Embodied Intelligence refers to the capability in machine intelligence where intelligent algorithms are integrated with the perception, action, and environmental interaction of physical entities, enabling machines to interact with the environment and solve problems in a more natural and intelligent manner. At the ITF World 2023 Semiconductor Conference, Jensen Huang, founder of NVIDIA, boldly predicted that the next wave of artificial intelligence will be embodied intelligence.

On September 22, our institution will host the 7th lecture of the 2023 "Next-Post-Future" AI4S monthly science series. This lecture will invite Professor Lu Cewu from Shanghai Jiao Tong University to deliver an academic talk titled "Embodied Intelligence - Perception (P), Imagination (I), Execution (E): PIE Framework and Exploration of Embodied Large Models."

The lecture will introduce the PIE framework for embodied intelligence. P (Perception) will cover the research on robots’ full perception and interactive perception. I (Imagination) will present the concept-driven simulation and reasoning framework for the physical world. E (Execution) will discuss the vision and research on general meta-operation skills. Based on these three modules, the exploration and preliminary achievements of the embodied PIE large model will be introduced. Finally, the research on embodied cognitive intelligence will be shared, focusing on how to verify the stable mapping relationship between neural behavioral and physical behavioral patterns.

【Guest Profile】

Lu Cewu is a professor and doctoral supervisor at Shanghai Jiao Tong University. He was awarded the title of "Overseas High-Level Young Talent" in 2016 and was named one of the "35 Innovators Under 35 in China" byMIT Technology Review(MIT TR35) in 2018. In 2019, he received the Qiushi Outstanding Young Scholar Award. He was a co-recipient of the Shanghai Science and Technology Progress Special Award (third completer) in 2020 and won the Ministry of Education Youth Science Award in 2022. His work was selected as one of the Best Papers at IROS (6/3579). In 2023, he received a nomination for the Best Systems Paper Award at RSS (one of four nominees) and the Science Exploration Award. As a corresponding author or first author, he has published over 100 papers in top journals and conferences such asNature,Nature Machine Intelligence, and TPAMI. He serves as a reviewer for journals includingScience,Naturesub-journals, andCellsub-journals, and has been a domain chair for conferences such as NeurIPS, CVPR, ICCV, ECCV, IROS, and ICRA. His research interests include embodied intelligence and computer vision.