Research
- Computation and information techologies. (PI: Jordi Jiménez)
- The origin and destination of the route that has to be optimized
- Characteristics of the Ship
- Meteorological Data and IMO Safety Guidelines
- >Environmental Protected Areas in the route
- Access to sensor data through mobile phones.
- Actuation and Control.
- Notifications System to receive alarms.
- Human Reality to empower the user experience.
The main research lines of CIMNE TIC consist of Artificial Intelligence technologies, IoT platforms, Decision Support Systems, GIS technologies, Blockchain and Web and App technologies covering applications in different engineering fields such as civil, aerospace, environmental, mechanical, telecommunication and bio-medical engineering.
Artificial Intelligence Technologies
Artificial Intelligence is one of the major research lines of CIMNE TIC. Our expertise consists of working with state-of-the art Artificial Intelligence technologies namely computer vision, deep learning, machine learning algorithms, route planning optimizers and data analytics tools to develop ground-breaking applications.
Predictive Maintenance Tools
CIMNE TIC develops predictive maintenance tools based on machine learning models to convert maintenance tasks from a preventive or corrective strategy to a predictive maintenance strategy. Using these Predictive Maintenance Tools maintenance is carried out only when they are necessary and infrastructures or manufacturing failures can be anticipated. This change provides a reduction in terms of costs derived from maintenance operations, greater security at critical elements, and the lengthening of the remaining useful life of every related component.
Deep Learning and Computer Vision Tools for object detection
CIMNE TIC investigates in Deep Learning and Computer Vision techniques to count objects in a scene and determine and track their precise locations, all while accurately labeling them.
One example of CIMNE TIC Object Detection Application is in the civil engineering field. Construction is one of the sectors where more accidents happen. An accident, additionally to the harm suffered by the worker, incurs in several costs for the company, from human resources costs to penalizations or fines derived from the accident. In order to tackle this problem, computer vision is one of the most used tools thanks to the current Deep Learning algorithms accuracy.
CIMNE has developed a prevention system based on Deep Learning Algorithms and Computer Vision technologies to detect the use of personal protective elements, specifically the helmet, vest and harness, through the image provided by a camera. Once a security fault is detected, the worker is tracked alongside for a few seconds to asses if the fault was momentarily or not.
Deep Learning and Computer Vision algorithms to monitor and track the human movements
CIMNE TIC is developing a virtual assistant able to correct in real time a person that is performing exercises in front of a camera. The subject can perform training exercises as well as rehabilitation exercises. A set of 17 keypoints such as knees, feet, head, etc. can be detected from a plain image and build the structure of the human body. The pose of the person is analyzed and a positive or negative feedback is given to the subject. A Deep Learning model has been trained in order to improve keypoints' detection performance.
Data analytics and Machine Learning
CIMNE TIC performs many diverse types of data analysis to reveal trends and metrics that would otherwise be lost in the mass of information. This information is used to optimize processes to increase the overall efficiency of a business or system.
One example of CIMNE TIC Data Analytics application is the customer churn prediction in the insurance sector.
Development of a Machine Learning powered system to predict customer decisions concerning the renewal of their insurance policies. The system, currently in its last development phase, is being integrated with the Spanish most used insurance broker management software, segElevia, which is developed by MPM Insurance Software Solutions. Also, internal tests are being carried out with several insurance brokers previously to the deployment in production.
Structural problems segmentation
CIMNE TIC investigates computer vision and deep learning algorithms for Structural Health Monitoring (SHM) to accurately estimate the state of deterioration of infrastructure.
Machine Learning for Early and Automated Disease Detection
CIMNE TIC investigates automated methods for the detection of neurological and respiratory illnesses through the extraction of voice features and by applying machine learning algorithms. Currently, we are investigating the automated detection of bulbar involvement in ALS patients and COVID-19 positives in close collaboration with the Universitat of Lleida (UdL), the Motoneuron Unit and Pulmonology Unit of the Bellvitge Hospital and the Internal Medicine Unit of the Hospital Universitari Arnau de Vilanova de Lleida.
Route Planning
This CIMNE TIC research line aims at giving a concrete contribution to the efficient, safe and environmentally friendly maritime transport through development of innovative e-navigational services to the shipping industry. For this purpouse, CIMNE TIC is developing a Route Optimization Service taking into account involuntary speed reduction due to wind and waves.
The inputs for the service are:
Web/App Platforms
IoT Platforms
CIMNE has wide experience developing on-demand cloud-based and on-premise software packages and related services that enable and support IoT services. CIMNE IoT platforms enable end-users to manage several devices and connections across multiple technologies and protocols and enables to combine device and connection data with specific customer data as well as data from third-party sources like social and weather data to create more valuable IoT applications.
Decision support systems in engineering
One of the main lines of CIMNE TIC is the integration and combination of internet technologies, IoT devices, GIS technologies and AI technologies with numerical and computational methods to develop Decision Support Systems (DSS). CIMNE TIC applies these DSS in different engineering fields such as civil, manufacturing, aerospace, environmental, mechanical, telecommunication and bio-medical engineering. Examples of DSS include DSS for risk management of floods in landscape and urban areas, risk management of water hazards in aquatic ecosystems and marshes, support to clinical interventions in cardiovascular diseases, prevention and treatment of epidemiological diseases and care support for handicapped and elderly people, among other applications.
Web Technologies
CIMNE TIC has a wide experience in the development of responsive modern web platforms using HTML5 technologies. Those client-side technologies include Ajax and JavaScript frameworks and the Document Object Model (DOM) to update regions of the page area without undergoing a full page reload to allow users to continue interacting with the page. In order to build websites that can run smoothly also in mobile phones, we use responsible architecture design, which is the most effective answer to the incompatibility problem of web pages across devices with varying sizes and resolutions. User interactions behavior is cross-device and continuous; therefore, it is also important to provide a fluent and consistent user experience for the same webpage across various devices.
App Technologies
CIMNE TIC has wide experience developing mobile applications (mobile APPs) for Android and iPhone/iPad.
The first mobile APP developed was a monitoring platform for mobile devices. It has been used in environmental and aeronautical sectors and has the following features:
The next applications developed by CIMNE TIC have been focused in tourism and health sectors.
Regarding applications related with tourism, CIMNE TIC has developed the Beaching App which was implanted in Ibiza two years ago and this year it has been introduced in other coasts, experiencing a high amount of downloads.
Beaching is an APP that allows to users (tourist, residents, visitants…) having information about the near places (offers from near establishments and shops, recommendations about beaches and their state, information about events near where you are, etc…).
CIMNE also investigates and develops Mobile APPs based on mobile sensor techniques (use the mobile as a sensor) to detect disintegrations, deformations, potholes in the pavement by using the accelerometers and GPS systems embedded in mobile phones. These monitoring techniques allow early detection of surface damage on the pavement.
