| Session | Presentation | Presenter | Affiliation | Title | Timeslot |
|---|---|---|---|---|---|
| Ses. 1 | Opening | Prof. Hailong Liu | NAIST, Japan | Opening speaking | 09:00-09:10 |
| Invited speaker 1 | Prof. Junmin Wang | UT Austin, US | Quantification of Driver Attentiveness and Impact of Visual Communication in Vehicle Lane-Keeping Automation | 09:10-09:40 | |
| Invited speaker 2 | Prof. Jia Hu | Tongji Uni., China | Towards socially-aware autonomous driving: An interpretable data-driven perspective | 09:40-10:10 | |
| Coffee break | 10:10-10:30 | ||||
| Ses. 2 | Invited speaker 3 | Prof. Takahiro Wada | NAIST, Japan | Computational Models of Human Self-motion Perception and Motion Sickness based on Subjective Vertical Mismatch Theory and Its Application to Vehicle Motion Comfort | 10:30-11:00 |
| Invited speaker 4 | Prof. Fang You | Tongji Uni., China | Design of Interactive Safety Warning Information Design of External Display Interface for Autonomous Driving Vehicles | 11:00-11:30 | |
| Invited speaker 5 | Prof. Daofei Li | Zhejiang Uni., China | Human-centered approach to automated driving: from decision-making, motion planning to control | 11:30-12:00 | |
| Lunch break | 12:00-13:00 | ||||
| Ses. 3 | Invited speaker 6 | Prof. Hailong Liu | NAIST, Japan | Causal Discovery from Psychological States to Walking Behaviors for Pedestrians Interacting an AV Equipped with eHMIs | 13:00-13:30 |
| Invited speaker 7 | Prof. Adriana Tapus | ENSTA Paris, France | TBD | 13:30-14:00 | |
| Invited speaker 8 | Prof. Jianmin Wang | Tongji Uni., China | IXAI: Generative Design of Automotive Styling Based on Inception Convolution with Explainable AI | 14:00-14:30 | |
| Invited speaker 9 | Bo Jiang | HUST, China | Advancing E2E-AD via Multimodal Planning, Reinforced Fine-Tuning, and Language Modality Integration | 14:30-15:00 | |
| Coffee break | 15:00-15:20 | ||||
| Ses. 4 | Poster 1 | coming soon... | 15:20-15:50 | ||
| Poster 2 | coming soon... | 15:50-16:20 | |||
| Poster 3 | coming soon... | 16:20-16:50 | |||
| Roundtable discussion | Interactive discussions for research and collaboration in HRI and HVI. | 16:50-17:35 | |||
| Closing | Asst. Prof. Hao Cheng | Uni. of Twente, Netherlands | Closing speaking | 17:35-17:50 | |
INVITED PRESENTATION 1 (20MIN + 10MIN)
Quantification of Driver Attentiveness and Impact of Visual Communication in Vehicle Lane-Keeping Automation
Prof. Dr. Junmin Wang
University of Texas at Austin, USA
Abstract
TBD
Biography
Junmin Wang is the Fletcher Stuckey Pratt Chair in Engineering and a Professor in Mechanical Engineering at University of Texas at Austin. In 2008, he started his academic career at Ohio State University, where he founded the Vehicle Systems and Control Laboratory, was early promoted to Associate Professor in September 2013 and then very early promoted to Full Professor in June 2016. In 2018, he left Ohio State University and joined University of Texas at Austin as the Accenture Endowed Professor in Mechanical Engineering. Professor Wang has a wide range of research interests covering control, modeling, estimation, optimization, diagnosis, and AI for dynamical systems, especially for automotive, vehicle, transportation, mobility, human-automation, robotic, energy storage, and manufacturing applications. Prof. Wang’s research contributions include the development of control and estimation methods that advance efficiency, driving safety, and emission performance of conventional, electrified, connected and autonomous/automated vehicles. He has five years of full-time industrial research experience (2003-2008) at Southwest Research Institute (San Antonio, Texas) where he was a Senior Research Engineer and led research projects sponsored by more than 50 industrial companies and governmental agencies worldwide. Prof. Wang’s research programs at UT Austin and Ohio State University have been funded by federal agencies and industrial companies such as National Science Foundation (NSF), Office of Naval Research (ONR), Department of Energy (DOE), U.S. Department of Transportation (DOT), National Highway Traffic Safety Administration (NHTSA), Army Research Lab (ARL), Texas Department of Transportation, GM, Ford, Honda, Tenneco, Eaton, Ftech, Denso, and others.
