Field Demonstration of Bendable Concrete Precast Pavement

The aim of this project is to reduce the construction time of road pavement and thus enhance the productivity of road construction. To achieve this goal, a new type of concrete that is bendable yet stronger and longer lasting than regular concrete is used for the creation of slim precast pavement slabs, thus halving the time needed for road works and new pavements. It is also more sustainable, requiring less maintenance. This new concrete material can greatly reduce the thickness and weight of precast pavement slabs, hence enabling speedy plug-and-play installation, where new concrete slabs prepared off-site can easily replace worn out ones (Fig. 1). In addition, this technology not only enables the construction industry to reduce labour intensive on-site work, enhances workers’ safety and reduces construction time, it also benefits road users by cutting down the inconvenience caused by road resurfacing and construction works.

This project is led by Assistant Professor Yang En-Hua from the School of Civil and Environmental Engineering in collaboration with Technical University of Munich, Jurong Town Corporation (JTC) and Land Transport Authority (LTA). The bendable concrete precast pavement will be scaled up for further testing over the next three years at suitable locations within JTC’s industrial estates where there will be heavy vehicular traffic.

(source: www.fhwa.dot.gov)

Web-Based 3D GeoData Modelling and Management System (GeM2S)

The main objective of the project is to establish a Web-based three-dimensional (3D) Geological and Geotechnical Data Modelling and Management System (GeM2S) to reduce construction cost and increase productivity for future underground construction projects in Singapore. A huge amount of geological and geotechnical data has been collected in the past. The 3D model and database will make use of existing borehole data as well as validated in-situ and laboratory data which can be updated with new data when these are available in the future.

Through the system, virtual borehole and cross-section can be created online as part of Building & Construction Authority’s Geoscience Information Sharing Portal; the geological conditions at a site can also be evaluated together with the geological or geotechnical model established. Uncertainties involved in the design parameters can be reduced in this way.

The proposed 3D GeM2S system can be used by both government agencies and industries for underground space planning or infrastructure developments such as buildings, roads, MRT, or underground caverns construction.

The project is led by Prof Chu Jian from the School of Civil & Environmental Engineering in collaboration with Building & Construction Authority, Land Transport Authority, Urban Redevelopment Authority of Singapore, NTU Earth Observatory of Singapore, Hong Kong University and British Geological Survey.

 

 

Figure: 1 Superhighway for Used Water Management (from PUB website)

Figure 2: Superhighway for Used Water Management (from PUB website)

Figure 3: Conceptual Design of a Robotic Platform

 

 

 

Test Bedding Research for Innovative Technologies in BCA SkyLab

To drive the research and development of energy-efficient building technologies in the tropics, an advanced rotatable testbed, BCA SkyLab, was developed in Singapore in collaboration with Lawrence Berkeley National Laboratory (LBNL). As the world’s first high-rise rotatable testbed for the tropics, this facility is an outdoor testbed sitting atop a 7-storey building, with a rotatable platform that can simulate various orientations of building. The testbed allows testing and evaluation of performance of various technologies in the tropical urban environment, including lighting, air-conditioning, façade systems, shading system and control systems.

Investigations will cover an integrated system that uses auto-dimming lighting control based on outdoor illuminance level through the digitally addressable lighting interface (DALI) combined with an automated daylight-redirecting blind system, active chilled beam system and thermo-chromic glass. The infrastructure developed and data obtained through this project will shed new insights in new energy-efficient technologies and building energy efficiency standards (enhancing from the current highest standard BCA Green Mark Platinum). Led by Asst Prof Wan Man Pun from the School of Mechanical & Aerospace Engineering, the research team also consists of researchers from Building and Construction Authority (BCA). 

Courtesy of Building and Construction Authority, Singapore
Picture credit: Surbana Jurong Pte Ltd

 

Project Paddington

In project Paddington funded by MINDEF, Asst Prof Erik Cambria from the School of Computer Science and Engineering develops and applies sentic computing for machine learning in autonomous systems e.g., for intelligent agents and robotic applications.

Sentic computing is a multi-disciplinary approach to natural language processing and understanding at the crossroads between affective computing, information extraction, and common-sense reasoning, which exploits both computer and human sciences to better interpret and process social information on the Web.

In sentic computing, whose term derives from the Latin 'sentire' (root of words such as sentiment and sentience) and 'sensus' (as in common-sense), the analysis of natural language is based on linguistics and common-sense reasoning tools, which enable the analysis of text not only at document-, page- or paragraph-level, but also at sentence-, clause-, and concept-level.

In the project, sentic computing will be exploited to understand sentiments and emotions in different modalities and different languages. Additionally, the project will also serve multiple purposes linked to sentiment mining, including human-robot interaction, emotional conversational agents, intention awareness, and domotics.

 

Figure above:  Flowchart of the Sentic Computing Framework. Text is first deconstructed into concepts. If these are found in SenticNet, linguistic patterns are applied. If none of the concepts is available in SenticNet, a machine learning approach is employed.

