Project 1: Parking control and simulation for intelligent transportation systems (Cybercar)
Supervisor: Dr. Zhou (zhouyuanqiang@sjtu.edu.cn)
In this project, the model-based optimization problem for car parking is solved. Using the MPC method, the parking trajectory of the car is offline optimized with state constraints and control limits. Then, the movies and animations are created with Matlab, as you can see at https://youtu.be/1Jf6t2YLmK8.
If you are interested in this project, have some questions, or can offer a new position, please feel free to contact me.
Project 2: Learning-based secure control framework for cyber-physical systems
My primary objective in this project is to develop a learning-based secure control framework for cyber-physical systems in the presence of sensor and actuator attacks. Specifically, we use a bank of observer-based estimators to detect the attacks while also introducing a threat detection level function. Once the attack mitigation is triggered, a two-player, zero-sum differential game is formulated and the underlying joint state estimation and attack mitigation problem is solved by using a reinforcement learning algorithm.
There are many potential examples of networked control systems in various application areas, such as smart power grids, intelligent traffic control systems, and so forth. With so much information collected from the communication network, control engineers can make precise and optimal control strategies emerging from intelligent control theory, starting from optimal control, adaptive control, robust control, and many other advanced forms of control algorithms. However, applying all these control strategies over a network with so many sensors is a challenging task.
This multidisciplinary project has a major area of communications, networking, and signal processing, which provides essential enabling and supporting technologies for different applications of dynamically networked systems. This project plans to develop the effective approaches and the core techniques that are available for the analysis, optimization, and implementation of distributed networked control systems. To ensure industrial relevance and impact, the advisors have established collaborations with many industrial partners to demonstrate these new approaches, e.g. large-scale electric power networks and large-scale, integrated water-saving irrigation systems. For more details, see the published papers on my publication page.