Centre for Intelligent Systems Research

CISR Research Seminar Series - 2011

  CISR presentation
blue star Professional development
yellow star Keynote lecture
purple star External presentation
red star No presentation

Seminars will be held at 12pm in the CISR Breakout Area (except where otherwise indicated)

Date Presenter Presentation/topic
Monday 12th December Husaini Aza Mohd Adam Haptic Visualisation and Rendering of 2D Images 

 Abstract

This research project investigates the ability for human participants to feel the visual information contained within 2D visual art. Haptic interaction is utilised as the basis for the development of mapping algorithms allowing participants to feel the visual information contained within 2D images. The human user will grasp a haptic interface with one or both hands and then 'feel' forces and vibrations representing colour, intensity and other visual elements of 2D images. The ability of visually impaired and able sighted users to perceive the visual information using the mapping algorithms and haptic interfacing will be evaluated.

Monday 5th December Imali Hettiarachchi Analysis of fMRI/EEG data to identify functional differences of brain 

 Abstract

Functional magnetic resonance Imaging (fMRI) and Electroencephalography (EEG) are non-invasive functional imaging modalities which records the brain dynamics. There are neurophysiology inspired models introduced for simulating brain's electrical activity imaged through EEG data and fMRI data. Dynamic Causal Models (DCM) is such a generative model consisting of a neuronal connectivity model and a forward model to how fMRI/EEG measurements are generated via source dynamics. DCM for EEG basically investigates how the coupling among brain regions is influenced by changes in the experimental context. The research will be focused on a DCM based analysis of the fMRI/EEG data to identify functional differences of the brain. In particular we will be looking at how different brain regions will be functionally integrating during a specific task. The DCM's have an input-state-output formulation and are well suited to use with Bayesian estimation techniques for fMRI/EEG data analysis. In the present work we use powerful nonlinear estimation technique called Sequential Monte Carlo (SMC) methods, also referred to as Particle Filters (PF). We also study incorporating other Bayesian techniques such as Markov Chain Monte Carlo (MCMC) adjoined with SMC methods to develop a robust estimation platform for fMRI/EEG data analysis. State filtering in nonlinear models similar to the DCM formulation is a well-studied issue. However, joint estimation of states and parameters of the model is still challenging due to the high number of parameters involved. As there is no ground truth data available for brain data, simulation results are used for the validation. Simulations for different input stimulations have shown robust estimation results which shows a high potential to be applied for the experimental data.

Monday 28th November James Zhang Development of an effective passenger screening algorithm 

 Abstract

Aviation security is an essential part of national transport security. Passenger screening is more involved than checked bag screening: it combines both hand luggage screening and body screening. Effective algorithms need to be developed to meet the challenge of stipulatory requirements under technological and budgetary constraints. In this research admission control algorithms from queuing theory are applied to a passenger screening system to make better use of the screening resources. The proposed method extends the existing algorithms in several ways. It allows for dynamic reclamation of screening capacity during operation and it is less susceptible to saturation problems in the fixed capacity formulation. The algorithm is tested in a realistic Quest simulation model under various scenarios and its effectiveness is verified.

Monday 21st November Isaac Winter Augmented Collision Detection using Stereo Imagery in an Unstructured Environment 

 Abstract

Stereoscopic teleoperation of remotely manipulated robots is a well-defined and researched field. However, the stereo cameras necessitated by this system can be utilised for much more than simply providing stereoscopic video for the operator. The stereo cameras can allow a 3D reconstruction of the environment to be generated from the video feed that can then be used for simulation with a kinematic model of the robot. This research is focusing on the development of a novel framework that, using stereo cameras already present for stereoscopic teleoperation, enables the real-time simulation of a teleoperated robot's movements in a virtual representation of the robot's environment. This simulation of the robot's movements in the virtual environment would enable the prediction of possible collisions and undesired interactions with the environment, therefore allowing the operator to be alerted to potential dangers before they occur.

Monday 14th November Lei Wei Function-based Visual and Haptic Rendering, Interaction and Collaboration in Shared Virtual Spaces 

 Abstract

Commonly, surface and solid haptic effects are defined in such a way that they hardly can be rendered together. A method for defining mixed haptic effects including surface, solid, and force fields is therefore proposed. These haptic effects can be applied to virtual scenes containing various objects, including polygon meshes, point clouds, impostors, and layered textures, voxel models as well as function-based shapes. Accordingly, a way how to identify location of the haptic tool in such virtual scenes as well as consistently and seamlessly determine haptic effects when the haptic tool moves in the scenes with objects having different sizes, locations, and mutual penetrations is also proposed. To provide for an efficient and flexible rendering of haptic effects, the author also proposes to concurrently use explicit, implicit and parametric functions, and algorithmic procedures. The above mentioned framework could also be extended to shared environments for haptic collaboration and tele-operations.

Friday 11th November Prof. Madjid Fathi External presentation  A new trend in Knowledge Management and -Integration using Rule Based Web 

 Biography

Madjid Fathi is a professor and Director in the Department of Electrical Engineering and Computer Science at University of Siegen. He is the director of Institute of Knowledge Based Systems and Knowledge Management (KBS & KM). He is also the founder and director of Research Center for Knowledge Management and Intelligent Systems (KMIS). His research interests are focused on Knowledge Management applications in Medicine and Engineering, Computational Intelligence and Knowledge Discovery from Text (KDT).

Madjid is the editor of Integrated Systems, Design and Technology published by Springer (2010) and also 5 edited books. He has more than 200 publications including 2 text books and 21 Journal publications, 4 paper awards and 170 platform presentations. Currently he has 2 books under preparation. He is a senior member of IEEE and member of editorial board of 5 respective journals. His latest keynote speeches since 2008 were at IEEE-EIT2008, IEEE-EIT 2009, KMIS 2009, IEEE-SMC (UK) 2009, KMI 2010 & 2011, IEEE-EIT 2011. He was a panellist at IEEE-IRI 2007, IC3K 2009, iiWAS2009.

