Projects

This is a brief list and summary of the projects undertaken in our group. If you are interested in any of the themes, please get in touch. We are always looking for new and exciting research challenges in computational spatial information science. Whether you are form industry, government, other universities or a student looking for a PhD, please get in touch.

We would love to branch and grow these streams with more funded and industry -supported projects - if you are interested, please get in touch!

We will progressively add separate sites for each of these projects, summarizing the research outputs.

Current funded projects

  • [2021] Walking the city: Digital infrastructure for pedestrian mobility . ARC Discovery project (DP170100153). CIs: Winter; Tomko.

    Pedestrian access, flow and management are critical for urban life. However, compared to other forms of mobility pedestrian mobility is significantly more complex. Currently, various incompatible pedestrian route graphs in both outdoor and indoor environments render any analysis biased and non-transparent. This project aims to solve this problem by developing a universal and necessarily hierarchical pedestrian route graph to support critical applications such as urban walkability (health), space and asset management (guidance, flow management), and public safety (evacuation). In contrast to conventional algorithms, we will take a novel approach based on human cognition to define this universal graph and then integrate topology and geometry.

  • [2021] Digital breadcrumbs feeding urban decision-making . Marsden Fund (Royal Society Te Aparangi, 20-UOA-056). CI:Katarzina Sila-Nowicka (U Auckland, NZ), AIs: Demsar (U St Andrews, UK), Tomko.

    Led by Dr Sila-Nowicka, UAuckland, NZ: In order to publish and use location data in a safe way, human mobility datasets have to be anonymised in a privacy-preserving way. Human mobility science has a huge potential to help the society find solutions to the most pressing contemporary problems – a recent example is contact tracing during the COVID-19 pandemic in many countries around the world – but that cannot be done without access to data. However, neither a unique methodology nor stable qualitative metrics for privacy protection of mobility data exist. There are a number of inconsistently-used metrics describing the quality of anonymization, but these metrics are hard to verify as they are tested on different datasets, thwarting reproducibility. This project will develop new tools for anonymization of movement data.

  • [2021] Indigenous Engineering: interpreting engineering foundations of Budj Bim . ARC SRI project (ARC SRI for Australian Society, History and Culture, SR200200227). CIs: Tomko, Prpic, Khoshelham, Lise-Pronovost, Langton, Bell.

    The overall aim of this project is to elucidate the traditional engineering processes that enabled the Gunditjmara to site, plan, construct and operate the elaborate eel trapping and aquaculture systems, water channels and settlements at the Budj Bim World Heritage Cultural Landscape. This project will enhance our understanding and conceptualising of Gunditjmara history, technological ingenuity and endeavour by revealing how the system evolved over time. We aim to (1) describe the extent and parameters of the Budj Bim aquaculture system, (2) develop a model of the productivity of the aquaculture system, (3) contribute sophisticated analytical methodologies for archaeology, and (4) provide a visual interpretation of how the site functioned.

    News:

    • May 2022: Lily Nash finished her honours thesis in archaeology on Least-Cost paths through the Budj Bim cultural landscape. Congratulations!
    • February 2022: Dr Brian Armstrong has joined the project

Past funded projects

We maintain many of these research streams – talk to us!

  • [2017-2021] Self-Healing Maps: protecting maps through automatic updating processes . ARC DP project (DP170100153). CIs: Winter; Rajabifard; Kealy; Khoshelham; Tomko; Kalantari.

    Our ability to sense the state of cities in real-time dramatically exceeds our ability to automatically and reliably integrate map data in real-time. Currently, no map can accurately reflect the changes on the ground as they happen and at the same time guarantee the map’s integrity. Quality assurance of map data is a lengthy and manual process. Map users such as emergency responders, traffic services and the public have to act based on out-dated rather than unchecked information. Map administrators lack mechanisms enabling responsive and autonomous updates of general-purpose maps from ad-hoc data sources with guarantees about resulting map integrity and quality. This project researched mechanisms to automatically monitor, correct and maintain data in dynamically updated multisource map databases, answering the main quesiton: What mechanisms must be combined to assure ongoing, real-time consistency of map data to deliver self-healing maps?

  • [2017-2020] Making human place knowledge digestible by computers . ARC DP project (DP170100109). CIs: Winter; Renz; Baldwin; Tomko, Kuhn.

    Questions of where and when are frequent in the search for information. Providing satisfactory answers to questions about where and when has become critical, with significant economic implications. This project researched how place refrences occur in human place questions, and how answers realte to them. It enabled to devise an automated, place-based question answering system ( see Demo).