A GRASP Lab seminar host and panelists Nadia Figueroa, Dinesh Jayaraman, and Mark Yim discuss robotics research, deployment challenges, and future opportunities at the University of Pennsylvania.
The host traces milestones from the first factory robot in 1961 to Shakey, early autonomous driving, Honda’s humanoid, drones, and today’s AI-enabled systems. He describes robotics as a growing but relatively small industry, with major uncertainty about what comes next.
Jayaraman argues that robots should combine learning from watching, imitation, and feedback, drawing on how people and animals adapt. A panelist sees language and vision-language models as tools for teaching robots and shaping their perception. Looking five years ahead, Figueroa expects more shared control between people and robots, while Jayaraman anticipates cheaper hardware that ordinary users might train for simple tasks.
Yim points to severe limits on tiny robots used inside the body, including scarce room for energy storage and memory. Panelists caution that impressive demonstrations do not establish reliable performance: robots must handle unpredictable people and environments, where the final stages of safety and robustness are difficult to validate. Figueroa also notes gaps between sensor speeds and control needs, along with fragile sensing and integration challenges.
Figueroa says people need ways to understand what a robot is doing and to guide it, potentially through language and clearer interfaces. Jayaraman favors interpretable representations, checks during multi-step tasks, and tests designed to expose failures. The discussion points toward lower-risk starting points such as toys and small-scale manufacturing, while safety requirements should vary with the task and the objects involved.
Briefs summarise talks and interviews; timestamps open the video at that moment. Claims in a talk are the speaker's own.
Dr. Dinesh Jayaraman, a faculty member in computer and information science at the University of Pennsylvania and GRASP Lab, discusses visual robot learning and ways to make robots more capable in uncertain settings.
Dr. Nadia Figueroa, a new GRASP faculty member at the University of Pennsylvania, presents her research on collaborative, human-aware robotics and the challenges of making robots learn, adapt, and work safely with people.