 Abstract

Besides existing technologies, Small and Medium-sized industries require novel approaches and integrated components to facilitate decision making and knowledge transfer. Knowledge Management integration in decision and problem solving endeavours provide a basis for utilizing human- and process oriented indicators towards optimizing organizational management process. We, humans, have the ability to distinguish not only between irrelevant and unimportant features but also between the important and essential ones. Knowledge browsing, retrieval and discovery (e.g. Knowledge Discovery in Databases - KDD and Knowledge Discovery from Text - KDT) through existing web technologies are arising challenges and complexities. Intelligent processing of contents and human competencies needs innovative and adaptive strategies. In this context, Rule-based Web empowers integrating semantic techniques i.e. Rule-based techniques for complex decision making, multi-lateral indicators and human collaboration. This practice-oriented strategy (entitled as WEB.X) is currently investigated at our research center KMIS particularly in knowledge transfer for Alzheimer Patients in cooperation with University-Hospital Cologne, and for adaptive process control in semiconductor manufacturing.

Monday 7th November

12:00pm
GTP Conference Room
(na1.014)
Prof. Hamid Garmestani External presentation  Application of Materials Design and Materials Informatics in Nano-Technology 

 Biography

Prof. Hamid Garmestani, faculty since 1991, is a Professor of Materials Science and Engineering at Georgia Institute of Technology. Dr. Garmestani is well known for developing methodologies in Microstructure Sensitive Design (MSD) framework that addresses an inverse methodology and innovations in various aspects of processing, structure-property relationships, and simulation-based design of materials. He has contributed years of work to statistical continuum mechanics for homogenization in composites and polycrystalline materials. He has recently been involved in Microstructure design of porous ceramic cathode materials resulted in the development and synthesis of porous gradient LSMO cathode microstructures for Solid Oxide Fuel Cell (SOFC). The theoretical modeling using statistical approaches provided a framework to identify the gradient and the required porosity in the ceramic structures. He and his team have also developed a synthesis methodology based on spray pyrolysis to fabricate the optimum gradient microstructure. His efforts in the microstructure design and processing of Carbon Based nano-tube composites has just resulted in a microstructure with superior mechanical properties (patent pending). Earlier in 2001, he has developed techniques in magnetic processing of Carbon Based Nano-tube composites using Shear Drag in the polymer matrix to produce highly oriented nano-tube composites using high magnetic field.

Professor Garmestani has been heavily involved in leadership roles in both the American Society of Mechanical Engineers Materials Division (Associate editor of ASME JEMT). He is a member of the texture, forming and composite committees of ASM and TMS. Dr. Garmestani has organized more than 26 workshops and symposia in the emerging subject of materials design (average of 3 in the last four years), He was awarded "Superstar in Research" by FSU-CRC in year 2000. He was also the recipient of the 2000 Engineering Research Award of the FAMU-FSU College of Engineering and recipient of the Faculty Award for Research from NASA. He is presently funded through DOE, Pacific Northwest National lab, and the NSF. Dr. Garmestani is a member of the editorial board of the Journal Engineering Materials and Technology, International Journal of Plasticity, Journal of Mechanics of Materials, Computers, Materials and Continua and Theoretical and Applied Multi-scale Modeling of Materials.

 Abstract

The field of materials and microstructure design has progressed significantly in the past two decades. Such efforts have been applied to a number of materials modeling and design efforts in multi-scale modeling and homogenization. The importance of such efforts in linking nano-structures to micro and macro scale efforts have been applied to a number of multi-physics problems with important applications in energy and advanced technology. Along one direction, materials selection has emphasized search of databases for properties or characteristics of responses that best suit a set of specified performance indices, often employing data mining and combinatorial search methods. Another stream employs modeling and simulation to support decision-based design, exploiting computational materials science and physics. The integration of computational materials science, computational materials design and simulation plus data processing in engineering systems design is central to present efforts in materials genome. Such efforts involve top-down and bottom up approaches. In systems design approaches, ranged sets of performance requirements are specified that place demands on multiple properties/responses of materials, often requiring more complex and sophisticated ways of trading off properties and responses than can be described by simple metrics. In this paper we will examine some of these efforts in top-down approaches as a basis for materials design on a nano and microscales. Examples are provided in energy production and in applications in Fuel Cells, solar cells and bio-materials.

Wednesday 26th October Khashayar Khoshmanesh Professional Development  Some tips to write journal papers - Part 2 

 Abstract

The ability to represent research results in an academic manner is crucial to a successful PhD, and more importantly for a successful research career. In this two-part seminar series, Khashayar will share his experience in writing conference papers and high-quality journal papers. He will talk about publication strategies, paper structure, presentation of results and the appearance of a good paper. Also, as a journal reviewer in his field, Khashayar will discuss the hot spots of a journal paper, which should be considered very carefully in order to best represent your research to reviewers and journal editors. Finally, Khashayar will show how to devise an appropriate strategy to manage your time and produce high quality publications.

Monday 24th October Bruce Gunn Combining Stochastic and Process models, in order to simulate Casthouse systems 

 Abstract

The scheduling of metal to different casters in a casthouse is a complicated problem, attempting to find the balance between pot-line, crucible carrier, furnace and casting machine capacity. In this presentation, a description is given of a casthouse modelling system designed to test different scenarios for casthouse design and operation. Using discrete-event simulation, the casthouse model incorporates a range of stochastic distributions for metal carriers, crucible movements, caster operation and furnace conditions. During initial model development, it was highlighted that furnace temperature is a dynamic variable that affects the system behaviour. In order to develop a credible model, the presentation will highlight how a stochastic model and a temperature model were combined to mimic production data.

Results from the model will be presented for a case study, which highlights the effect different parameters have on overall casthouse performance. This case study uses past production data from a casthouse to validate the model outputs. The results suggest that along with metal preparation times and caster strip-down/setup, the temperature evolution within the furnaces is one key parameter in determining casthouse performance.

Wednesday 19th October Khashayar Khoshmanesh Professional Development  Some tips to write journal papers - Part 1 

 Abstract

The ability to represent research results in an academic manner is crucial to a successful PhD, and more importantly for a successful research career. In this two-part seminar series, Khashayar will share his experience in writing conference papers and high-quality journal papers. He will talk about publication strategies, paper structure, presentation of results and the appearance of a good paper. Also, as a journal reviewer in his field, Khashayar will discuss the hot spots of a journal paper, which should be considered very carefully in order to best represent your research to reviewers and journal editors. Finally, Khashayar will show how to devise an appropriate strategy to manage your time and produce high quality publications.

