题目：Bayesian Robotics: Self-creating Robots and Machines Learning to Adapt
报告人：Dr. Andre Rosendo (Shanghai Tech. University)
时间： 2019年10月23日（星期三），下午4:30-5:00
地点： 广东工业大学大学城校区工学二号馆731会议室
Bayesian Robotics: Self-creating Robots and Machines Learning to Adapt
Dr. Andre Rosendo
Shanghai Tech. University
Abstract
Biological forms started primitively and, eventually, some of these forms evolved and reached intelligence. Although current artificial intelligence (AI) explores learning techniques to create thinking machines through simulations, I argue that the interaction with the real-world is essential for the convergence in true intelligence. With this in mind, I present the Bayesian Learning framework applied to robots, either by a robot capable of creating other robots and improving their performance over time, or, more recently, by our latest experiments incrementally cutting one of the legs of a robot and using Bayesian Optimization to adapt the gait of that robot. I will talk about robots capable of simultaneously altering their morphology and control and compare this method with other design ramifications, such as solely altering control parameters. The future of robotics remains uncertain, but the capacity of adapting their design and altering their morphology to fulfil tasks more effectively can be a driving force for future Robotics.
Biography
Dr. Andre Rosendo received his Bachelor at Bahia Federal University, and Master and Ph.D. degrees from Hokkaido University and Osaka university, in 2011 and 2014, respectively. In 2014 he was hired as a postdoctoral researcher by the Adaptive Robotics Laboratory in Osaka University, in 2015 he became a Research Associate at the Bio-inspired Robotics Laboratory at the University of Cambridge, and he is currently a Tenure Track Assistant Professor at Shanghai Tech. University.
His research output was featured in one special interview with Discovery Channel, two IEEE Spectrum article, he has published more than 30 scientific papers in more than 10 IEEE conferences. He volunteered as a program committee for MOD 2017, IEEE ROBIO 2017, ICAROB 2018, a keynote speaker for ICINCO 2017, and currently hold an Erdos number of 5. He was Guest Editor for the IEEE Robotics and Automation Magazine and a referee for Science Robotics, IEEE Transactions on Robotics, IEEE Trans. on Mechatronics, IEEE Intl. Conf. on Intelligent Robots and Systems, and IEEE Intl. Conf. on Automation and Robotics. His research interests are related to evolutionary robotic systems and to robotic locomotion by means of hopping, walking and running.
