I2R's 20th Anniversary: Elevating Singapore's Science and Technology Capabilities through Innovative R&D

A*STAR’s Institute for Infocomm Research (I²R) celebrates its 20th anniversary this year. Over the past two decades, I²R has contributed its infocomm R&D capabilities in artificial intelligence (AI), connectivity and cybersecurity to many public agencies and local companies. Through impactful collaborations, it has co-developed smart solutions to enhance manufacturing, healthcare, urban solutions, sustainability and more, for a better future for Singapore and Singaporeans.

Infocomm technologies for safer and more efficient air, land and sea transport

Predictive Maintenance System
Developed as part of the SIA-I²R joint lab collaboration, the predictive maintenance solutions can identify sensor data recorded during flights and predict signs of failure in critical aircraft components with high accuracy using AI techniques. This allows for timely maintenance, reducing the risks of flight delays by detecting recurring defects and predicting component failures in aircrafts. 

Air Traffic Management System 
I²R and the Civil Aviation Authority of Singapore (CAAS) collaborated to establish a joint Air Traffic Management (ATM) research laboratory to facilitate scientific research on ATM to achieve innovation and optimisation in operations and further enrich Singapore’s ATM ecosystem through R&D. Areas include automatic speech recognition to facilitate ATM operations, data analytics for ATM performance measurements and improvements and the development of advanced visualisation technologies.

Developed based on the team’s research on the best achievable automatic speech recognition (ASR) performance for air traffic control officers (ATCOs) and pilots, a standalone demo of the ASR module for tower and approach control has been incorporated in the LORADS Human Machine Interface (HMI) for experimentation in simulated operations.

CoopeRative and UnIfied Smart Traffic SystEm (CRUISE) 
Developed in collaboration with Singapore’s Land Transport Authority (LTA), CRUISE is a next generation smart traffic light control system that improves traffic and pedestrian flow by using artificial intelligence (AI) and intelligent traffic light algorithms. Coupled with real-time data of the physical presence of vehicles and pedestrians, harnessed from existing traffic systems, global positioning  system (GPS) and sensors, it can optimise traffic light and pedestrian crossing timings to benefit both drivers and pedestrians.

Perimeter Intrusion Detection System 
Developed in collaboration with ST Electronics, the Perimeter Intrusion Detection System can precisely locate intrusion along the patented sensor cable with real-time perimeter monitoring.

The optical fibre cable solution requires no power or electronics in the field, providing immunity against electromagnetic interference (EMI) and lightning, and can be quickly deployed on any barriers, including fences, walls, drain gratings and sliding gates. It is currently deployed at Changi Airport and SBS Transit bus depots to enhance safety and security. 

Winner of IES Prestigious Engineering Achievement Award 2013 and ASEAN Outstanding Engineering Achievement Award 2013

Multi-purpose All Terrain Autonomous Robot (MATAR) 
Developed in collaboration with the Home Team Science and Technology Agency (HTX) and Singapore Police Force (SPF), the Multi-purpose All Terrain Autonomous Robot (MATAR) has evolved over the years. Powered by state-of-the-art hardware and software, the autonomous patrol robot augments police operations on the ground such as patrolling or routine checks, so that frontline officers can focus on more critical operations like responding to emergencies.

A later version, MATAR 2.0, was equipped with sound and visual sensors with 360° video capabilities to detect suspicious activity on the ground. The latest version of the robot, MATAR 3.0, has been further enhanced to capture a wider field of views with better reception and accuracy and additionally equipped with a tethered UAV for aerial surveillance in crowded situations.

MATAR has been trialled in various public places in Singapore and used to support police operations during major events, including the National Day Parade and Singapore Formula One Grand Prix.

Fish Activity Monitoring System (FAMS) 
Developed in collaboration with PUB as part of a joint lab, FAMS was an innovative water quality monitoring technology that used video imaging to remotely monitor the well-being of fishes at key points of the water loop, which is still in use to date through our start-up ZWEEC Analytics. 

Winner of the Honour Award (Applied research category) at the biennial Asia Pacific Regional Project Innovations Awards in 2012

Infocomm technologies for better automation and productivity 

SG Translate
Developed in collaboration with Singapore’s Ministry of Communications and Information (MCI), SG Translate is a customised neural machine translation engine powered by AI algorithms that was first developed for government use. It was further incorporated by MCI into SG Translate Together where the public can use the dedicated web portal for accurate translation across our four official languages, including localised terms.

