The IoT Software Engineering group aims to bridge the gap between the research community and the industry community, and to bring the most brilliant research ideas to solve real-world challenges.
The group currently has 6 PhD students (1 PhD is co-supervised with UTS and 3 MRES-PhD students). Their research topics are: “Trustworthy AI-human interaction in cyber-physical systems”, “Privacy preserving for intelligently connected vehicles”, “Detecting anomaly behaviour of elderly people using non-intrusive sensor fusion”, “Securing LiDAR based autonomous driving systems”, “Leveraging data-driven metamorphic testing to improve robustness of DNNs”, and “Real-time anomaly detection of autonomous systems using online learners”.
The group has several projects in the past three years that addressed major issues in autonomous systems: “Securing modern vehicle systems” ($17,100), “Dynamic safety analysis for cyber-physical systems” ($18,000), and industry projects related to edge computing and micro-services “Customizable and efficient development and deployment of micro-service” ($143,500), “Trustworthy AI-human interaction in cyber-physical systems (Data61/CSIRO)” ($105,000).
The group’s recent paper “A survey on security issues in services communication of microservices-enabled fog applications” has been recognized as a top 20 most read paper (2017-2018) in Concurrency and Computation: Practice and Experience.
Selected work with real-world impact:
- Haiming X, X. Zheng, et al. A Hybrid Method Combining Markov Prediction and Fuzzy Classification for Driving Condition Recognition. IEEE Transactions on Vehicular Technology, 2018. [IF=4.06]
This paper is the first of its kind to use feature engineering and machine learning to solve a fundamental challenge in electric vehicles, which is to extend the mileage sustainable by in-car battery before requiring recharge. The first author was a PhD student co-supervised by Dr. Zheng in Tsinghua. With this work as one of the foundation, after graduation the student was engaged by some venture capitals to establish a company producing electric vehicle power system. The experience gained from the work will aid Dr. Zheng to resolve the machine learning related challenges in the real-world.
- M. Fu, X. Zheng, et al. DaLiM: Machine Learning Based Intelligent Lucky Money Determination for Large-Scale E-Commerce Businesses. International Conference on Service-Oriented Computing, 2018. [CORE A]
This paper is a pioneer to apply machine learning and feature engineering inside Alibaba Group to determine lucky money for online customers in “Double 11 Global Shopping Festival 2017”. The feature engineering part is very thoroughly conducted and with the evaluation of a few hundred million customers. The approach significantly reduces the campaign cost and boosts the online sales. The first author has been working very closely with Dr. Zheng on the solution and has been promoted to senior management role due to the outstanding contribution made through this work. The experience gained from feature engineering and machine learning will also help Dr. Zheng to resolve machine learning related challenges in the real-world.