INVITED PRESENTATION 2 (20MIN + 10MIN)
Towards socially-aware autonomous driving: An interpretable data-driven perspective
Prof. Dr. Jia Hu
The College of Transportation Engineering, Tongji University, China
Abstract
Autonomous Driving (AD) vehicles still struggle to exhibit human-like behavior and socially interaction awareness in highly dynamic and interactive traffic scenarios. The key challenge lies in AD’s limited ability to interact with surrounding vehicles due to a lack of understanding of the underlying social interaction mechanisms. To address this issue, MPCFormer is introduced, which is an interpretable data-driven approach for socially-aware autonomous driving. To the best of the knowledge, MPCFormer is the first to explicitly model the dynamics of social interaction mechanism between vehicles. A novel interpretable data-driven modeling paradigm is introduced to ensure modeling interpretability. Afterwards, a socially-aware Transformer-based encoder-decoder architecture learns the social interaction dynamics from natural driving data. By integrating the learned social interaction dynamics into motion planning, the AD can proactively interact with surrounding vehicles. The planning safety is guaranteed by social interaction mechanism awareness and the safety constraints. Open-looped evaluation demonstrates that with the socially-aware capability, MPCFormer achieves the lowest trajectory prediction errors, from ADE@1s (0.12m) to ADE@5s (0.86m). Close-looped evaluation confirms the modeling interpretability of MPCFormer. Meanwhile, MPCFormer achieves the best planning success rate of 94.67%, improves driving efficiency by 15.75%, and saves lane change distance by 49.01%.
Biography
Jia Hu works as a ZhongTe Distinguished Chair in Cooperative Automation in the College of Transportation Engineering at Tongji University. Before joining Tongji, he was a research associate at the Federal Highway Administration, USA (FHWA). He is an Associate Editor of the IEEE Transaction on Intelligent Transportation Systems, IEEE Transaction on Intelligent Vehicle, American Society of Civil Engineers Journal of Transportation Engineering, IEEE Open Journal in Intelligent Transportation Systems, an assistant editor of the Journal of Intelligent Transportation Systems, an advisory editorial board member for the Transportation Research Part C, an associate editor for IEEE Intelligent Vehicles Symposium since 2018, and an associate editor for IEEE Intelligent Transportation Systems Conference since 2019. Furthermore, he is a vice chair of the SAE-China Intelligent Transportation Committee.
INVITED PRESENTATION 3 (20MIN + 10MIN)
Computational Models of Human Self-motion Perception and Motion Sickness based on Subjective Vertical Mismatch Theory and Its Application to Vehicle Motion Comfort
Prof. Dr. Takahiro Wada
Graduate School of Science and Technology, NAIST, Japan
Abstract
The importance of enhancing motion comfort in vehicles has been growing with the rapid development of automated vehicles and the increasing use of in-vehicle digital devices. Understanding human motion comfort requires a quantitative framework of self-motion perception, including motion sickness under various motion exposures such as vehicle dynamics. In this talk, I will provide an overview of recent research on modeling motion sickness and self-motion perception based on the Subjective Vertical Mismatch Theory, which has attracted increasing attention in recent years. Furthermore, I will present applications of these computational models to the study of motion perception and comfort in vehicles, including automated vehicles.
Biography
Prof. Takahiro Wada received his Ph.D. degree in Robotics from Ritsumeikan University in 1999. He then joined the Department of Robotics at the same university as a Research Associate. In 2000, he moved to Kagawa University as an Assistant Professor and was promoted to Associate Professor in 2007. In 2012, he became a Full Professor at the College of Information Science and Engineering, Ritsumeikan University. Since 2021, he has been a Full Professor at the Nara Institute of Science and Technology. His research interests include human–machine systems, human modeling, and human–robot interaction. He has served as Chair of the IEEE SMC Society Technical Committee on Haptic Shared Control and as Co-Chair of the IFAC Technical Committee on Human-Machine Systems.
INVITED PRESENTATION 4 (20MIN + 10MIN)
Human-centered approach to automated driving: from decision-making, motion planning to control
Asso. Prof. Dr. Daofei Li
Zhejiang University, China
Abstract
Autonomous driving (AD) is advancing rapidly. By leveraging state-of-the-art sensing, decision-making, and control technologies, the most advanced AD features currently available are capable of providing assistance close to SAE Level 3 on complex and congested urban roads (some manufacturers refer to this as Navigate on Autopilot, or NOA). However, given the diversity and unpredictability of driver and passenger needs, there is still a long way to go before fully trustworthy and satisfactory AD functions can be achieved. This talk focuses on personalized demands of drivers and passengers in terms of safety, comfort, and traffic efficiency, presenting our explorations in decision-making, motion planning, and motion control. The discussion includes research on personalized decision-making tailored to drivers' style preferences and motion sickness-aware motion planning considering passenger comfort.
