TI & IA
INFORMATION TECHNOLOGIES & ARTIFICIAL INTELLIGENCE

Information Technologies & Artificial Vision

 

We have more than 10 years’ experience developing systems in which we apply Information Technologies & Artificial Intelligence for customers in a wide range of industries.

From the start, our approach is to design and implement flexible, robust, and easy-to-maintain solutions.

Information Technologies (IT)

According to the Polytechnic University of Valencia, IT is responsible for the acquisition, processing, storage and dissemination of “information” regardless of its format (voice, images, text, numeric) through the combination of computer and communication techniques.

In Orbita Ingenieria we apply the design and architecture characteristics detailed below, with a methodology in constant evolution that is defined internally and implemented in each project.

 

IT | Design characteristics

arquitectura n capas con orientación a dominio
  • Object oriented programming (OOP)
  • Applying design patterns
  • Modular design
  • Layered architecture
  • Best practices guide
  • Infrastructure
  • Base classes

— Abstractions

— Extensions

— Wrappers

 

 

“GUÍA DE ARQUITECTURA N-CAPAS ORIENTADA AL DOMINIO CON .NET”. César de la Torre Llorente, Unai Zorrilla Castro, Miguel Ángel Ramos Barroso, Javier Calvarro Nelson. 2010.

It | Architecture

  • Domain-Oriented Design (DDD)

— Provides an application SOLID model

domain driven design

 

  • Micro-services architecture

CI/CD strategies

  • Dependency injection
  • Automatic Deployment (DevOps)
  • Aspect oriented programming (AOP)

— Interceptors

  • Database-agnostic (ORM)
  • Cloud Architecture and Services (IaaS)

— Amazon Web Service (AWS)

  • Audit services
  • Uncoupling
“DOMAIN DRIVEN DESIGN”. Eric Evans. 2004

— Message queues

  • Object Mapping 
  • Localisationdevops Orbita
  • Background Jobs
  • Fault tolerance

— Replication

— Distribution

  • Logging
  • Testing
  • Templates
  • Monitoring

 

“DevOps”. Basado en ilustración vista en juancarlosabaunza.com

 

microservicios
Basado en “.NET MICROSERVICES: ARCHITECTURE FOR CONTENARIZED .NET APPLICATIONS “. Cesar de la Torre, Bill Wagner, Mike Rousos. 2019

 

 

Artificial Intelligence (AI)

Our AI projects can be divided into three categories, including smart cameras, PC-based systems and Deep Learning based systems. The right choice depends on multiple variables such as performance, scalability, user interface, or cost.  We advise and adapt the most appropriate solution.

 

AI | Intelligent Cameras

Devices that combine a camera, a vision algorithm processor, and the integration of various interconnection technologies with PLCs or other devices. These are stand-alone devices and do not require interaction with a PC, only for programming or monitoring purposes. Since their performance is reduced, their functionality is limited and inflexible they are recommended to address basic problems.

 

 

 

 

 

AI | PC based systems

PC based systems offer the highest degree of flexibility. In this type of system, it is possible to integrate a set of sensors of various technologies: matrix cameras, linear cameras, 3D sensors, hyperspectral cameras. Therefore, a system is necessary that offers a degree of computing adaptable to the needs of the system.

In Orbita we are specialists in this type of system, since its flexibility and power allows us to address the most complex projects.

 

  • Vision sensors

 

 

 

 

  • Lighting systems

  

 

 

 

  • Vision libraries on which we develop our algorithms:

 

 

 

 

 

 

AI | Deep Learning based systems

Deep Learning based systems have been hugely popular in recent years. They are based on neural networks, and are therefore, automatic learning techniques applied to image processing.

They offer a new paradigm with respect to traditional techniques, since they do not require the definition of complex decision algorithms, but from a large volume of labelled images they are able to decide autonomously. This allows them to address complex problems where there is great variability, both in the environment as for the objects to be analysed.

Despite its popularity, its application is not always viable for all problems detected in industrial processes and therefore we consider that the application in conjunction with Deep Learning techniques together with traditional techniques is the key to the success of our systems. Our team is expert on these systems and we have successfully addressed several projects with these technologies, both in classification, object detection, anomaly detection, defect detection and more.

 

 

 

 

The Orbita solutions where these technologies are applied are:

Orbita ControlOrbita Trace

Orbita Industrial IT

 

 

 

 

 

 

 

 

 

Products | Information Technologies

 

 

Products | Artificial Intelligence

 

 

More information? ¡Contact us!