Monday 17th October Mats Isaksson Analysis and Synthesis of Parallel Manipulators with Rotation-Symmetric Arm System 

 Abstract

Parallel manipulators with rotation-symmetric arm system possess all the typical advantages of parallel robots, such as high acceleration, high-accuracy and high stiffness. Additionally, the rotation-symmetric arm system leads to a large workspace in relation to the manipulator footprint. This presentations reviews previously proposed manipulators in this category. One subclass of these manipulators has additional good qualities like low inertia and high resonance frequencies. These favorable qualities are achieved using only 5-DOF passive links and by mounting all actuators on the non-moving base of the manipulator. The common feature of previously proposed manipulators of this subclass is identified and several novel 3- and 4-DOF members are derived.

Monday 10th October Wael Abdelrahman VirtualGRASP: A Case Study for Multi-Point Haptics 

 Abstract

In this talk, we discuss a novel multi-point haptic study that enable the user to interact and grasp deformable models accurately. The multipoint interaction is done through a custom haptic gripper that extend the number of interaction points to two. The system is physically loyal as it incorporates a data-driven algorithm that replicates real objects material behaviour using artificial neural network (ANN). The system can perform in a real time manner as the material model is built offline and online calculations are limited to the simulation of the trained ANN.

Monday 3rd October Matthew Watson Reality, Enhanced: Towards vision based object recognition for Augmented Reality 

 Abstract

Augmented Reality (AR) seamlessly enhances a user's reality by registering virtual, computer-generated content interactively into their real time perception of freespace. A true AR system locates specific placeholders in freespace to register virtual information upon. The way pose is determined differentiates Augmented Reality (AR) as a subset of Mixed Reality (MR), though both terms are often mistaken to be interchangeable. This presentation overviews the current trends of vision based tracking and recognition, and explores my current research in precise object recognition and pose estimation for real time systems.

Monday 26th September Michael Johnstone Parallel Calculation of Order Statistics on GPU's 

 Abstract

High breakdown regression techniques, while robust against data outliers, are often computationally expensive. Graphical processing units (GPU) offer a means to alleviate this burden. In this work we compare alternative methods for the parallel calculation of order statics on a GPU and introduce an improved method based on the minimisation of a convex function.

Monday 12th September Shady Mohamed Optimal data fusion for nonlinear systems with correlated noises 

 Abstract

This research is concerned with the data fusion problem for systems suffering from the possibility of missing measurements. The systems under consideration are nonlinear. The required linearisation takes place before the state estimation and data fusion is provided. This researh presents the recursive data fusion filter for measurements obtained from two sensors. Two scenarios are considered, when both sensors lose the measurements in the same random manner and when each sensor has its own data loss rate. The noise covariance in the observation process is allowed to be singular which requires the use of generalized inverse. The noise between both sensors is assumed correlated and the means to compute this correlation is provided.

Monday 5th September Marwa Hassan Aly Hassan Prediction Interval Modelling for Heteroskedastic Processes 

 Abstract

GARCH time series model has played a very vital role in treating heteroskedasticity of data errors, when a long-time horizon forecast is required. GARCH is a successful forecasting technique for conditional variances and volatility changes combined with time series data analysis . The GARCH model succeeded in addressing the problem of heteroskedasticity of the data, which was not explained by prior regression models, such as ordinary least square regressions. This allowed for a remarkable contribution in the field and proven of a great interest, especially in the field of financial applications.

The purpose of this study is to quantify heteroskedasicity and improve the accuracy of prediction interval modelling. A novel heteroskedasticity index was introduced to measure how heteroskedastic the data is and the applications are to be extended to various types of heteroskedastic data.

Monday 29th August Hamid Abdi Performance analysis of constrained human motions and fault-tolerant robotics 

 Abstract

In this talk, at first a fault-tolerant control strategy is introduced based the behaviour of humans with constrained motions and mathematical methods of fault tolerance. The aim is to add the advantages of human behaviours dexterity to the mathematical methods of fault tolerance. In order to do this, a neural network is trained from the difference between the joint profiles of a human arm and a robotic arm. This neural network is then used to introduce the fault tolerance strategy.

Then, the talk will continue by introducing a design methodology for optimal fault-tolerant manipulators. A class of optimal Jacobian matrices based on geometric properties are proposed. This class of Jacobian matrices are equally fault-tolerant with regard to worst case relative manipulability and worst case dexterity of manipulators. The optimal Jacobian matrix is achieved by symmetric distribution of n+1 points in an n dimensional sphere.

Monday 22nd August Sara Farag Towards a Parameterless 3D Mesh Segmentation 

 Abstract

Segmenting 3D meshes into distinct components improves both the perception and processing of these meshes. The smaller segments are usually easier to process and can be associated with semantics or geometric features. This can be used in 3D parametrization, animation, deformation transfer and many other 3D graphics applications. This work focuses on the generation of automatic 3D segments, which requires no object dependent parameters and only tuning parameters that can be fixed for different object classes. The automatic method assumes the segmentation is done based on the minima rule of the cognitive theory because it is considered the closest to human approach. Automating such a process is challenging due to the variety and complexity of the input. Most of the literature requires many object specific parameters to achieve their goals. Examples of used parameters are segment count, segment size threshold and initial seeds on certain vertices. This requires the presence of the user in the loop and he is required to do some manual processing before the automatic technique can start. The proposed method transforms the problem into the 2D domain and calculates the required cut lines by processing different pose projections and fusing the results.

Friday 19th August

9:15am
GTP Conference Room
(na1.014)
Robert Reed External presentation  How to make your first million dollars through innovation - from a practitioners perspective 

 Biography

Robert has a Bachelor of Commerce and Business Administration (Deakin University) 1988.

Having gained extensive experience by working in several "grass roots" positions in Express Promotions, while completing his degree, Robert took on the role of General Manager in 1989 and Managing Director in 1997. When Robert took on the General Manager's position, Express Promotions consisted of two small geographically separate companies, both with an uncertain future. They employed a total of 35 people, and turned over less than $5 million between them.

Today, Express Promotions and FE Technologies are strong, consolidated companies that are world competitive in their fields, with a turnover of $20 million and with 80 employees.

Robert's current role as CEO includes responsibility for managing an efficient organization with effective reporting structures, and appropriate training and development programs. Some of Robert's key achievements in the organisation include forming critical alliances with key library LMS providers, and RFID equipment component suppliers. Robert is the driving visionary behind FE Technologies' RFID success in the library arena, and more recently has spearheaded our successful ventures into retail RFID applications and new fields such as RFID vehicle tracking and chipless RFID technology development.

