontology readme | Write a readme about the ontology and how to get started using the Protégé. | ⏳ | |
| Write a Contributing readme | Write about how someone can contribute to this project using this repository. | ⏳ | |
| List all the service routes | List the routes created in the servers showing the HTTP method and what is used for. | ⏳ |
Basically the system architecture diagram flows like this:
Here I will explain each layer briefly:
-
Model: The model is the definition of the objects that will be persisted to the database and also used by the system as a normal object manipulation. In this layer I created a Generic Model where were created all the common characteristics that all objects will have (e.g. ID and some set and get functions), and all the other models inherit from this Generic Model these characteristics, so that we can access the inherited characteristics and also create others that are specific for this object/model.
-
Database: The database I used for this this project is the MySQL, but it could be anyone, it was a choice based in the popularity and the highly use of this database by enterprises. To manage the database, the MySQL Workbench was used, To understand more about configuration, see the Getting Started page.
-
DAO: The DAO (Data Access Object) is a pattern for data persistence. This pattern allow us to uncouple the application from the data persistence layer. It's also good for code management, since a layer does not "disturb" the other when trying to update a code from a layer. This pattern allows us to configure the rules to persist a model to the database. Similar to the Model Layer, I created a GenericDAO which holds all the rules that are common to all the other DAOs that I created, so all of them will inherit from this GenericDAO and eventually can also implements specific rules.
-
Service: The service is a layer created in order to uncouple the application to the database and server responsibilities, so that the layers could be divided and configured in different hosts. The service uses the Model and the DAO layers to create routes that can be triggered through RESTful HTTP requests.
-
Controller: The controller layer in this case is used to call the Service routes and then update the view when requested by it. It needs to use the Model to "understand" how to interact with the objects returned by the services. It also connects to the Ontology Service and Ontology Model that will work similarly to the Service and Model layers.
OBS: In some frameworks the concept of the Controller changes. Some of them abstracts the Service layer and uses the Controller as the routes definer. So in this case the Controller offer the RESTful routes and the View layer interact with the Controller through the RESTful API. For example, in javascript you can use the Adonis.js as the server to offer the Service through the Controller layer, and use the React.js to construct the View layer and the Axios.js to make requests to the Adonis server. -
View: The view is the frontend HTML tags that will communicate with the Controller to get the updated data received by the requests to the services. The view pages are located at SistemaWeb-I2oT/src/main/webapp/pages. When you run the Tomcat server, it will serve the services and also the view pages, and you are able to access the routes through some RESTful API caller (as Insomnia) to make GETs and POSTs HTML requests and use the browser to access the pages.
-
Ontology: The ontology is where some objects are configured and classified in order to infer new information about them and return to the system which will execute some decision. This ontology is an extension of the IoT-Lite, a light ontology for IoT devices, we took this ontology and created some more classes that we needed to use within this project. The ontology is located in SistemaWeb-I2oT/src/main/java/ontology/i2otology.owl.
-
Ontology Model: It's similar to the Model layer, the difference is that we define the characteristics thinking about the Ontology and not the Database, because of that, these models cannot inherits from the GenericModel once the Ontology model characteristics are different from the Database model characteristics.
-
Jena Library: Is the library created by the Apache used to communicate with the Ontology file. Here you can see more information about the documentation. This library also offers some methods where we can activate the reasoner and make some inferences inside the ontology.
-
Ontology DAO: It's similar to the DAO layer, but it will not use the database to persist the data, it will use the library Jena to communicate with the ontology file (i2otology.owl).
-
Ontology Service: Similar to the Service, but will serve the ontology routes using the Ontology DAO layer and the Ontology Model layer.
In this section I'll show the project working locally. Well, as said in the beginning of this readme, this project is a prototype so is not a complete project which has all the webpages created and working. Another things is that the project was built using the portuguese language, so most of the terms and messages are in portuguese.
The main page I worked to validate this project and to publish about this proposal is the Smart Objects page, where was built thinking in as a social network page. So this is the visual:
In the right side of the image you see the list of objects registered in the MySQL database. In the left side we have the "Mensagens" (Messages) that is where the user will see some warning about the objects coming from the system that consulted the ontology using the reasoner. And the other component "Objeto" (Object) shows the smart object information when we click in one of the objects in the list. The page is updated every 60 seconds in order to verify constantly if there is some warning.
-
Page updating: Every 60 seconds that the page updates, the Controller verifies the Ontology Service through the route
http://localhost:8080/I2oT/rest/smartObjectOntology/status/[smartObjects]passing the actual list of objects. So the ontology service will verify the ontology (i2otology.owl) through the Ontology DAO layer using the reasoner option of the Jena framework. So the response will be a list of messages, if the list is empty that means that there is no warning, so the "Mensagens" component will be empty. -
Smart Object clicking: When clicking on a smart object listed, the Controller will consult some information to the database through the Service layer (route:
http://localhost:8080/I2oT/rest/smartObject/{id}) , and after that will send this object to the Ontology Service layer (route:http://localhost:8080/I2oT/rest/smartObjectOntology/info/{smartObject}) requesting an ontology analysis of this object. The return will be all the object status, showing who moved this object, where it is right now, if this person is allowed to carry this object, if the object can be in this location and if the object material is not dangerous. So the Controller after having the information ready, delivers to the View.
Go to the Getting Started page
- Jauberth Abijaude
- Péricles Sobreira
- Fabiola Greve
- Levy Santiago
You can ask me anything, here is my email: levyssantiago@gmail.com, I have done this project for a little while so I don't remember much everything I did, but I'm willing to help with any problems you might have about configuring the project to your machine.