Hardware Podcasts

Episode Thirty-Nine: Maria Bauza Villalonga

Robot Talk · 3 Mar 2023 · 20m · Episode 39

What this episode covers

Maria Bauza Villalonga, a research scientist at DeepMind, discusses her work on robotic generalization and the path toward artificial general intelligence (AGI). She explains how probabilistic models allow robots to simulate outcomes and transfer skills across tasks, drawing on her PhD research at MIT’s Manipulation Lab and the Amazon Robotics Challenge. The conversation covers the importance of tactile sensing for force control, the role of embodiment in achieving true intelligence, and the scaling advantages DeepMind provides for robotics research.

Most relevant to

large enterprises

Topics

Mentioned

Companies
DeepMind, Massachusetts Institute of Technology, ABB
Products
AlphaGo, GPT
Also
Amazon Robotics Challenge

Related episodes from other shows

Other shows on the same ground, closest first.

  • The Robot Report Podcast · 1h 18m

    Practical Applications of Embodied AI and Humanoid Robots

    Dr. Russ Tedrake of Toyota Research Institute and Scott Kuindersma of Boston Dynamics discuss their collaboration to integrate TRI's large behavior models into the Atlas humanoid robot, focusing on using imitation learning and data to enable general-purpose manipulation skills. The episode also covers the FIRA 2024 ag-robotics event, highlighting Advanced Farms' apple-harvesting robot, the Reservoir Farms incubator, and new ventures like Rotate 8 and Blue White's autonomous tractor solutions.

    Robotics & autonomyAI & machine learning
    7 chapters

  • The Next Byte Podcast · 23m

    177. Did A Shrimp Fry This Rice? No, A Robot Did

    Daniel and Forbode discuss Stanford University's 'Mobile Aloha' platform, a bimanual robot designed to learn household tasks through imitation learning rather than extensive simulation. They highlight the platform's ability to generalize from a static dataset of 825 operations and improve success rates by 90% after roughly 50 human-led demonstrations. The hosts emphasize the significance of reducing the cost of such robotic systems from hundreds of thousands to around $30,000, making general-purpose robotics more accessible for future commercial and domestic use.

    AI & machine learningRobotics & autonomy
    7 chapters

  • The Robot Report Podcast · 1h 05m

    From Teleoperation to Autonomy: Inside Boston Dynamics' Atlas Training

    Scott Kuindersma, VP of Robotics Research at Boston Dynamics, explains how the company trains the Atlas humanoid robot using Large Behavior Models (LBMs) and teleoperation data. The team collected 20 hours of teleoperation data to enable Atlas to perform complex bi-manual manipulation tasks, such as picking and placing Spot parts, while adapting to real-time disturbances. Kuindersma details the four-stage pipeline of data collection, annotation, model training, and evaluation, highlighting the critical role of simulation data and human demonstrations in generalizing these skills. The episode concludes with plans to test Atlas in Hyundai facilities and further explore AI-centric dynamic behaviors.

    Embedded firmwareRobotics & autonomy
    10 chapters

  • EE Times Current · 48m

    Robots Need Physical, Not Just Artificial, Intelligence

    Emeritus Prof. Rodney Brooks, CTO of Robust AI and founder of iRobot and Rethink Robotics, discusses the necessity of physical intelligence in robotics, arguing that reactive systems and force control are superior to pure position control or simulation. He critiques the current hype around humanoid robots and large language models, emphasizing that successful products like the Roomba and Baxter industrial robot relied on pragmatic engineering and series elastic actuators rather than complex AI. The episode concludes with a discussion on the societal implications of automation and the enduring value of fundamental hardware principles.

    Robotics & autonomyMechanical engineering
    7 chapters

  • The Robot Report Podcast · 1h 04m

    General Robotics Is Betting on Modular Intelligence, Not One Robot Brain

    Sai Vemprala, CTO of General Robotics, explains how the company’s Grid platform addresses the gap between robotics research and real-world deployment by using modular, robot-agnostic intelligence rather than a single monolithic 'robot brain.' He details how Grid combines specialized models for perception, grasp prediction, and motion planning into reusable skills that can be adapted to various robotic form factors, from humanoids to wheeled manipulators. Vemprala also discusses the platform's architecture, which separates heavy reasoning on the cloud from safety-critical operations on the edge, and emphasizes that enterprises retain full ownership of their proprietary data and custom skills.

    Connected products & IoTAI & machine learning
    8 chapters

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.