(a) Time-Critical Edge Applications in Autonomous Vehicles (Dr. Kecheng Yang)
With the advances in robotics, autonomous systems have become increasingly pervasive and are able to provide considerable computing power at the edge. Meanwhile, a significant portion of data and applications in such autonomous systems are time-sensitive. This project focuses on the infrastructures and the schedulers in particular in such systems to improve their abilities to support time-critical applications. The Robot Operating System (ROS) has been widely used to develop autonomous systems, and its successor ROS2 was proposed for enhanced real-time support and performance. Students will experience how the applications are implemented on top of ROS or ROS2, investigate the computing resource allocation schemes, and perform system-level optimization to improve the real-time performance of the applications. The outcomes will be evaluated by both simulation and actual implementation on autonomous model cars.
(b) Wearable Sensing System for Personalized Fall Detection at the Edge (Dr. Anne H.H. Ngu)
Fall is a major public health problem leading to loss of life independence, premature mortality, and major hospitalization cost. Wearable devices, especially smartwatches that pair with smartphones, are increasingly a platform of choice for deploying digital health applications. This project focuses on creating a personalized fall detection system on low-cost smartwatches that can detect both “hard” and “soft” falls. In particular, students will experiment with different machine learning personalization strategies, different model validation strategies and scalable mechanism for data collection for personalization. Students will be exposed to skills such as Android programming, machine learning with Python, non-SQL database and edge computing using smartwatches and phones.
(c) DLP: Dynamic and Lightweight Protection towards the Security of IoT Devices (Dr. Tao Hou)
With the evolvement of sophisticated and low-cost chips and the ubiquity of wireless networks, Internet of Things (IoT) has been extensively deployed in various domains, including smart home, wearable, health care, manufacturing, agriculture and military. The variety and novelty of different IoT applications indeed improve the efficiency and service quality of our daily lives. However, it also incurs a range of security concerns. As IoT devices may gather geographical information, monitor users’ privacy activities, and record clients’ biometric features, two of the critical concerns are how to ensure the operational stability and preserve the confidentiality of sensitive data. Intuitively, cryptography methods can be applied to protect the operation and encrypt all the conversation between IoT devices. However, as IoT devices are usually featured with limited computational capacity and limited power, they may not afford expensive cryptography operations by conventional methods like AES or RSA. To cope with the limited resources of IoT devices, this project aims to develop a dynamic and light-weight scheme to protect IoT devices against attacks like Distributed Denial-of-Service (DDoS) and Malicious Eavesdropping.
(d) Cognitive-Neuroscience Inspired Reinforcement Learning for Cyber Physical System (Dr. Heena Rathore)
Cyber-physical systems (CPS) are becoming more and more significant in our daily lives in the form of internet of things (IoT) devices, medical devices, or connected vehicles. CPS differentiate themselves by their capacity to adapt and learn by assessing their surroundings, learning patterns, and correlations. Machine learning plays an important role in achieving these inherent capabilities of CPS. Typically, the performance of ML models is dependent on the selection of an optimal model, and large amounts of training and quality data to operate efficiently. Thus, we require novel models to overcome the above-listed limitations and be able to multitask and coordinate to improve system performance and enhance CPS productivity. CPS are intrinsically equivalent to human beings in their ability to sense the environment and make real-time decisions which can be achieved using reinforcement learning models. This research project aims to develop novel reinforcement learning models inspired by human cognition for CPS applications. The students will hone their skills in developing mathematical models and simulating these computational reinforcement learning models.
(e) Data Science in Healthcare and Medical Research (Dr. Lu Wang)
Healthcare systems are changing in the era of big data. Advances of artificial intelligence (AI) in healthcare make it possible for healthcare providers to sift through tremendous amounts of information efficiently, which eventually help them take care of their patients better. There are various types of health information ranging from medical literature to pathology reports. Albeit the last few years have witnessed an explosive increase of healthcare data in terms of volume, variety and veracity, it is insufficient to build a robust prediction model in various scenarios due to time, geographical and domain inherent constraints. How to develop and apply Machine Learning (ML) methods that can efficiently utilize Electronic Health/Medical Records (EHRs/EMRs) is significant to facilitate decision making of physicians in their clinical practice.
In addition to ML methods, inspired by the motivation of human-in-the-loop, Human-Centered AI (HCAI) for Data Driven Decision Making (D3M) addressing healthcare/medicine problems attracts more attention to improve the situation awareness and the quality of decisions. More specifically, interactive Machine Learning (iML) improves the ML prediction by looping human experts in the learning process and integrating human expertise. From another perspective, eXplainable Artificial Intelligence (XAI) and trustworthy AI in healthcare systems not only improve the uptake of ML model but also increase physician trust in ML prediction for clinical decision making.
In this project, you will have opportunities working with the real healthcare and medical data including EHRs/EMRs for multiple cognitive disorders and chronic diseases collaborating with physicians, clinicians and psychiatrists, etc.