Artificial Intelligence and machine learning.
Focusing on developing distributed algorithms which can be deployed across a network to solve the industry needs.

The algorithms developed in the AIML lab are designed to work in internet of things, hyper-scale and AR/VR environments.

The TSSG’s Artificial Intelligence and Machine Learning lab focuses on developing distributed algorithms which can be deployed across a network to solve the needs of industry. The emergence of Artificial Intelligence as a disruptive technology over recent years is driving a market for AI applications across many sectors including agriculture, financial services and transport. The algorithms developed in the AIML lab are designed to work in internet of things, hyper-scale and AR/VR environments.

The team in the AIML lab takes inspiration from the cognitive abilities of the human brain and designs algorithms, process and applications that models its behaviour. The AI team are building lightweight AI algorithms that can vary their complexity depending on the urgency of the decision to be made in a similar fashion as to how the human brain can concentrate on complex tasks in short bursts. These algorithms are at the fore of edge-based AI algorithm design where balancing device constraints against decision performance is key. Similarly the AR/VR team are attempting to model how the human brain processes virtual content to trigger emotive responses. This research is key to developing personalised dynamic content that maps to your personality and mood and will potentially lead to the development of content for the treatment of certain mental disorders.

There are three thematic areas to the AIML Lab and they include:

  • Fog Data Analytics.
  • Artificial Intelligence Algorithms and Applications.
  • Augmented Reality and Virtual Reality.

The maturation of the Internet of Things and Big Data technologies has produced some highly innovative services and applications. However, many application scenarios are constrained by low energy consumption or low latency requirements where it is necessary to utilise processing resources closer to the source of the data.

This Fog Data Analytics team focus on developing algorithmic processes that take advantage of available computing resources to deliver the best results in constrained environments. The objective of this research is to split up complex applications and process individual components using the most appropriate resources possible. The technologies delivered here work at the intersection of the Internet of Things, High Performance Computing and Fog Networking to produce highly novel applications for future society.

A prime example of this is our R&D activities in the precision agriculture sector where using technology, farmers are identifying and monitoring efficiencies in their environment. The key piece of technology our teams are working is using advanced Fog Data Analytics algorithms that infer fine-grained movement and behavioural patterns of dairy cows using energy efficient sensors. This research project is contributing to a highly novel animal wearables platform where further inferences are derived by our algorithms based on herd behaviour and interactions with their environment.

Artificial Intelligence

Researchers
  • Eric Robson
  • Philip O’ Brien
  • Deirdre Kilbane
  • Christine O’ Meara
  • Gary McManus
Projects
  • CareLink
  • Smart Appi
  • Inspiration

Deep Learning algorithms have dramatically transformed the field of Artificial Intelligence in recent years and is not only driving a new breed of interaction models and interfaces to existing applications and services but also pushing the cognitive intelligence of technology forward at a rapid pace.

The A.I team are focusing on building cognitive intelligence algorithms to produce services capable of reasoning about their current environment. We are currently focusing on distributing AI algorithms over low powered mobile devices to use cognitive services that will compensate users with degenerative brain conditions. The work conducted by this team is driving the A.I. strand in the TSSG Brain Initiative project.

The core skills of the AI team comprise of data scientists that devise new distributed algorithms along with data engineers who build the distributed infrastructure which scales to the needs of the end users. The team have deep machine learning backgrounds and are also currently working on forecasting technologies for the Irish dairy sector and market segmentation technologies for the US insurance sector.

Source: www.siliconrepublic.com

Augmented Reality and Virtual Reality

Researchers
  • Ian Mills
  • Steve Barnes
Projects
  • VR Glove
  • EngageNet
  • SenseAR

Augmented Reality and Virtual Reality technologies are achieving mass adoption in numerous sectors including education, manufacturing and gaming. Using a HMD “Head Mounted Display”, VR creates an environment via software that immerses the user in a virtual world which suspends belief. AR is similar but uses holograms which are overlaid on the user’s real environment and enhances or adds to what they can see via data displays or 3D models. These technologies produce new immersive human machine interaction paradigms that are not only transforming how we consume information but also the speed of how we comprehend higher level concepts.

THE RESEARCH IN THE ARVR LAB FOCUS ON TWO ON MAIN ACTIVITIES

AR/VR Communication Platforms: Future telepresence systems will not only require seamless communication in single or multi-user scenarios but also allow users to interact and collaborate in augmented or virtual environments. The ARVR researchers are delivering a next generation enterprise and personal communication system at a much lower cost and barrier to entry to traditional systems. They are investigating pre-scanning and real-time user capturing systems via depth sensing camera hardware, 3D mesh modification representation and enhancement, 3D mesh transmission and the use of adaptive level of detailing systems to manage and maintain quality of service between variable connections and disparate end user hardware systems.

NeuroVR As VR hardware improves through new technology and future iterations of existing hardware, users will experience increasingly immersive experiences. Our research into the area of Neuro VR is currently investigating the relationship between the brain and the reactions within nerve clusters to specific and adaptive virtual stimuli. The results of this research could potentially lead to new neuro-rehabilitation and treatment techniques in addition to bench marking and improving the level of immersion within VR worlds or applications.