Monday 15th August Vu Le Proactive Task Assignment for Labour Control in Manufacturing 

 Abstract

Optimising the utilisation of skilled and non-skilled worker is a preliminary requirement in order to improve the efficiency and productivity in manufacturing. This is a challenging management task that requires a detailed analysis of different scenarios, a comprehensive understanding in the capability of individual personnel and specification of resource availability. This study examines a supervisor labour control strategy with real-time job assignment that optimises worker task execution and minimise processing error. The control manages multi-labour groups and multiple work cells. It utilises a Closest Contractor (CC) selection rule from the global position information of every individual resource inside a facility to assign the next job to an appropriate worker. A job consists of multiple tasks that are required to carry out. The controller interprets the requirement of these tasks, it provides information about a task to the worker and derives navigation path to this task. The control strategy described in this paper contains a reactive real-time job assignment and path navigation features, which can be used to enhance the existing Material Resource Planning (MRP), Manufacturing Resource Planning (MRP2), Enterprise Resource Planning (ERP) and the Just In Time (JIT) system in manufacturing environments.

Monday 8th August Prof. Anthony Maciejewski External presentation  Kinematically Redundant Robots: The Promise of Human-Like Dexterity 

 Biography

Anthony A. Maciejewski was born in Cleveland, Ohio on July 19, 1960. He received the B.S.E.E (summa cum laude), M.S., and Ph.D. degrees in Electrical Engineering in 1982, 1984, and 1987, respectively, all from The Ohio State University under the support of a National Science Foundation (NSF) graduate fellowship. From October of 1985 to September of 1986 he was an American Electronics Association Japan Research Fellow at the Hitachi Central Research Laboratory in Tokyo, Japan where he performed work on the development of parallel processing algorithms for computer graphic imaging and simulation.

In 1988, Prof. Maciejewski joined the faculty of Purdue University as an Assistant Professor of Electrical Engineering. He was promoted to Associate Professor in 1993 and Full Professor in 1998. In August of 2001 he joined Colorado State University where he is currently a Professor and Head of the Department of Electrical and Computer Engineering. His research and teaching interests center on the analysis, simulation, and control of kinematically redundant robotic systems. His current work focuses on how kinematic redundancy can be utilized to design failure tolerant robotic systems for remote operations. He directs a research laboratory and has developed graduate courses in these areas. His commitment to education resulted in his receiving four teaching awards. His research has been supported by NSF, Sandia National Laboratories, DARPA, NASA, National Imagery and Mapping Agency, Missile Defense Agency, Non-lethal Technology Innovation Center, the NEC Corporation, Caterpillar, AT&T, H-P, Intel, Wolf Robotics, and the TRW Foundation.

Prof. Maciejewski has served on the editorial boards for the IEEE Transactions on Robotics and Automation; IEEE Transactions on Systems, Man, and Cybernetics (Part A); Intelligent Automation and Soft Computing; Pattern Analysis and Applications; Journal of Robotics; Journal of Automation and Mobile Robotics; International Journal of Robotics and Automation and was co-guest editor for a special issue on Kinematically Redundant Robots for the Journal of Intelligent and Robotic Systems. He is a Fellow of the Institute of Electrical and Electronics Engineers "for contributions to the design and control of kinematically redundant robots," and has served as the VP for Financial Activities, AdCom member and Secretary for its Robotics and Automation Society. He was Technical Program Co-chair for the 1997 International Conference on Robotics and Automation (ICRA), Technical Program Chair for ICRA 2002 and has served on over 60 other conference program committees. He is also a member of the Association for Computing Machinery.

 Abstract

The vast majority of robots in use today operate in very structured environments, e.g. in factory assembly lines, and possess only those limited motion capabilities required to perform specific tasks. While these robots can outperform humans in terms of speed, strength, and accuracy for these tasks, they are no match for the dexterity of human motion. Part of a human's inherent advantage over industrial robots is due to the large number of degrees of freedom in the human body. Articulated, i.e. jointed, motion systems that possess more degrees of freedom than the minimum required to perform a specified task are referred to as kinematically redundant. In an effort to mimic the dexterity of biological systems, researchers have built a number of kinematically redundant robotic systems, e.g. anthropomorphic arms, multi-fingered hands, dual-arm manipulators, and walking machines. While these systems vary in their appearance and intended applications, they all require motion control strategies that coordinate large numbers of joints to achieve the high degree of dexterity possible with redundant systems. This talk will discuss the issues that arise when designing such strategies, frequently drawing on the use of the singular value decomposition, including the characterization of redundancy, the quantification of dexterity, and the development of efficient and numerically stable motion control algorithms that simultaneously optimize multiple criteria. In addition, the ability of kinematically redundant robots to sustain component failures and yet still complete an assigned task will addressed, thereby extending the application of robots to environments that are unacceptable or inaccessible to humans.

Monday 1st August Mojdeh Nasir Modelling Pedestrian Wayfinding Behaviour under Normal Conditions 

 Abstract

This presentation focuses on research into pedestrian wayfinding behaviour in public environments during non-panic situations. Wayfinding refers to the cognitive and behavioural abilities of an agent to follow a path from the current location to the destination. A control theory approach from cognitive science perspective, to devise a model will be discussed. A utility-based framework that considers uncertainty is proposed. The aim is to generate an accurate prediction of pedestrian trajectories between two points in a familiar environment. As pedestrian interaction with the environment is an important feature of wayfinding behaviour, we study the spatial effects of the environment on pedestrian wayfinding patterns.

Monday 25th July Kyle Nelson Vision-Based Geometric Measurement via Uncalibrated PTZ Cameras and Laser Projection 

 Abstract

Despite the range of laser-camera measurement systems available none are sufficient to automatically measure individual geometric features during part manufacture without time-consuming complete 3D reconstruction. This seminar introduces a novel in-process geometric measurement system comprising uncalibrated Pan-Tilt-Zoom (PTZ) cameras and a laser scale projection. The use of high-zoom cameras and a structured laser scale feature allows accurate and repeatable performance from a distance of up to 9m. This research discovered that the projection of a predefined laser pattern in the immediate vicinity of interest features to assist with size and location estimation is viable measurement technique. Experimental results show automatic and detailed analysis of part geometry to an accuracy of approximately 1mm.