Biography
Daofei Li received the B.S. degree in Vehicle Engineering from the Jilin University, Changchun, China, in 2003, and the Ph.D. degree in Vehicle Engineering from the Shanghai Jiao Tong University, Shanghai, China, in 2008. Since June 2008, he joined the Institute for Power Machinery and Vehicular Engineering, Faculty of Engineering, Zhejiang University (ZJU), Hangzhou, China. He was a Visiting Scholar with the University of Missouri-Columbia in 2011, and later with the University of Michigan, Ann Arbor, Michigan from 2014 to 2016. He is currently Associate Professor with ZJU and directs the Research Group of Human-Mobility-Automation. His research interests include vehicle dynamics and control, driver model and autonomous driving.
INVITED PRESENTATION 5 (20MIN + 10MIN)
Design of Interactive Safety Warning Information Design of External Display Interface for Autonomous Driving Vehicles
Prof. Dr. Fang You
The School of Arts and Media, Tongji University, China
Abstract
With the rapid advancement of autonomous driving technology, the communication and interaction design of external display interfaces has become a critical factor in ensuring traffic safety and enhancing human–vehicle coexistence. External displays are not only required to convey the vehicle’s intentions and status but also to provide pedestrians, cyclists, and other road users with clear and intuitive safety warnings. This talk focuses on the design of interactive safety warning information for external display interfaces in autonomous vehicles, addressing the functional positioning and interaction logic of such interfaces, the content and visual representation of safety information (including color, symbols, and dynamic cues), the application of multimodal interaction and context-awareness, and the validation of design effectiveness through user experiments and simulation studies. Building on theoretical analysis and case studies, the report proposes an AI-driven, scalable, and psychologically grounded design framework to support both the social acceptance and traffic safety of autonomous vehicles.
Biography
Fang You is a Professor and Ph.D. supervisor at the School of Arts and Media, Tongji University, where she leads the User Experience Laboratory and the Automotive Interaction Design Laboratory. She is also a Fellow of the Royal Society of Arts (FRSA). Her main research interests include cognitive interaction design, intelligent cockpit design and evaluation, hybrid space information, and embodied design. In recent years, she has led more than ten major projects, including the National Natural Science Foundation of China, the National Social Science Fund, projects funded by the Ministry of Science and Technology, and several provincial-level grants. She has published over 80 papers, authored five books and two textbooks, and holds 17 invention patents along with numerous utility and design patents. She has established long-term collaborations with companies such as Huawei, Baidu, and SAIC, as well as universities and research institutes in Germany and Australia. Her research outcomes have been widely applied in intelligent cockpits and human–computer interaction.
INVITED PRESENTATION 6 (20MIN + 10MIN)
Causal Discovery from Psychological States to Walking Behaviors for Pedestrians Interacting an AV Equipped with eHMIs
Asso. Prof. Dr. Hailong Liu
Graduate School of Science and Technology, NAIST, Japan
Abstract
In recent years, autonomous driving systems have been applied not only to cars but also to smaller robotic mobility platforms, such as Autonomous Personal Mobility Vehicles (APMVs). These autonomous vehicles (AVs) frequently encounter pedestrians during automated driving, especially APMVs, which are typically used in sidewalks and other shared spaces. To enhance the interaction experience between pedestrians and autonomous vehicles (AVs) and to reduce potential risks, it is essential to investigate pedestrians’ walking behaviors in such encounters and to understand the underlying psychological processes. In this talk, I will first introduce a study that investigates the causal relationships between pedestrians’ subjective feelings and their walking behaviors during interactions with an APMV equipped with an external human–machine interface (eHMI). Based on the causal factors identified in this study, I will then present an approach for improving pedestrian–autonomous car interactions.
Biography
Hailong Liu (IEEE S'15 - M'19 - SM'25) received his M.Eng. and Ph.D. degrees in Engineering from Ritsumeikan University, Japan, in 2015 and 2018, respectively. From Apr. 2016 to Mar. 2018, he was a JSPS Research Fellow for Young Scientists (DC2). He worked as a researcher at Nagoya University from Apr. 2018 to Oct. 2021. In Nov. 2021, he joined Nara Institute of Science and Technology (NAIST), Japan, as an Assistant Professor and was promoted to Associate Professor in Feb. 2024. His research interests focus on human factors and machine learning in intelligent transportation systems. He is a Senior Member of IEEE and holds memberships in IEEE ITSS, IEEE RAS, and IEEE SMC. He also serves on the Human Factors in ITS Committee of IEEE ITSS. In addition, he is a member of JSAE, JSAI, and SICE.
