嘉宾介绍
汪建平
香港城市大学计算机学院院长、主任教授
Prof. Jianping Wang is a Chair Professor of Computer Science and Dean of the College of Computing at the City University of Hong Kong. Previously, she served as the Associate Provost from 2023-2024 and Assistant Provost from 2020-2023 at the City University of Hong Kong. Her research interests are autonomous driving, networking systems, and cloud computing. Her research has been sponsored by various funding agencies and transferred to industry companies. Her team has developed state-of-the-art solutions for the security and safety of autonomous driving, winning the 1st place in the 2025 Waymo Open Dataset Interaction Prediction challenge, the 1st place in the CVPR 2024 Argoverse 2 Multi-Agent Motion Forecasting Challenge, the 1st place in the 2024 Waymo Open Dataset Sim Agents Challenge, the 1st place in the 17th F1-Tenth Autonomous Grand Prix, the 1st place at the Grand Prix Metaverse Autonomous Driving Challenge 2023, and the 1st place in the Argoverse 2 Multi-agent Motion Forecasting Challenge 2023. She is a fellow of IEEE and a fellow of AAIA. She is among the top 2% in the Stanford list of the world's most-cited scientists in 2020-2024. She was and is an associate editor for many journals such as IEEE Transactions on Mobile Computing, IEEE Transactions on Cloud Computing, IEEE Transactions on Parallel and Distributed Systems, etc. She served on the IEEE fellow committee (2024-present).
内容抢先读
Despite significant advancements in autonomous driving technology, ensuring safe navigation in all scenarios remains a critical challenge. In this talk, we will present our recent work aimed at achieving safe autonomous driving through three interrelated pillars: rigorous validation, reliable driving, and resilient security. Together, these elements form a comprehensive and cohesive approach to safety. First, we will outline our validation methodologies designed to rigorously test autonomous driving systems before their deployment on public roads. To supplement costly real-world testing, we have developed simulation tools that create high-fidelity, safety-critical traffic scenarios and realistic behaviors of traffic agents, as well as mixed-reality testing approaches. Second, we will discuss our efforts focused on ensuring the reliable driving of autonomous vehicles in dynamic and uncertain environments. Our research focuses on prediction and motion planning, including both single-vehicle solutions that enable vehicles to accurately predict the future trajectories of nearby agents, as well as infrastructure-aided motion planning. Lastly, while the first two pillars address failures that arise naturally from complex traffic scenarios, we also explore the threat of malicious attacks from adversaries and the corresponding defence mechanisms. We will introduce our findings on attacks targeting autonomous vehicles' prediction and online map construction tasks. Additionally, we will present a diagnostic tool that identifies the root causes of system failures, thereby facilitating more effective recovery efforts. We have evaluated all our security-related approaches on real autonomous vehicles.
嘉宾介绍
Zsolt Szalay
布达佩斯技术与经济大学汽车技术系系主任
Zsolt Szalay is an electrical engineer and economist, Head of the Department of Automotive Technologies at the Budapest University of Technology and Economics (BME). He earned his degree in electrical engineering in 1995 and in economics in 1997 and completed his PhD in 2002 at BME. Between 2021 and 2023, he studied innovation management in the MIT REAP program. A member of the Hungarian Academy of Engineering since 2009, he joined its Board in 2023. He has been active in automotive innovation for over 30 years.He began his career at Knorr-Bremse, working on electronically controlled braking systems, later becoming project manager for fourth-generation EBS systems. He then founded his own automotive research company. Today, telematics solutions from Inventure Automotive are used in over 1 million vehicles across 85 countries.Between 2016 and 2023, he contributed to the planning and implementation of the ZalaZONE Automotive Test Track. A faculty member at BME since 2005, he has led the department since 2015. He founded and leads the BME Automated Drive Lab research group. His research focuses on automated vehicle control and combining virtual and physical testing methods for mobility systems.
内容抢先读
The presentation will give an insight into the current EU research directions in ADS and ADAS through the activities of the BME Automated Drive Lab. Here, the research goals include saving lives by avoiding traffic accidents through automated vehicle control, mobility for all that is not only safe but also cyber-secure, as well as deep traffic digitisation and dynamic HD maps through real-time digital twins for a safer and more efficient transport. Some of the research results have led to BME spin-off companies, as BME promotes entrepreneurship and commercialisation, since academic research should not only have a scientific impact, but also an economic, and particularly a social impact.
讲座信息
时间：2025年7月19日（周六）09:30—11:30地点：清华大学李兆基楼A247会议室
请通过下列二维码报名，主办方将于7月18日17:00前通过邮箱发送确认信息。