Monday 18th July Chintha Handapangoda Modelling of Laser Doppler Flowmetry 

 Abstract

Laser Doppler Flowmetry (LDF) is a new method of continuous and noninvasive measuring of microcirculatory organ perfusion (blood flow), utilizing the Doppler shift of laser light as the information carrier. LDF is useful in the clinical assessment of blood flow in disciplines such as dermatology, plastic surgery and gastrointestinal surgery. Multichannel flowmeters have enabled simultaneous perfusion monitoring of several organs. Existing Laser Doppler flowmeters, however, assume a linear relationship between the output of the meter and the blood flow rate and hence do not provide accurate measurements for high flow rates such as that in kidneys. The objective of the proposed research is to model Laser Doppler flowmetry more precisely, taking into account the nonlinear effects due to multiple scattering, and to use pulse excitations that will result in a more accurate profile of the flow. This work involves the extension of governing stationary equations such as the photon transport equation to moving media together with a coupling to the Doppler shifted signal.

Tuesday 12th July

3:30pm
CISR Breakout Area
Ron Mack External presentation  Commercialisation Australia - A New Approach to Innovation 

 Biography

Ron has run a successful management consulting practice for more than 30 years, delivering a broad range of services to public and private sector organisations throughout Australia and overseas. His core skills are in business strategy, market research and technology commercialisation. He has also held interim CEO and directorship positions with technology start-ups, private and public companies, particularly in the food, agribusiness, electronics, engineering and manufacturing sectors.

Ron is currently contracted to Commercialisation Australia as one of their 20 national Case Managers. His role is to guide applicants through the grant process, assess applications on behalf of the CA Board, and work with grantees to ensure successful commercial outcomes. Prior to joining CA, he spent two years with CSIRO as Technology and Innovation Adviser, where he provided specialist advice on technological innovation for Australian small to medium enterprises, including the establishment of the Victorian Direct Manufacturing Centre. He has a Bachelor of Chemical Engineering (Hons) and Bachelor of Economics.

 Abstract

This presentation discusses the structure of Commercialisation Australia's grants programs, eligibility and merit criteria, and provides examples of local early-stage and established companies who are participating in this program. The CA program will be of interest to Victorian Direct Manufacturing Centre (VDMC) industry participants and research providers, commercialising direct manufacturing products, processes or other innovations.

Monday 11th July

CISR Haptics Lab
(na1.421)
John McCormick and Kim Vincs Using Motion Capture in Performance, Movement Analysis and Classification 

 Abstract

Motion Capture has traditionally been used mainly as a recording and reproduction tool, especially in the movie and game industries. Mocap is also a very useful tool for enabling interaction within simulation environments and for analysis of many aspects of human performance. In this seminar Kim Vincs and John McCormick will describe recent research into the use of motion capture to assist in recognition, classification and defining key signatures within complex dance movement. We will also present a short excerpt from a performance currently in development, featuring Erin, a dancer with the Melbourne Ballet Company.

Monday 4th July Zoran Najdovski  

 Abstract

Monday 4th July

9:30am
CISR Breakout Area
Prof. Chee Peng Lim External presentation  Computational intelligence-based systems and their real-world applications 

 Biography

Chee Peng Lim obtained the Bachelor of Electrical Engineering (1st class) degree from University of Technology, Malaysia in 1992, the MSc. (Eng.) in Control Systems (Distinction) and Ph.D. degrees, both from the University of Sheffield, UK, in 1993, and 1996, respectively.

His research interests focus on the design and development of computerized intelligent systems in the domains of pattern classification, data mining, fault detection and diagnosis, condition monitoring, medical prognosis and diagnosis, as well as manufacutring process optimization. He collaborates closely with researchers in the international arena, whereby he received the Australia Endeavour Executive Award, 2009, and Associaiton of Commonwealth Universities (ACU) Titular Fellowship, 2008 (both at University of South Australia, Australia); Strategic International Research Fund, 2007 (at University of Auckland, New Zealand); Commonwealth Fellowship, 2003 (at University of Cambridge, UK); Fulbright Scholarship, 2002 (at University of California, Berkeley, USA); and Japan Society for the Promotion of Science (JSPS) Research Fellowship, 2002 (at Kyushu University, Japan). The potential of his research work for commercialization has also been recognized, whereby he received a Gold medal and a Special Award at the British Innovation Show, UK, 2007, and a Gold medal at the Geneva International Exhibition of Inventions, Switzerland, 2006.

He has published more than 185 technical papers in journals, conference proceedings, and books. He has received 7 best paper awards at international conferences. He has also edited 2 books and 12 special issues in journals, and served in the editorial board of 5 international journals. He was a visiting professor at University of South Australia from July 2008 to July 2009, and is currently a professor at University of Science, Malaysia.

 Abstract

Computational Intelligence (CI) is a broad discipline that encompasses many different paradigms, e.g. artificial neural networks, fuzzy systems, evolutionary algorithms. In this seminar, the main characteristics of CI are introduced, and the benefits of utilizing CI-based methodologies for the design and development of computerized learning systems are explained. The applicability of CI-based systems to a number of real-world problems is exemplified and discussed, viz. condition monitoring of complex systems, failure mode and effect analysis of manufacturing processes, optimization of production schedule, minimization of product waste, as well as decision support of medical diagnosis and prognosis tasks. Results are presented to demonstrate the potential usefulness and effectiveness of CI-based systems in undertaking complex real-world problems.

Monday 27th June Ahmad Hossny Handling Uncertainty of Scheduling and Resource planning 

 Abstract

Many researches have tackled scheduling problems in different contexts for a variety of objective functions. Most of them state the assumption that input data is certain and fully trusted which is not matching with the reality where the data are not certain most of the time due to environment changes, unexpected behaviour, uncontrollable parameter, or error of input data. The target of such research is to adapt scheduling algorithms to possible uncertainty of data and to minimise the uncertainty harmful effect to the objective function and to decision maker.

The proposed research depends on modelling the uncertainty into mathematical (interval), probabilistic or fuzzy representation, and then update the scheduling algorithm to use the interval algebra or credible intervals or fuzzy intervals as replacement to standard numeric arithmetic. This is assumed to minimise the total uncertainty of the schedule and the uncertainty of the objective function. It is a bit difficult to provide solid mathematical proof that the proposed technique will minimise the uncertainty, so we depend on computational analysis by testing many examples with different scenarios and compare the results.