Integrated Systems for Future Air Traffic

Increase in air traffic density is a major challenge in Air Traffic Control (ATC) worldwide. Air traffic density is forecasted to double by 2025 (ICAO, 2006).  However, current ATC systems are approaching maximum capacity and existing ATC practices are unlikely to be able to sustain the expectant growth (CANSO, 2012). This may cause inevitable burden on air traffic controllers (ATCOs) and ultimately compromise air traffic safety.

The project led by Prof Chen Chun-Hsien of the School of Mechanical and Aerospace Engineering and team in collaboration with reseachers from Fraunhofer IDM@NTU and partners from Civil Aviation Authority of Singapore, Shenzhen University and Chiba University aims to develop and evaluate the future work place for ATCOs.

In the planned new work place, new human computer interfaces and 3D visualisations of flight data would be developed to let ATCOs interact and better comprehend the large amount of time-critical information shown by the displays in the control room (Figure 1). For example, details of aircraft flight paths essential to fast and robust flight management and planning would be displayed in 3D.  Potential conflicts would also be highlighted on the screen.

Next, a series of user studies would be performed to investigate the trust, dependence, workload, performance and situation awareness of the ATCOs. Evaluation would be done real-time using bio signals such as EEG obtained from brain computer interfaces to give more reliable measures for all studies (Figure 2). Through doing this, a better understanding of ATCOs’ working patterns can be obtained and automation support can be extended appropriately, improving the capability to meet future air traffic growth.

The project is an initial effort to improve current aviation infrastructure to enhance performance of ATCOs and reduce risks of air traffic control incidents.

References:

ICAO. (2006). Report of the Second Meeting of Automatic Dependent Surveillance - Broadcast (ADS-B) Study and Implementation Task Force (ADS-B SI TF/2). Bangkok, Thailand: International Civil Aviation Organization.

CANSO. (2012). Accelerating Air Traffic Management Efficiency. A Call to Industry. Paper of ATM Global Environment Efficiency Goals for 2050. Civil Air Navigation Service Organization.

Figure 1. Rendering of the future air traffic work place with touch interfaces

and 3D flight management

 

Figure 2. The ATC setup for the user study

 

 

Digital Manufacturing: Embracing the ‘Factory-of-the-Future’

Economic globalisation and urbanisation has continued to fuel new trends and demands in manufacturing and supply chain networks. In response to challenges and to seize growth opportunities in the manufacturing and service sectors, both of which are key industries and growth engines of the Singapore’s economy, the School of Computer Science and Engineering teamed up with A*STAR SIMTech through a SIMTech-NTU joint Laboratory to harness computational intelligence and complex systems technologies to address complexities in manufacturing operations and supply chain networks that will position Singapore for “Factory-of-the-Future” and maintain leadership as master facilitative control hub for supply chain in Asia Pacific.

Notably, the Joint Lab deploys the Large Engineering Supply Chain Adaptive System (LesCaS), a NTU Licensed decision support and optimisation software designed for large scale logistics and supply chain management, and the Algorithm Development Environment for Problem-Solving (ADEP), another NTU patented system with wide range of heuristics and mathematical programming tools and codes self-configured optimiser, in its research.

The 5-year SIMTech-NTU joint laboratory programme which is funded at $2.2m (and supported by 20 PhD A*GA scholarships) led by Prof Ong Yew Soon, Director of SCSE Centre for Computational Intelligence, had attracted industry interest. The research of the joint lab is aligned with SCSE’s research foci from Computational Intelligence, Big Data Analytics, Multi-Agents Modelling and Simulation to Complex Systems Optimisation.

Two of the “Factory-of-the-Future” themed projects currently undertaken at the SIMTech-NTU joint Laboratory are: “Master Facilitative Control Tower for Risk Management of Complex Supply Chains” and “Multi-Objective Vehicle Routing for Last Mile Logistics”.

In the project on Risk Management of Complex Supply Chains prompted by the increase in scale, connectivity and vulnerability of supply chains, Prof Ong and teams from the School and SIMTech ditches the traditional approach for novel bottom up analysis of complex supply chain networks. The aim is to introduce innovations for more efficient complex manufacturing, smart production operations and resilient supply chain systems. The team’s new approach investigates how a complex supply chain works as a whole, the interplay of factors and components in complex, uncertain and varied scenarios. The technologies and models developed from the research could also be applied to the aerospace, maritime, chemical and other hi-tech industries.

Figure 1. Complex Manufacturing and Supply Chain Environment

For the project on Multi-Objective Vehicle Routing for Last Mile Logistics which saw partnership with The Logistics Institute Asia Pacific, NUS and SMU, the focus is to develop multi-objective, dynamic, eco-friendly collaborative vehicle routing for last mile logistics in the city area. Last mile logistics is currently regarded as one of more expensive, least efficient and most polluting sections of the entire logistics chain in the urban environment. Through novel route planning, routing, scheduling and optimisation algorithms and eco-indicators, the research helps logistics service providers improve efficiency, reduce environmental impact (fuel consumption and emissions) and enjoy cost and time savings.

