Work packages

New intelligent algorithms

This work package focuses on research on the design and operational mechanisms of intelligent algorithms in open environments. Through formal modeling and formal specification, we aim to study how to formally describe the operating scenarios of intelligent algorithms, model human-machine mixed intelligent systems, and develop formal specification methods for the trustworthiness properties of intelligent algorithms.

Trust assurance

This work package undertakes research on developing trust assurance technologies for intelligent systems in open environments. The main focus will be on the establishment of rigorous methods guaranteeing trustable perception components, knowledge and models in open environments, as well as the generation of intelligent system control strategies and runtime monitoring methods based on reinforcement learning.

Simulation platforms

This working package will investigate simulation platforms for autonomous driving, formal design of control systems, and safety verification under typical scenarios. Further, it will conduct research on the application of reinforcement learning methods for intelligent control on software-defined satellites and validate the effectiveness and feasibility of the methods through satellite control algorithm test platforms and semi-physical simulation test platforms.