Robot Talk · 6 Jan 2023 · 36m · Episode 31
Professor Chris Atkeson from Carnegie Mellon University discusses the divergence between robot 'brains' and 'bodies,' arguing that large language models will solve the cognitive side while soft robotics must solve physical safety. He explains how his research on inflatable 'balloon robots' inspired the design of Baymax in Disney's Big Hero 6, prioritizing intrinsic safety through low mass and soft materials. Atkeson also explores the concept of 'human-aware environments,' where sensors are embedded in home infrastructure rather than just the robot itself, and discusses the ethical implications of data ownership and privacy in these systems.
early-stage companies
Other shows on the same ground, closest first.
Hosts Daniel and Forbode discuss two robotics projects: a pizza-making robot from MIT, Carnegie Mellon, and UCSD that uses a teacher-student AI framework to learn dough manipulation, and an autonomous wheelchair study by EPFL that highlights safety risks and the need for assisted navigation in crowds. They analyze the engineering challenges of soft robotics, sensor integration, and the legal implications of human-robot interaction.
Brittany Postnikoff (Straithe), a researcher and community manager at Great Scott Gadgets, discusses her master's thesis on robot social engineering and her upcoming PhD research into privacy risks posed by mobile robots in public spaces. She explains how physical robots like Roombas and Knightscope security bots exploit human trust through anthropomorphism and authority to collect sensitive data, often without user consent. Postnikoff also details her role at Great Scott Gadgets, where she manages open-source contributions for hardware tools like the HackRF and the upcoming USB toolkit Luna, emphasizing how hardware sales fund open-source development.
Berndt Børnich, founder and CEO of 1X, explains the company's strategy to deploy the NEO Beta humanoid robot in homes to gather diverse real-world data for AI training, arguing that home deployment is necessary to achieve the scale and data diversity required for general-purpose intelligence. He details the robot's design principles, which prioritize inherent safety through low-energy, compliant mechanics and soft materials to ensure it is safe for human interaction, contrasting this with industrial robots. The discussion covers the challenges of regulatory compliance, the importance of public perception, and the broader societal impact of labor abundance through robotics.
Dr. Jan Liphardt, Stanford bioengineering professor and co-founder of Open Mind, discusses the transition of robotics from sci-fi to practical application through his company's modular AI architecture, OM1. He explains how Open Mind’s approach uses discrete, Lego-like functional blocks rather than monolithic models to enable transparency, debugging, and personalized social intelligence in robots like Iris. Liphardt highlights the rapid maturation of hardware components, such as reliable robot hands and self-charging capabilities, predicting widespread household deployment by late 2026, while emphasizing safety, transparency, and the importance of human-robot interaction in education and companionship.
Andrew Singletary, CEO, and Amir Sharif, COO of 3Laws, discuss their company's evolution from academic research at Caltech to commercializing safety software for robotics. They explain how their core technology, control barrier functions, provides a deterministic safety layer that safeguards AI-driven physical systems, contrasting this with traditional heuristic-based safety. The conversation covers their product lines, Supervisor Pro and Supervisor for ROS, and how these solutions help companies achieve certification and reduce costs by enabling the use of non-safety-rated sensors.
The summary, chapters and topics are generated from the episode’s audio by this directory, not written by Robot Talk. The episode itself belongs to its publisher.