Figure 2: Eco-friendly Collaborative Last Mile Logistics

The H-Man: Planar Robot for Arm Rehabilitation after Stroke

Click here for video clip.

Robot-aided therapy has been shown to require less effort from both therapists and patients and yet be effective in achieving intensive goal-directed repetitive arm motor retraining, which in turn improves motor control and strength. However, current robot-aided therapies involve complex, bulky and tethered machines which are expensive.

The H-man (H-shape cabled differential Manipulandum) designed and developed by Assistant Professor Domenico Campolo of the School of Mechanical and Aerospace Engineering in collaboration with researcher at the Tan Tock Seng Hospital, is a novel, compact, portable and inexpensive robot which with its computer-controlled force field generation that assist or resist a patient’s motion, aims to aid rehabilitation therapy of stroke patients to reduce motor weakness and impairment and improve coordination of upper limbs.

One appealing feature of the H-Man device is its easy deplorability in hospitals, at home and community centres thereby enabling early discharge and continuous monitoring and training of patients.

In the later phases, Prof Campolo and his team will further optimise the device to develop more solutions such as design of the rehabilitation video games to make the H-Man more attractive for licensing/spin-off purposes.

Specialised Structural Cells for Growing Street Trees in Constrained Planting Spaces

Roots of a tree are responsible for uptake of nutrients and keeping the tree stable. Urban trees are faced with the problem of congested growing space for their root system. The increase of infrastructures limits root spread, thus undermining the health and stability of trees.

Professor Harianto Rahardjo and Associate Professor Leong Eng Choon from the School of Civil and Environmental Engineering collaborate with Professors in the School of Materials and Science Engineering, NTU and National Parks Board Singapore on designing a cellular structure which consists of a load-bearing skeleton and lightly compacted soil mixture, called structural cell.

The structural cell replaces the highly compacted sub-base under the carriage way and part of the subgrade, hence creating an ideal environment for root growth and spreading. The structural cell also attempts to enhance storm water management by controlling infiltration rate of the soil mixture and discharge of accumulated water. The soil mixture is designed as a layered filter to prevent the fine particles migration, thus minimising maintenance.

The research launched in December 2012 is expected to achieve results in 3 years and test-bedded at some selected sites in Singapore.

A Vein Pattern Identification System

Leveraging on the advancement in imaging technology, digital photographs can be enhanced to reveal the vein pattern underneath a person’s skin. Such patterns are unique and immutable; it can be turned into useful tool for identifying a subject of interest. This technique is especially useful when other biometric traits of the perpetrator are not recovered at the scene of crime, except digital pictures e.g. in cases involving paedophiles – where the suspects’ faces are usually obscured or masked terrorists, etc.

Traditionally, it was difficult to use cutaneous vascular patterns for forensic identification, because they were nearly invisible in color images. The limitation was overcome by using Optical-based Vein Uncovering (OBVU) computational method that is sensitive to the power of the illuminant and does not utilise skin color in images to obtain training parameters to optimise the vein uncovering performance. The OBVU method however only supports manual verification.

In collaboration with the CID of the Singapore Police Force, Singapore Prison Service and researchers from the Los Angeles Biomedical Research Institute and University of Nottingham Ningbo, China, Assistant Professor Adams Kong from the School of Computer Science and Engineering and his team worked on two new schemes to overcome the limitations in the OBVU method.

Specifically, a colour optimisation scheme is used to derive the range of biophysical parameters to obtain training parameters and an automatic intensity adjustment scheme is used to enhance the robustness of the vein uncovering algorithm. In the event the images are JPEG-compressed, the de-blocking algorithm specifically designed for the skin is applied. The team also developed an automatic matching algorithm for vein identification. This algorithm can handle rigid and non-rigid deformations and has an explicit pruning function to remove outliers in vein patterns. The proposed algorithms were examined on a database with 300 pairs of colour and near infrared (NIR) images collected from the forearms of 150 subjects. The experimental results were encouraging and indicated that the proposed vein uncovering algorithm performs better than the OBVU method and that the uncovered patterns can potentially be used for automatic criminal and victim identification. Prof Kong’s team also introduced a full body imaging system that is effective in collecting and managing images and personal information from suspects and inmates

The full-body imaging system

• Original Image
• egmented skin image
• Corresponding IR image
• Result from (b)

Industry

NTU College of Engineering (CoE) views its strategic alliances with external agencies as an extension of itself. We have a long working history with industrial partners, for economic impact and academic excellence. We are constantly looking to developing fruitful partnerships that generate productive, commercially viable solutions for some of the world’s greatest challenges. CoE works with companies including Rolls-Royce, Bosch, Infineon, Lockheed Martin, ST Microelectronics, and THALES to name but a few.

For organisations which are keen to engage us on such cooperative efforts, whether for research collaborations or sponsorship, we are most pleased to talk to you. Please write to Assoc Dean (Research) at vd-res-coe@ntu.edu.sg