Welcome to Su Rong's Homepage
Cyber
Physical Intelligent Systems Research Group Webpages Enter
Here!
Ongoing/New
Projects
Project 1: Robust Autonomous Mission Planning for a Fleet of Heterogeneous UxVs (NRF CARTIN and Naval Group, NTU PI, August 2022 - July 2025)
Project 2: Real-time Resilient Fleet Management and Production Planning for Cobot-AMR Systems (A*STAR IAF ICT, SHARE@NTU, NTU Co-PI, P4.1 PI, February 2021 - February 2024)
Project 3:
Design of Security Functionality for V2X Networks (A*STAR IAF
PP, NTU Co-PI, WP5 PI, November 2019 - October 2022)
Project 4: V2X Network-Enabled Traffic Analysis and Smart Traffic
Signal Control for Large Traffic Networks (A*STAR IAF PP, NTU Co-PI,
WP4 PI, November 2019 - October 2022)
Abstract: Mobility,
literally speaking, refers to the ability to move from one location to
another. It is key to the functioning of a livable and sustainable
community. An efficient people-mover system is the backbone of a smart
city that has been drawing worldwide attention. In anticipation of
population growth and demographic changes, it is vital to develop an
integrated and sustainable transport system that meets the diverse
needs of the burgeoning population. The key lies in our ability to
harness the capabilities of information communication technologies,
factor resilience into infrastructure planning and management, retrofit
existing infrastructure to promote greener commuting modes and develop
innovative technologies in a timely manner to respond to people of all
age groups. With all advanced transport related technologies achieved
so far, fundamentally we are still facing the main challenge of how to
ensure safety, comfort and affordability in terms of travel (time or
money) cost, environmental footprint, and social impact. In this
project we shall address two key problems from a systems and control
perspective, i.e., how to discover and understand people’s travel needs
and commute patterns at a societal level via traffic analysis and
prediction, and how to use traffic signal control, which is essentially
a group control mechanism, to enhance safety, comfort and affordability
of daily travels in a large complex traffic network.
Past Projects
Project 1: Resilient Task Planning for Layout Changes in Smart Manufacturing (NRF Delta-NTU Corp. Lab, SMA Track PI, SMA Project PI, January 2020 - December 2021)
Project 2: Hierarchical Modelling and Real-time Operation Planning for Low Volume High Mix Reconfigurable Manufacturing (NRF Delta Corporate Lab@ NTU, SMA Track PI, SMA-RP2 Project PI, July 2016 - June 2021)
Abstract: In this project we
shall address both modelling and optimization challenges related to low
volume high mix manufacturing, and overcome the drawbacks of existing
modelling and optimization techniques, i.e., the lack of
computationally efficient real-time scheduling tools, which guarantees
requirement compliance or behaviour correctness, and the lack of
modelling tools that can automatically adapt to new system
functionalities and user requirements after system reconfiguration. For
this purpose we aim to achieve the following objectives: (1) to develop
a hybrid hierarchical modelling framework, which combines both
finite-state automata and programming models to ensure expressiveness,
verifiability, modularity and computational friendliness for low volume
high mix reconfigurable manufacturing. (2) To
develop
a novel modeling and optimization framework, which
facilitates
drag-and-play functionalities in smart manufacturing.
Project 3: Scalable and Smart Building Energy Management Systems (NRF BCA, GBIC, PI, July 2016 - June 2019)
Abstract:
Singapore
has mandated aggressive energy efficiency targets to be
realized by 2030 – 35% improvement over 2005 levels (set by Sustainable
Singapore Blueprint), and 80% of buildings Green Mark certified (set by
the Building and Construction Authority). In this project we propose an
intelligent, efficiency centric BEMS architecture inspired by modern
Computer Science methods in data analytics, inference, and scheduling.
Our architecture embeds intelligence in the BEMS and realizes
significant energy savings. It is distributed, computationally
efficient, and scalable. We expect that our strategy will help realize
Singapore’s aggressive targets for energy efficiency, while respecting
human thermal comfort and air quality constraints. Most critically, we
will argue that our solution enables rapid and cost-effective
deployment both in new buildings and for retrofits. It is adaptive to
occupancy and environment changes and robust by supporting fault
detection and isolation, and learns from all available sensor data.
