管理科学系学术讲座（7月11日）

时   间：2025年7月11日（周五） 10:00-11:30

地   点：管理学院思源楼524室

主题：Machine Learning for Embodied Artificial Intelligence in Multi-robot Coordination

主讲人：李庆标 澳门大学助理教授

主持人：Jelmer Pier van der Gaast 复旦大学管理学院副教授

Abstract: This talk explores advanced machine learning applications in embodied artificial intelligence, focusing on decentralized multi-robot coordination.

In multi-robot coordination, we tackle scalability challenges in decentralized path planning for urban transit and logistics. We utilize Graph Neural Networks (GNNs), adept at processing topological models and non-Euclidean data, to build an information-sharing mechanism within a decentralized multi-robot system. Combining GNNs with imitation learning, we train a low-power framework approximating optimal algorithms without high computational demands. Our physical experiments involve a purely visual, machine learning-assisted navigation technology. This technology encodes relevant viewpoint information from panoramic cameras scattered in the environment through convolutional networks and communicates it to mobile robots via a GNN. Trained via imitation learning, the model effectively guides robots to their destinations, achieving performance close to the best motion algorithms. Our experiments demonstrate its universality in guiding robots in previously unseen environments with various sensor layouts.

Our work demonstrates the potential of integrating advanced computer vision and machine learning techniques to enhance the capabilities of embodied artificial intelligence in complex and dynamic environments, improving real-time decision-making, navigation, and overall system performance.

Bio:

Dr. Qingbiao Li currently serves as an Assistant Professor (Independent PI) in the Department of Electromechanical Engineering at the University of Macau, and as a affiliated Assistant Professor in the Department of Computer Science. He is also an Associate Researcher (Associate Senior Level) at the Zhuhai UM Science & Technology Research Institute and the Hengqin Advanced Institute of the University of Macau. Previously, he conducted postdoctoral research at the Oxford Robotics Institute, University of Oxford, UK. During his Ph.D., he studied at the Prorok Lab in the Department of Computer Science at the University of Cambridge under the supervision of Prof. Amanda Prorok. His research focuses on swarm motion planning and graph neural networks. Before joining the University of Cambridge, Dr. Li pursued his master's degree at the Hamlyn Centre, founded by Prof. Guang-Zhong Yang and Sir Ara Darzi at Imperial College London, where he researched medical robotics and modern medical technologies for image-guided interventions. Prior to joining the Hamlyn Centre, he graduated from a 2+2 joint program between the University of Edinburgh's School of Engineering and South China University of Technology. His work has been published in top-tier international journals and conferences such as Nature Machine Intelligence, RA-L, IJCARS, ICRA, and ICLR. He has also served as a committee member for *AAMAS 2023 - Blue Sky Track* and AAAI 2022, as well as a reviewer for IJRR, T-RO, AURO, RAL, ICRA, and IROS. In 2020, he received the Wiseman Prize from the University of Cambridge's Department of Computer Science and the 2023 Chinese Government Award for Outstanding Self-Financed Students Abroad. His representative work was awarded the Publication of the Year by the University of Cambridge's Department of Computer Science and the Best Paper Award at the CoPerception Workshop of ICRA 2023. His research has garnered over 1,200 citations on Google Scholar, with a single paper on graph neural network-based swarm robot path planning cited more than 330 times.