A best of episode where Russ interviews computer scientist and electrical engineer, Dorsa Sadigh. They had a fantastic conversation about the work she’s doing to train robots to better understand humans, and as she shares, it turns out that one key to this work is better understanding human behavior. If you’re curious about how we’re going to make human-robot interaction work, this is a great episode to tune into again. Enjoy.
Episode Reference Links:
Connect With Us:
Chapters:
(00:00:00) Introduction
Host Russ Altman introduces guest Professor Dorsa Sadigh, a computer scientist and electrical engineer at Stanford.
(00:02:24) Bridging Robotics and Human Psychology
The journey from focusing solely on robotics to incorporating human behavioural insights to enhance robot design and functionality.
(00:05:31) Blending Cognitive Science and Robotics
The integration of cognitive science with robotics to build better computational models of human behaviour.
(00:07:35) Addressing Suboptimal Human Decision Making
Exploring human decision-making biases and their impact on robot interaction, with focus on bounded rationality and prospect theory.
(00:10:39) Robot Adaptations to Human Imperfections
How robots can adapt to human imperfections in collaborative tasks, using theories from behavioural economics.
(00:14:57) Training Robots and Humans for Better Interaction
Strategies for teaching both humans and robots to improve their interactions, including active teaching and understanding each other’s capabilities.
(00:18:41) Partner Modelling in Robotics
Insights into partner modelling in robotics, demonstrated through a scalable model in an air hockey game setup.
(00:21:54) Complex Multi-Agent Interactions
Addressing the dynamics of multi-agent systems like traffic, where autonomous and human-driven vehicles interact.
(00:24:11) Robots in Healthcare
The role of robots in healthcare, particularly in assistive technologies, and the challenges and advancements in this area.
(00:26:26) Conclusion
Connect With Us:
Episode Transcripts >>> The Future of Everything Website
Connect with Russ >>> Threads / Bluesky / Mastodon
Connect with School of Engineering >>>Twitter/X / Instagram / LinkedIn / Facebook
