Ashiq Anjum

Title Intelligent Digital Twins: Artificial Intelligence and HPC meet Virtual Reality to offer Real Time 3D Analytics

The ability to visualize a system outcome before it happens is extremely valuable and the proposed Virtual Reality (VR) system makes this much easier and accessible. Engineers will be able to design better products while customers can see final products pre-production, ultimately saving everyone time and money. This talk presents a distributed Virtual Reality (VR) system for data driven model creation, AI based adaptive model evolution and high performance model execution for large scale immersive design and analytics. Creating VR visualisations from real world distributed data and analytics offers a unique opportunity for next generation analytics in many engineering, scientific and medical applications. This work aims to provide a distributed VR system for development and deployment of VR models across the networks. This will enable programmers and users to specify and create mathematical and engineering models of engineering applications in VR space. Users will be able to analyse data in 3D environment by producing real time models and visualisations as the data is captured/analysed by distributed engineering teams working on different aspects of engineering models. Dynamically integrating data, algorithms & analytics to a VR environment needs piecing hundreds of thousands of objects together and will need careful understanding of geometrical and system models to minimise the computational burden. To overcome this problem, we use high performance in-memory systems where we can store the components of the models in a distributed shared memory while users are working on their models. This research project enables distributed users, with minimal effort or specialist knowledge, to view, create and edit data within VR and deliver applications for use in areas such as engineering design, visualisation and immersive training. This requires distributed engineering and design teams to build collaborative VR models, analyse and integrate their own datasets to the VR models using AI and big data analytics algorithms and intelligently evolve, update and visualise these VR models in real time as the data sources undergo changes. This distributed VR analytics system offers real time collaborative design and development of engineering models and is one of the first attempts to offer VR enabled data analytics and their Immersive Visualisation. Our proposed solution offers a massive technological challenge bringing together several highly technical skills areas such as VR, distributed systems, AI and data science to deliver a virtual space where users can meet and work collaboratively.


Ashiq Anjum is a professor of distributed systems and director of the data science research centre at the University of Derby, UK. His areas of research include data intensive distributed systems and high performance analytics platforms for continuous processing of streaming data. Prof Anjum has been part of the EC funded projects in distributed systems and large scale analytics such as Health-e-Child (IP, FP6), neuGrid (STREP, FP7) and TRANSFORM (IP, FP7) where he investigated resource management and optimization issues of large scale distributed systems and provided platforms for high performance data analytics. He has been investigating large-scale distributed systems and analytics platforms for the LHC data in collaboration with CERN Geneva Switzerland for the last fifteen years. Before starting an academic career, I worked for various software multinational companies for around ten years. He secured grants from industrial partners, Innovate UK, RCUK and other funding agencies for investigating high performance video analytics systems for producing intelligence and evidence for medical, security, object tracking and forensic science applications. He is also closely working with healthcare providers, hospitals and pharma companies in investigating high performance analytics systems for distributed clinical intelligence and integration, iterative genome analytics and precision medicine. He has been actively working in collaboration with rail companies to investigate how rail infrastructures and services can benefit from Internet of Things (IoT) and real time analytics by intelligently analyzing streams of data arriving from rail networks to increase accuracy, reliability and capacity of rail infrastructures and services. In addition, he has also been investigating ways to model the rail networks as a distributed Graph System and provide adaptive scheduling and resource management systems. Thanks to a large grant from Innovate UK, he has been working with a leading VR provider to enable real time visualization of 3D engineering models and distributed algorithms in a Virtual Reality environment. This work allows distributed parties involved in large scale collaborative engineering projects to identify potential conflicts or required changes at the design stage, rather than during manufacturing, when they’re extremely costly to put right.