Thursday, July 25, 2024, 10am

As robots become increasingly integrated into human society, ensuring their safety is a paramount concern that has not been thoroughly addressed. This thesis tackles crucial questions on providing formal safety guarantees for robots operating in complex real-world scenarios and enabling these robots to perform diverse tasks intelligently and safely. By developing novel methodologies that build upon safe control, model-based control, and large language models (LLMs), my research bridges the gap between theoretical safety guarantees and practical implementation. This thesis presents three significant contributions: creating reliable safety measures and strategies, extending safety assurances to complex robotic systems, and achieving reliable open-world task execution. The benefits of this research are substantial, offering a framework for deploying robots that can safely and efficiently interact with their surroundings, thereby enhancing their utility and trustworthiness in various applications and fostering their integration into human society.

Thesis Committee:
Changliu Liu (Chair)
Yuejie Chi
Yorie Nakahira
Andrea Bajcsy
Chuchu Fan (Massachusetts Institute of Technology)

In Person and Zoom Participation.  See announcement.

Event Type: Thesis Orals
Room Number: In Person and Virtual
Building: Newell-SImon 3305 and Zoom
Speaker's Name: TIANHAO WEI
Speaker Website: www.ri.cmu.edu…
Speaker's Professional Title: Ph.D. Candidate, Department of Electrical and Computer Engineering, Carnegie Mellon University
Talk Title: Safeguarding and Empowering General Purpose Robots
Affiliations: Robotics Institute (RI)
Organization(s): Department of Electrical and Computer Engineering, School of Computer Science