Tuesday 21st June

12:30pm
CISR Breakout Area
Khashayar Khoshmanesh Professional Development  Preparing high quality journal papers - Part 2 

 Abstract

The ability to represent research results in an academic manner is crucial to a successful PhD, and more importantly for a successful research career. In this two-part seminar series, Khashayar will share his experience in writing conference papers and high-quality journal papers. He will talk about publication strategies, paper structure, presentation of results and the appearance of a good paper. Also, as a journal reviewer in his field, Khashayar will discuss the hot spots of a journal paper, which should be considered very carefully in order to best represent your research to reviewers and journal editors. Finally, Khashayar will show how to devise an appropriate strategy to manage your time and produce high quality publications.

Monday 20th June Burhan Khan Haptically-Enabled Dental Training Simulator 

 Abstract

Technological advancement has had a profound effect on all spheres of our lives. It further develops with time as solutions are sought for the various obstacles encountered in the areas of science and education. One such example is the field of dentistry. In this field it is essential for students to gain sound practical experience before they are able to perform dental procedures on patients. Faculties of medicine and dentistry around the world are faced with the single challenge of training their students in this regard. Training procedures involve a great deal of practice and also needs a lot of resources like mannequins for jaws, teeth, gums, dental instruments and most importantly, the practice hours. Most of these resources are not reusable as they are made of non-renewable materials. It is for these reasons that Haptics has found a major application development in this field, thereby significantly reducing the hassles and related expenses. This presentation highlights the work done in this field to date, demonstration of the dental training simulator developed at CISR, its affordability and future developments and implementations.

Tuesday 14th June

12:30pm
CISR Breakout Area
Khashayar Khoshmanesh Professional Development  Preparing high quality journal papers - Part 1 

 Abstract

The ability to represent research results in an academic manner is crucial to a successful PhD, and more importantly for a successful research career. In this two-part seminar series, Khashayar will share his experience in writing conference papers and high-quality journal papers. He will talk about publication strategies, paper structure, presentation of results and the appearance of a good paper. Also, as a journal reviewer in his field, Khashayar will discuss the hot spots of a journal paper, which should be considered very carefully in order to best represent your research to reviewers and journal editors. Finally, Khashayar will show how to devise an appropriate strategy to manage your time and produce high quality publications.

Monday 13th June Behnam Ganji Energy Management in Hybrid Electric Vehicles 

 Abstract

A hybrid electric vehicle is a complex system consisting of many individual components. In order to achieve an efficient model, it is necessary to understand how the model is accurate and what level of approximation is acceptable for the application. The model must be capable of delivering the right compromise between accuracy and computational time. These objectives can be obtained not only from engineering knowledge and ingenuity, but also from experience and intuition. Accordingly, in this presentation at first an introduction to the different power train configurations and the concept of hybridization is presented.

In order to select an appropriate model for the energy management a study based on drive cycle analysis is carried out. A trade-off between conventional, series hybrid, parallel hybrid is drawn and the results of simulation are presented.

The different method of control in HEVs is reviewed and the format of an optimal problem is constructed. Then an Adaptive Cruise Control (ACC) integrated with an optimizer is revealed. This system is implemented with a modelled parallel hybrid electric vehicle. Based on the model, a cost function is defined with respect to the consumed energy. Genetic algorithm (GA) is used to solve the cost function for real highway data.

Monday 6th June Abbas Khosravi Research on Prediction Intervals: Recent Findings 

 Abstract

Prediction Intervals (PIs) have been widely used in literature for quantification of uncertainties associated with point forecasts. As PIs provide much more information than exact predicted values, they can be effectively and efficiently integrated into expert systems for optimal decision-making. Through this presentation, recent trends and findings in theory and application of PIs are investigated. Innovative methods for rapid construction of reliable PIs and their optimization (making them narrower with a greater coverage probability) are discussed. Also, some results on their application in different fields are presented and discussed.

Monday 30th May Mohammed Hossny Short Tales of CISR's Research 

 Abstract

This presentation highlights three research topics. First, image fusion techniques for HDR imaging is discussed. Second, VIDEO2IMAGE fusion is proposed for minimal radiation fluoroscopy. Finally, Content-based AR Marker is discussed as well as moving the AR rendering scheme from recognition to interpretation.

Monday 23rd May Khashayar Khoshmanesh Dynamic Analysis of Drug-Induced Cell Death Using Microfluidics 

 Abstract

Quantification of programmed and accidental cell death provides useful end-points for the anticancer drug efficacy assessment. Cell death is a stochastic process, and the opportunity to dynamically quantify individual cellular conditions is advantageous over the commonly employed static assays. This presentation describes the development and application of a microfluidic, dielectrophoretic (DEP) cell immobilisation platform for the real-time analysis of cancer drug-induced cytotoxicity. Microelectrode arrays were designed to generate weak electro-thermal vortices that support efficient drug mixing and rapid cell immobilisation at the delta-shape regions of strong electric field formed between the opposite microelectrodes. This technology was applied for the dynamic analysis of hematopoietic tumour cells that represent a particular challenge for real-time analysis due to their dislodgement. The present study provides a comprehensive mechanistic rationale for accelerated cell-based assays on DEP chips using real-time labelling with cell permeability markers. The work provides data on the complex behaviour of dying cells in the DEP field and the overall bioassay performance. Results indicate that DEP cell immobilisation technology can be readily applied for the dynamic analysis of investigational drugs in hematopoietic cancer cells.

Monday 16th May Aung Kyaw Soe Micro-fabricating and Interfacing of Neurons on microfluidic chip to Silicon and investigating learning 

 Abstract

The research is micro-fabricating and Interfacing of Neurons on microfluidic chip to Silicon. Using that tool, how learning and memory form in neuron networks cultured in vitro responding to stimuli and state will be investigated. It is to design and develop an integrated platform for neuron cells culturing on a chip with arrays of microelectrodes, neurons are to be close loop interfaced to embedded and real time systems and data acquisition and stimulation software. The integrated device will not only facilitate investigating dynamics of neuronal networks as it cannot only record but stimulate neuron networks via electrodes to condition (train) them with Hebbian learning. It is to deal with extra-cellular action potentials - voltage spikes found by Hodgkin-Huxley.

One objective is to replace micro electrode array (MEA) dish and systems those are expensive, bulky, and not portable limiting neuromorphic computing systems research, constraining the development of 'rat brained' robots driven by neuronal spikes recorded from dissociated cultured neuron networks. Proposed chip can form the core of biosensors, bioMEMS, real time PCR chips and low end personal preventive diagnostic devices to detect biomarkers. It can be used as cell culture analog such as "neuron networks on a chip" alike lung on a chip for drug testing model and it can be used to study neurons and networks cultured and survived long periods of time in vitro.