Project 4: Distributed Model Predictive Routing and Scheduling for Minimizing Network-wise En-Route and Airport Arrival Delay in Air Traffic Flow Management with an Eulerian-Lagrangian Flow Dynamic Model (CAAS, PI, June 2015 - June 2018)
Abstract: The goal of this research is to develop a distributed model-predictive air flow management approach (in terms of flight routing and scheduling), which is applicable to a large-scale air traffic network, and can mitigate the network-wise en-route and airport arrival delay caused by reduction of link capacities during abnormal situations such as severe weather in some regions or airspace restrictions due to military operational needs.
Project 4: Distributed Model Predictive Routing and Scheduling for Minimizing Network-wise En-Route and Airport Arrival Delay in Air Traffic Flow Management with an Eulerian-Lagrangian Flow Dynamic Model (CAAS, PI, June 2015 - June 2018)
Abstract: The goal of this research is to develop a distributed model-predictive air flow management approach (in terms of flight routing and scheduling), which is applicable to a large-scale air traffic network, and can mitigate the network-wise en-route and airport arrival delay caused by reduction of link capacities during abnormal situations such as severe weather in some regions or airspace restrictions due to military operational needs.
Project 5: Energy Aware Heterarchical Task Routing and Scheduling in a Complex Material Handling System (MoE Tier 1, PI, March 2014 - February 2017)
Abstract: We will develop a formal modelling formalism and a corresponding heterarchical routing and scheduling approach, which aims to improve performance of an MHS while reducing the incurred energy consumption and environmental impact. This approach should remain effective for a large complex system (i.e., high scalability), allow multiple routes and schedules within a strategy as long as they lead to the same performance (i.e., high flexibility), and be able to undertake real-time rerouting and rescheduling to cope with changes in the system caused by component failures or client requirement adjustments (i.e., high adaptability). A simulation tool for multi-vehicle scheduling for material handling can be found here.
Project 6: Demand-Response in Building HVAC Scheduling (NRF Singapore-Berkeley Building Efficiency and Sustainability in the Tropics (SinBerBEST), NTU PI, January 2012 - March 2017)
Abstract: Singapore has mandated aggressive energy efficiency targets to be realized by 2030 – 35% improvement over 2005 levels (set by Sustainable Singapore Blueprint), and 80% of buildings Green Mark certified (set by Building and Construction Authority). Energy efficiency measures, however, must guarantee thermal comfort and indoor air quality constraints. Any viable approach must be cost-effective in deployment to diverse building stock, and be scalable to large structures with, e.g., 500 zones or more. We propose a novel computationally efficient, scalable air distribution scheduling/control approach inspired by Computer Science methods. We expect that our strategy will realize Singapore’s aggressive targets for energy efficiency in variable air volume (VAV) ACMV systems (depicted in Figure 1), while respecting human comfort and air quality constraints. Most critically, we will argue that our solution enables rapid and cost-effective deployment both in new buildings and for retrofits. It is adaptive to occupancy and environment changes, and robust by supporting fault detection and isolation.
Project 7: Electrical Machine Design for More Electrical Aircraft in Hostile Environment (NRF Rolls Royce Corporate Lab @ NTU, NTU PI, August 2013 - July 2018)
Abstract: A conventional aircraft typically has multiple secondary power distribution systems, namely the mechanical, hydraulic, pneumatic and electrical. Consequently, this leads to a complex power distribution. Moreover, each of them is required to have an appropriate redundancy for maintaining reliability. To overcome these shortcomings, the manufacturer of the aircraft in recent years trends towards the More Electric Aircraft (MEA) concept. The MEA has the key advantages of reducing the complexity, improving the efficiency and reliability. Boeing’s 787 Dreamliner and Airbus’ A380 are both successful examples where the more electric technologies have been adopted. In this project we will design an APU within a confined space in an aircraft, where high temperature operation and cooling issues will be key factors in determining which aspects of electromagnetic design to optimize. The electrical machine in the APU will be optimized through design and analysis to achieve high efficiency, high gravimetric and volumetric power density. The thermal property and robustness issues will be taken into consideration. To minimize the drag losses when operating the machine at high-speed, the structures of the rotor will be optimized.
