Friday, Apr 12, 2019
10 AM - 11 AM
EB 3105 Engineering Building
As robots are envisioned to become ubiquitous in our personal and work environments, there have been growing interests in developing intuitive ways for lay users to interact with such autonomous systems, raising several challenging questions. For instance, can we command robots in natural language? Can robots describe what they observe or explain what they have done or plan to do? How can we train our robots to understand rich semantic context, utilizing a vast amount of sensor data that is available in multiple modalities? How do we guarantee that our robots abide by social norms within a human society, e.g., how do we design our learning algorithms to prevent from developing inappropriate biases and stereotypes? In this talk, we will discuss various hurdles in addressing these questions in several examples of robotics applications including social navigation, autonomous driving, disaster response, military robots, and creative arts. We will also discuss general limitations of datasets and evaluation metrics in interdisciplinary research and propose alternative directions.
Jean Oh is a Systems Scientist at the Robotics Institute at Carnegie Mellon University. Jean is passionate about creating persistent robots that can co-exist and collaborate with humans in shared environments, continuously learning to improve themselves over time through training, exploration, and interactions. Toward this general goal, her current research is focused on the topics at the intersection among vision, language, and planning in robotics. Jean has been leading several robot intelligence tasks in government, defense, and commercial projects in various problem domains including soldier-robot teaming, self-driving cars, disaster response, eldercare, and arts. Jean’s team has won two Best Paper Awards in Cognitive Robotics at IEEE International Conference on Robotics and Automation (ICRA) in 2015 and 2018 for the works on following natural language directions in unknown environments and socially-compliant robot navigation in human crowds, in 2015 and 2018, respectively. Jean received her Ph.D. in Language and Information Technologies at Carnegie Mellon University, M.S. in Computer Science at Columbia University, and B.S. in Biotechnology at Yonsei University.
Prof. Joyce Chai