INVITED PRESENTATION 7 (20MIN + 10MIN)
Title：TBD
Prof. Dr. Adriana Tapus
ENSTA Paris, France
Abstract
TBD
Biography
Adriana Tapus is a Professor at ENSTA Paris, member of the Institut Polytechnique de Paris, in the Autonomous Systems and Robotics Laboratory of the Computer Science and Systems Engineering Unit (U2IS). In 2011, she obtained Habilitation (HDR) for her thesis titled "Towards a personalized Human-Robot Interaction". She obtained her PhD in Computer Science from the Swiss Federal Institute of Technology in Lausanne (EPFL), Switzerland in 2005 and her degree in Computer Engineering from Polytechnic University of Bucharest, Romania in 2001. She has worked as a Research Associate at the University of Southern California (USC), where she was among the pioneers in the development of socially assistive robotics, contributing mainly in machine learning, human modeling, and human-robot interaction. Prof. Tapus is Associate Editor of the International Journal on Social Robotics (IJSR), ACM Transactions on Human-Robot Interaction (THRI), and IEEE Transactions on Cognitive and Developmental Systems (TCDS) and is on the steering committee of several major robotics conferences (IROS, ICRA, HRI, RO-MAN, etc.). She has more than 150 research publications and received the Romanian Academy Award for her contribution in assistive robotics in 2010. In 2016 she was nominated as one of the “25 women in robotics you need to know about”. She has been the main coordinator of several national and international projects. Since 2019, she is also a founder member of the RoboticsByDesign lab. Since 2019, she is also the DIrector of the DOctoral School of IP Paris.
INVITED PRESENTATION 8 (20MIN + 10MIN)
IXAI: Generative Design of Automotive Styling Based on Inception Convolution with Explainable AI
Prof. Dr. Jianmin Wang
The School of Arts and Media, Tongji University, China
Abstract
This presentation will introduce IXAI — a generative automotive styling design framework based on Inception Convolution and Explainable Artificial Intelligence (XAI). The method integrates multimodal data processing, Inception Convolution-based target detection, and Grad-CAM visualization to achieve efficient, transparent, and highly accurate styling generation. Based on this framework, we have developed a creative tool, IXAI-CAR, capable of rapidly generating high-quality automotive styling design images. Experimental results show that the method achieves an accuracy of 98.3% and improves the SSIM image quality metric by 8.24%, significantly outperforming traditional approaches. The study demonstrates that IXAI not only optimizes the design process and output quality but also provides an efficient and transparent solution for human-computer collaboration in complex design tasks, advancing the application and development of intelligent design tools in the automotive styling domain.
Biography
Jianmin Wang, Ph.D., is a Professor and Vice Dean at the School of Arts and Media, Tongji Universityand a Fellow of the Royal Society of Arts (FRSA). He currently serves as Director of the Artificial Intelligence Media Research Center and Head of the User Experience Laboratory and Automotive Interaction Design Laboratory at Tongji University. His research interests focus on human–computer interaction, automotive virtual simulation, and big data visualization. He has led more than twenty national and provincial research projects, including the National 863 Program, the National Natural Science Foundation of China, and the National Science and Technology Support Program, as well as collaborative projects with leading enterprises such as Huawei, Mitsubishi Electric, and China Telecom. He holds 44 authorized invention patents, has published over 80 papers, and authored several monographs. His achievements have been recognized with the National Science and Technology Progress Award (Second Prize), the Ministry of Education Science and Technology Progress Award (First Prize), the Guangdong Science and Technology Progress Award, and the Guangdong Dingying Science and Technology Award. He has also served as a member of the Ministry of Education’s Teaching Steering Committee for Animation and Digital Media, and as a standing member of the Human–Computer Interaction Committee of the China Computer Federation (CCF).
INVITED PRESENTATION 9 (20MIN + 10MIN)
Multimodal Understanding for Autonomous Driving
Bo Jiang
HUST Vision Lab, China
Abstract
In this talk, I will present our recent research in several areas of autonomous driving (AD). I will begin with our latest multimodal end-to-end models, VADv2 and DiffusionDrive. Next, I will introduce RAD, a closed-loop, interactive reinforcement fine-tuning framework based on 3D Gaussian Splatting. Then, I will share our work on incorporating language modalities into AD. This includes Senna, a dual-system vision-language-action model, and AlphaDrive, a planning-oriented VLM based on RL and reasoning. Finally, I will outline our vision for future directions in driving-oriented VLAs and interactive driving enabled by language-based instructions.
Biography
Bo Jiang is a Ph.D. candidate at Huazhong University of Science and Technology, advised by Prof. Xinggang Wang and Prof. Wenyu Liu. His research focuses on autonomous driving and multimodal understanding. He has published 4 papers in top-tier conferences and journals, with over 840 citations on Google Scholar. His representative works include the VAD series and the MapTR series, which have garnered over 3k stars on GitHub and are widely used in both academia and industry.