RadiLogic 
Developed in collaboration with Tan Tock Seng Hospital, RadiLogic is an AI-powered system that expedites the chest X-Ray screening process of patients suspected of having COVID-19 pneumonia. The system can analyse each chest X-ray within three seconds and highlight abnormal cases that show signs of pneumonia with up to 96% accuracy, which greatly enhanced the efficiency and accuracy of the diagnosis process so that patients can receive the appropriate care more promptly.

Winner of Singapore Computer Society (SCS) IT Leader Award 2021 and National Healthcare Innovation and Productivity Award (HIP) 2021

SoundEye 
SoundEye has developed sound-activated emergency monitoring systems that incorporate technology that was developed at I2R. Powered by sound-recognition technology and sensors like thermal infrared and ultrasonic detectors, they can detect abnormality such as falls or calls for help from elderly people living alone to alert and get help from caregivers and emergency services in a timely manner.

Remote Vibration Monitoring System
Developed in collaboration with FOSTA Pte Ltd, the IoT-based monitoring system allows workers real-time, remote access to on-site data such as vibrations and noise data by facilitating machine-to-machine communications over the Internet. Workers can access the data remotely from the office at regularly scheduled intervals, and also do timely reporting through the automated collection of data, saving them the hassle to go to the physical site about 40 times a week to manually retrieve data, leading to an increase in productivity by 50%.

Silt Imagery Detection System (SIDS)
Developed in collaboration with PUB as part of a joint lab, the Silt Imagery Detection System (SIDS) used image analytics technology for real-time detection of silt discharge and CCTV downtime, where the relevant parties were automatically alerted upon detection of silt discharge or image problems. It was implemented across hundreds of construction sites to monitor Earth Control Measures (ECM) in a manpower-efficient manner. 

A*STAR’S I²R’S PIONEERING TECHNOLOGICAL INNOVATIONS

Early 1990s  -  World's first search engine for libraries
I²R (then Information Technology Institute (ITI)) developed the world’s first search engine for libraries using the Z39.50 protocol, which is an international standard on communications protocol for database search. It was first deployed across Singapore’s public libraries in the early 1990s to facilitate interlibrary catalogue searches and loans, where information can be remotely accessed easily over the internet.

2000 - World's first commercial surgical planning system

The Dextroscope was the world’s first commercial surgical planning system created by Volume Interactions Pte Ltd, a spin-off company from I²R (then Kent Ridge Digital Labs (KRDL)). Based on virtual reality principles, it was used to plan surgical procedures and offers a detailed, three-dimensional view of anatomical relationships and pathology in patients. It was used in the planning of several craniopagus twin separations, including the Zambian twins (1997) and German twins (2004) at Johns Hopkins Hospital, as well as the Nepali twin's separation (2001) at Singapore General Hospital.

2001 - World's first automatic video editing software
muvee Technologies, a spin-off of I²R (then KRDL) invented the world's first automatic video editing software for Windows, which can analyse videos automatically for human faces, scene boundaries and more using advanced signal processing techniques.

It was first launched in 2001 for computers, before being adapted for mobile on the Symbian 3 platform for the world's first video camera phone, the Nokia 7610. muvee Technologies’ automatic video editing solutions are now integrated into various global applications.

2010 - World's first adaptive lossless audio streaming technology

The world's first adaptive lossless audio streaming technology, which uses the MPEG-4 Scalable to Lossless (SLS) audio standard developed by I²R, was launched in 2010 by A*ccelerate (then Exploit Technologies Pte Ltd (ETPL)). MPEG-4 SLS is capable of compressing any audio file into less than half of its original size without any loss in quality, making high-quality music accessible anytime and anywhere, such as on mobile phones and audio streaming platforms.

It taps on the earlier Advanced Audio Zip (AAZ) technology developed by I²R, which was adopted by MPEG international standards and officially published as an international standard in ISO/IEC 14496-3:2005/Amd.3:2006 in 2006 – a significant achievement where standards were typically dominated by big industry players back then.