In addition to merits and features of intended research platform, there will be process innovation for the prototyping will be done using commercial off the shelf (COTS) and open source software. Research platform will be prototyped using Open source software, design and commercial off the shelf (COTS) hardware as another goal is to facilitate neural engineers, neuroscientists and neuromorphic researchers with affordable "neurons on a chip" platform and accessible processes. The low cost lab on a chip tool will enable them to advance their fields of research. It can also advance embedded systems research as real biological neurons can be embedded inside Silicon and software based simulated or emulated artificial neurons or vice versa.

Although this research is primarily for in vitro neuron networks and silicon/software interfacing but systems developed as part of research can be reused for in vivo neuron-silicon interfacing by replacing the electrodes by changing planar electrode films on substrate into invasive electrodes to be inserted into the cortical regions of brain or spinal cords to record signals. In vivo electrode arrays interfaced to light weight, low power embedded systems with wireless communication capabilities can be used in cybernetic research transforming insects and bats as micro aviation vehicles for search and rescue missions.

Monday 9th May Samer Hanoun Pareto Archived Simulated Annealing for Single Machine Job Shop Scheduling with Multiple Objectives 

 Abstract

In this work, the single machine job shop scheduling problem is studied with the objectives of minimizing the tardiness and the material cost of jobs. The simultaneous consideration of these objectives is the multi-criteria optimization problem under study. A metaheuristic procedure based on simulated annealing is proposed to find the approximate Pareto optimal (non-dominated) solutions. The two objectives are combined in one composite utility function based on the decision maker's interest in having a schedule with weighted combination. In view of the unknown nature of the weights for the defined objectives, a priori approach is applied to search for the non-dominated set of solutions based on the Pareto dominance. The obtained solutions set is presented to the decision maker to choose the best solution according to his preferences. The performance of the algorithm is evaluated in terms of the number of non-dominated schedules generated and the proximity of the obtained non-dominated front to the true Pareto front. Results show that the produced solutions do not differ significantly from the optimal solutions.

Monday 2nd May Marzieh Asgari 3D Particle-Based Cell Modeling for Cell Microinjection 

 Abstract

Force feedback for microrobotic intracellular injection has many beneficial implications. In particular, a haptic device can provide force feedback to an operator for training purposes. This work introduces a 3D particle-based model to simulate cell membrane mechanics during cell injection. The model is based on Voigt material and fluidic incompressibility properties of cell interior. As part of a parameter estimation a preliminarily work to measure deformation of a real cell is presented. The model is validated using experimental data of zebrafish embryo microinjection. The results demonstrate that the developed cell model is capable of estimating zebrafish embryo deformation and force feedback accurately.

Monday 18th April Ali Ghanbari Micropillar-based on-chip system for force pattern analysis in C. elegans locomotion 

 Abstract

Investigating the correlation between locomotive behaviour and genes has raised an lot of interest in life science. C. elegans, a soil-dwelling multicellular eukaryotic nematode with its genome completely sequenced, is an established experimental genetic system to investigate the relationship between genes and motion behaviour. This talk will address an automated tracking and force measurement system for C. elegans in motion, based on microscopy computer vision. A polydimethylsiloxane (PDMS) device is microfabricated to allow C. elegans to move in a matrix of micropillars in a channel. An image processing algorithm has been developed to track the worm and deflecting micropillars automatically in purpose of quantifying multi-point locomotive forces of a moving C. elegans. The developed on-chip system is able to visually resolve forces with sub-micro Newton resolution.

Monday 11th April Quan Zen (Danny) Ang Multi-point haptic interaction for mediator interfaces: A novel approach to teleoperation 

 Abstract

Multi-point haptic interaction is essential to achieving a wider range of haptic interaction. Firstly, a new multi-point haptic platform is introduced enabling the user to grasp virtual objects using two fingers. The low-cost platform utilises two Phantom Omni haptic devices and provides effective grasping and manipulation of virtual objects. The multi-point platform is integrated with CHAI 3D and its operation presented. Haptic guidance for multi-point haptic interaction is then considered. Haptic virtual fixtures modelled on potential fields are introduced to guide the user in multi-point grasping tasks. Simulation demonstrate the ability for the user to be guided to predetermined grasp points. Finally, the concept of multi-point mediator interfaces is introduced. Such an interface allows an operator to control various robot functions through multi-point haptic interaction. A method to control robotic manipulators by virtually grasping and moving individual linkages is introduced.

Wednesday 6th April

10:30am
CISR Breakout Area
Will Vandenberg External presentation  Simulation in the Mining Industry 

 Biography

Since joining TSG Consulting in 2002, Will has been fundamental in the success of the BHP Billiton Iron Ore and Ravensthorpe Nickel studies. He has been involved in every significant expansion project and contributed to Business Improvements studies, Integrated Planning, Third Party Rail Access Studies and marketing projects. Since his promotion to Manager in 2009, Will has led a team of consultants on the BHP Billiton Iron Ore projects.

Prior to joining the TSG Team, Will worked as a Process Engineer with Rio Tinto's Technical Services division. This role saw Will accountable for conducting preliminary feasibility studies into the application of a new process optimisation package for mineral processing plants within the Rio Tinto group. During his time at Rio Tinto, Will gained experience in system analysis, process modelling, simulation and in conducting site trials across a range of areas within the mining and mineral processing industry.

While at Rio Tinto, Will formed part of a site-based research group at Comalco's Tasmanian aluminium smelter in the role of Process Engineer. The research group was involved in the development of robust and energy efficient aluminium reduction cells. The principal focus of Will's role was the evaluation of experimental cell performance and the application of this analysis to achieve improved process control. The research group was also involved in designing and conducting experimental trials, presenting findings to relevant staff and managing the performance several non-routine construction projects.

Will has also formed part of a project team involved in the high-level modeling of Hamersley Iron Operations, which aimed to develop a better understanding of how the system output - the iron ore stockpiles at the port - were affected by various changes to operational policy with respect to mining, ore processing and transport. Tools were developed, in particular discrete event simulation models, which identified the key sources of process variation and bottlenecks and their contribution to final product throughput and quality. Will's role within the project team included sourcing and analysing appropriate operational data and performing discrete event simulation of the process.

