Research

Cyber-Physical Systems

My research is focused on foundations of Cyber-Physical Systems (CPS). CPS are next-generation computer systems which sense and actuate physical environment. Their examples are autonomous driving vehicles, unmanned aerial vehicles, and intelligent surgery robots. CPS can potentially revolutionize entire our life.

One important aspect of CPS is real-time systems because CPS interact physical entities which are sensitive to time. Real-time systems are systems whose correctness depends on their temporal aspects as well as their functional aspects. In Real Time scheduling, its performance is evaluated by timeliness on timing constraints (deadlines). On the other hand,  speed/average case performance/throughput, which is used for performance metric in traditional system research,  are less significant. Its key property is predictability on timing constraints.

Machine Learning Security and Autonomous Driving 

Federated learning is a learning method that collects only learned models on a server to ensure data privacy. This method does not collect data on the server but instead proceeds with data directly from distributed clients. Because federated learning clients often have limited communication bandwidth, communication between servers and clients should be optimized to improve performance. Federated learning clients often use Wi-Fi and have to communicate in unstable network environments. However, as existing federated learning aggregation algorithms transmit and receive a large amount of weights, accuracy is significantly reduced in unstable network environments. In this study, we propose the algorithm using particle swarm optimization algorithm instead of FedAvg, which updates the global model by collecting weights of learned models that were mainly used in federated learning. The algorithm is named as federated particle swarm optimization (FedPSO), and we increase its robustness in unstable network environments by transmitting score values rather than large weights. Thus, we propose a FedPSO, a global model update algorithm with improved network communication performance, by changing the form of the data that clients transmit to servers.  

Mixed Criticality Embedded Systems

Many safety-critical real-time systems such as avionics and automotive consist of multiple functionalities with different criticality. For example, an Unmanned Aerial Vehicle (UAV) consists of flight-related (high-critical) functionalities and mission-related (low-critical) functionalities. An increasing trend is to integrate multiple components with different criticality into a single shared platform, called Mixed-Criticality (MC) systems, in order to reduce manufacturing cost.