Project 8: Intelligent Power Allocation and Scheduling for More Electrical Aircraft (NRF Rolls Royce Corporate Lab @ NTU, NTU Co-PI, August 2013 - July 2018)
Abstract: The concept of More-Electric Aircrafts (MEA) emphasizes on the utilization of the electrical power to power the non-propulsive aircraft systems. The major advantages of having MEA are the optimization of aircraft performance and reduction in operating and maintenance costs. Also, MEA enables the reduction in pollutant emission, easing the environmental concerns caused by air travel. However, MEA poses challenges to the aircraft electrical system. This is because conventional aircraft power distribution and management system manage only to present demands. Due to the increasing number of electrical loads, the dynamics of power flow on the electrical bus has been significantly altered. For example, the power electronics loads may produce regenerative power equal to the peak power for the durations of 20 – 200ms and peak-to-average power ratios may exceed 5-to-1 across 50-500ms. This dynamic characteristic of electrical loads will affect the reliability of power system. Also, without effective power management strategy, more auxiliary power sources (energy storage device), such as battery or super-capacitor, are required to improve the system performance when electrical loads demand high power in a short time. This leads to over-provision of storage capacities in order to allow for multiple concurrent loads and future technology insertion. These challenges show the need of intelligent power management system in optimizing the power management of a civil aircraft’s power system. In this project, an intelligent power management for civil aircraft will be developed, based on a much smaller scale of power system architecture. The main objective is to develop an intelligent power management system that is scalable and flexible to be applied to systems with different levels of complexity.
Project 9: Task Model Construction in Interface for Human-Robot Interaction (A*STAR Industrial Robotic Program, NTU Co-PI, September 2013 - August 2016)
Abstract: The gola of this project is to develop techniques that can automatically transform informal user requirements input from user friendly HMI into formal ones which is suitable for task model construction and verification, and automatically generate and verify a task reference model, which ensures safety and operational optimality. The techniques shall allow a user to online update task requirements and see the resulting task reference model, which shall enhance human-machine interaction.
Project 10: Distributed Adaptive Urban Traffic Signal Control based on V2X Information Infrastructure (EDB NTU-NXP Smart Mobility Test Bed Program, NTU Co-PI, April 2015 - March 2019)
Abstract: Traffic congestion has become one of the major challenges for metropolitan growth. In addition, the waste exhaustion from massive vehicles trapped on the road also imposes environmental issues, which, for example, have caused significant social and health problems in China. In this project we will investigate how to improve traffic situations in a densely populated region by using intelligent traffic control with Vehicle-to-infrastructure (V2X) information technologies. In the first stage of research, we aim to develop a distributed traffic-responsive scheduling architecture for urban traffic signal control, which consists of a set of local processing centres running in parallel that communicate with each other for neighboring traffic information to refine their own local traffic signals. It takes real-time traffic measurements via road sensors and/or V2X communication infrastructure, estimates vehicle driving patterns (such as speeds, turning ratios and link densities), and generates the corresponding green time schedule for each junction aiming for minimizing the network-wise total waiting time. To deal with traffic uncertainties, a model-predictive strategy will be adopted. In the second stage, synthesized traffic-responsive traffic light schedules and the corresponding predicted link speeds are sent back to individual vehicles via V2X, which affect vehicles future route plans. Such updated route plans will be fed back to the traffic light control centers via V2X to enhance existing traffic light schedules. Our ultimate goal is to form a closed loop between traffic light schedulers and traffic signal end users (i.e., vehicles), whose joint efforts will eventually lead to an intelligent and highly adaptive road traffic management system.
Project 11: Distributed Model Predictive Routing and Scheduling for Minimizing Network-wise En-Route and Airport Arrival Delay in Air Traffic Flow Management with an Eulerian-Lagrangian Flow Dynamic Model (CAAS, NTU PI, April 2015 - March 2018)
Abstract: The goal of this research is to develop a distributed model-predictive air flow management approach (in terms of flight routing and scheduling), which is applicable to a large-scale air traffic network, and can mitigate the network-wise en-route and airport arrival delay caused by reduction of link capacities during abnormal situations such as severe weather in some regions or airspace restrictions due to military operational needs.
Project 12: Integrated Fibre Optic Sensor Based Monitoring System for LNG Terminals (NRF EIRP3 Gas Technology Grant Call, NTU Co-PI, May 2015 - April 2018)
Abstract:
The objective
of the proposed project is to develop a monitoring system using
integrated fibre-optic sensors for safe operation of LNG terminals. The
focus of the research works in this project are: 1) develop simulation
models to investigate rollover and leak characteristics for LNG tanks
and gas pipelines; 2) develop the advanced low temperature polyimide
(PI)-coated Fibre Bragg Grating (FBG) sensors and online calibration
method for single end fibre-optic Raman Distributed Temperature Sensors
(RDTS); 3) integrate RDTS and FBG sensing sub-systems to a LNG tank
safety monitoring system; and 4) develop risk assessment methods for
LNG monitoring system based on the developed models and measurement
data to evaluate potential incidents.