In 2002, Will completed a research degree with the Department of Engineering, Australian National University in Canberra, Australia. This 2 year degree included a strong focus on collaborative projects with the key industrial partner, Telstra. Will's thesis, entitled "Demand Forecasting for Resource Planning", focused on identifying the limits to predictability of future installation demand and fault occurrence in regional Australia and how this knowledge could be used to improve Telstra's long-term resource planning and short-term scheduling activities. As a result of Will's study, conducted in partnership with Regional Service Operations Resource Team, and subsequent successful field trials, major changes to the resource planning and scheduling functions have been rolled out nationwide in less than a year. Both performance improvements and significant cost savings, estimated to be in excess of $10million in the first year, have been achieved in the regional operations. As a result of this success work is underway to adapt the initiatives for use in urban regions.

 Abstract

Monday 4th April Wael Abdelrahman Data-Driven Computation Of Contact Dynamics During Two-Point Manipulation Of Deformable Objects 

 Abstract

This study represents a preliminary step towards data-driven computation of contact dynamics during manipulation of deformable objects at two points of contact. A modeling approach is proposed that characterizes the individual interaction at both points and the mutual effects of the two interactions on each other via a set of parameters. Both global as well as local coordinate systems are tested for encoding the contact mechanics. Artificial neural networks are trained on simulated data to capture the object behavior. A comparison of test data with the output of the trained system reveals a mean squared error percentage between 1% and 3% for simple interactions.

Monday 28th March Shady Mohamed Robust Rapid MRI 

 Abstract

The Magnetic Resonance Imaging (MRI) has proved to be a very useful medical imaging technique. The main drawback of this technique is that the scanning time is high due to physical limitations. This long scanning duration limits the number of patients served daily which creates backlogs. This research aims at increasing the speed of the MRI scanning process. The results of this research will significantly reduce the scanning time by at least one third. This research does not consider redesign of MRI machines; rather it considers designing a robust MRI image reconstruction algorithm from fewer MRI measurements.

Friday 25th March

2pm
GTP Conference Room
(na1.014)
Prof. Peter Branyi External presentation  VirCA (Virtual Collaboration Arena www.virca.hu

 Biography

Peter Branyi Ph.D, D,Sc is the head of the 3D Internet based Control and Communication laboratory, and is a professor at Budapest University of Technology and Economics. He initiated and organized the foundation of the ITM Norwegian and Hungarian laboratory. He received the International Dennis Gabor Award and the Youth Prize of the Canadien and International Constituency Group of Sigma Xi. He had visiting research positions at various Universities. His main interest is the TP model based polytop representation for qLPV and LMI based modern control methodologies.

 Abstract

VirCA is a loosely coupled modular, 3D Internet based interactive virtual environment for collaborative manipulation of robots and other hardware or software equipments. It has the following advantages:

3D Internet Collaboration:

VirCA provides a platform where users can build, share and manipulate 3D content, and collaboratively interact with real-time processes in a 3D context, while the participating hardware and software devices can be spatially and/or logically distributed and connected together via IP network. The VirCA 3D virtual reality component can work together with the cutting edge stereoscopic 3D display technologies including the immersive 3D CAVE systems.

Augmented collaboration:

The 3D content and processes in VirCA can be synchronized with the real world, which allows the combination of reality and virtuality in the collaboration arena. This allows VirCA users to virtually interact not only with other users, but also with existing, remotely operated hardware and software, such as robots, sensors, or actuators. This type of semi-virtual interaction allows the users to build distributed systems consisting of real and virtual parts at the same time.

Knowledge Plug and Play:

VirCA's uniform, modular, RT-Middleware based framework allows to organize components from different sources into a system using a browser-based graphical programming environment. This allows the plug and play exploitation of the knowledge that is embedded into the academy, industry, or community developed components, and simplifies the creation of new, state of the art solutions. In our vision, RT-Middleware components and VirCA-enabled software pieces (CyberDevices) implementing standard generic or even fully customized interfaces could be collected in a community website. System integrators could use these functional building blocks for free or even on commercial basis similarly to the popular online application stores.

Beyond cutting edge technology:

Since the virtual environment in VirCA is generated by computers, all information has a corresponding internal representation in the system. Therefore, in this environment everything can be measured or detected, and various kinds of information can be extracted via virtual sensors, which need not even exist in the real world today. Such combination of real and virtual sensors can be very powerful.

Wednesday 23rd March Prof. Paul Brown External presentation  Art Plus Science - Science Plus Art 

 Biography

Prof. Paul Brown is a Brisbane-based artist and writer who has been involved in art, science and technology collaborations and especially the computer-based arts for over 40 years. His artwork was recently included in the "Digital Pioneers" exhibition in London's Victoria and Albert Museum. He is currently an honorary visiting professor of art and technology at the Centre for Computational Neuroscience and Robotics (CCNR), School of Informatics, University of Sussex, UK and the Synapse Artist in Residence at the CISR where he is a member of the team investigating the application of haptic devices to enable the blind to experience 2-D artworks.
www.paul-brown.com
www.brown-and-son.com
www.informatics.sussex.ac.uk/research/groups/ccnr
www.vam.ac.uk/exhibitions/future_exhibs/Digital%20Pioneers/index.html

 Abstract

In this talk Prof. Brown will describe two projects: his ongoing work using cellular automata to navigate very large geometric permutative sets and; an ongoing project at the CCNR that is attempting to use evolutionary and adaptive systems to make a robot that can draw.

Monday 21st March Hamid Abdi Design of optimal fault tolerant manipulators 

 Abstract

There are a lot of research on the safety and reliability of robots. Fault tolerance increases the reliability and safety of robots. Robots in surgical tasks, in hazardous material disposal tasks, and in exploration of see and space are a few common applications of high reliable robots. In addition, coexistence of human and robots has recently opened serious questions about the safety and reliability. This talk aims to addresses the optimal fault tolerance in serial robotic manipulators. The main focus of talk is about design of optimal fault tolerant manipulators. A novel method for design of local optimal fault tolerant Jacobian matrices for manipulators is presented and a class of Jacobian generators is introduced based on symmetric geometries. It is shown that the optimal fault tolerant Jacobian matrices can be obtained via symmetric geometries such as pyramids. The concept of zonotopes is used in the literature of fault tolerant robotics to indicate an exact connection between the fault tolerance and specific class of symmetric geometries for serial manipulators.

 Photos

Hamid Abdi

 

Deakin University acknowledges the traditional land owners of present campus sites.

27th March 2012