Andreas Orthey Podcast

Everyone talks AI, but what is intelligence, actually?Join me and Prof. Oliver Brock, director of the Science of Intelligence excellence cluster, for a deep-dive into the science of intelligence, robotics, protein folding, and his startup side quests.Here are my three personal highlights from this podcast:- Oliver believes that intelligence manifests itself through behavior, and that behavior depends on a small set of principles. One of those principles, he believes, is active interconnection, where information flows between different parts of a system, guided by optimization.- He recounts his truly inspiring life-story, including how we grew up in the 80s in West Berlin, what he did when the wall came down, how he found his way into Stanford University, and how he became the CTO of a company raising $200M in funding.- By linking protein folding to robotics, Oliver explains how Deepmind's AlphaFold achieved a breakthrough in finding stable protein folding states, what Levinthal's paradox entails, and that necessity has driven nature to excel at protein folding.Outline02:14 Growing Up in Berlin in the 80s08:26 Path Towards Academia17:17 Founding AllAdvantage27:23 Professorship in Amherst, MA29:15 Protein Folding and AlphaFold37:54 Interactive Perception44:42 Intelligence, Behavior, and Braitenberg Vehicles47:21 Definition of Intelligence49:02 Multiple Intelligences53:33 Intelligence as a Process55:15 Work Turing Test57:19 Principles of Intelligence: Active Interconnection01:05:59 Science of Intelligence01:10:16 Future of RoboticsThanks and CreditsSpecial thanks to the Science of Intelligence excellence cluster, especially Maria Ott, for providing the podcast room for this episode and help with setting everything up.Unfortunately, the audio of this episode got scrambled during recording due to a microphone malfunctioning. I had to reconstruct roughly 90 percent of this episode using AI tools. To do this, I used ElevenLabs.io to generate a transcription of the scrambled audio (which was surprisingly good) and then used their voice cloning tool to regenerate the audio in Oliver's voice. Unfortunately, the voice cloning introduced an accent to Oliver's voice, so the voice you hear is not the original voice of Oliver. However, the content is preserved and has not been altered from the original. Furthermore, this process led to some mismatch between lips and audio, which made the video stream unusable. I used different tools like facefusion.io, sync.so and diff2lip to finally restore the lips to be in synchronization with the audio from the voice cloning. The final video still has some audio and video artifacts plus the additional accent for Oliver's voice, but it was the best that could be achieved with the tools available to date.Books MentionedMind Children by Hans MoravecComputers & Thought by Edward FeigenbaumThe $2 Million Cough by Greg Hoffman [About the company AllAdvantage]Vehicles: Experiments in Synthetic Psychology by Valentino BraitenbergAction In Perception by Alva NoëEpisode LinksScience Of Intelligence: https://www.scienceofintelligence.de/Oliver Brock Homepage: https://www.tu.berlin/en/robotics/about-rbo/prof-dr-oliver-brockRobotics and Biology Laboratory: https://www.tu.berlin/en/roboticsWikipedia AllAdvantage: https://en.wikipedia.org/wiki/AllAdvantageThe Work Turing Test: https://citeseerx.ist.psu.edu/pdf/75d3340c6dbd7bcd8c4b4c617acb5d932b03f4baPodcast LinksYoutube: https://www.youtube.com/andreasortheySupport: https://buymeacoffee.com/aortheyPodcast Website: https://aorthey.com/src/podcast

Majid Khadiv is Professor for AI Planning in Dynamic Environments at Technical University Munich (TUM). In this podcast, we talk about multiple topics connected to the future of humanoid robotics, including:- Which company is winning humanoid robotics?- What are the major application areas for humanoid robots?- The power of visualizing dynamical equations- How to recover on flat terrain by only a single step into the future.- The ideal education of students who want to study humanoid robotics.Enjoy the podcast!Outline01:05 Getting into University in Iran03:28 Visualizing Mathematical Equations13:35 PhD in Iran19:55 From Iran to Germany26:11 Humanoid Locomotion32:41 From PhD to Postdoc42:40 The Ideal Education48:55 Biggest Societal Problems50:06 Warfare and Technology53:10 Applications for Humanoid Robots54:49 General-Purpose vs Specific-Purpose Machines57:33 Moravec's Paradox01:01:20 Main Bottlenecks for Humanoid Locomotion01:12:23 Model Predictive Control01:14:35 Problem of Infinite Choice01:17:32 Contact Before Motion or Motion Before Contact01:20:05 Which Company Will Win Humanoid Robotics?01:23:16 When We Will Get Room Cleaning Robots01:24:11 Data Acquisition01:28:33 Quadrupeds at CERN01:30:21 Cleaning Rooms01:32:28 Problems With Unemployment01:35:48 Basic Income01:39:53 AI Taking Over01:41:24 Humanoid Robot Personalities01:43:24 Favorite Philosophers01:49:55 Current Research and Longterm VisionEpisode LinksMajid Khadiv Homepage: https://www.professoren.tum.de/en/khadiv-majidMajid Khadiv Lab Homepage: https://www.ce.cit.tum.de/en/aipd/home/Podcast LinksYoutube: https://www.youtube.com/andreasortheyX/Twitter: https://x.com/andreas_ortheySpotify: https://creators.spotify.com/pod/profile/andreasorthey/LinkedIn: https://www.linkedin.com/in/aorthey/Support: https://buymeacoffee.com/aortheyPersonal Website: https://aorthey.comPodcast Website: https://aorthey.com/src/podcast

The global drone market is projected to reach 260 billion USD by 2030 [1]. Are you curious about the technology driving this revolution?I recently sat down with Wolfgang Hönig, professor for multi-robot coordination at the TU Berlin, for an in-depth discussion on the future of robot drones. In our podcast episode, we discuss- When will drone package delivery become a Reality?- How to overcome the complex physics of drone flight- Coordinating swarms of drones for maximum efficiency - Why bird-like drones could be a game-changer- How to build a cutting-edge drone research lab- How drones can revolutionize search and rescue operationsCheck out our full conversation below!Content00:35 Research Goals01:17 Motivation to do Robotics Research01:58 Early Life in Dresden03:20 First Programming Experiences05:20 From Computer Science to Embedded Systems06:36 Master Thesis with NVIDIA07:38 Software Engineer at NVIDIA09:11 Lessons Learned at NVIDIA10:17 Favorite Coding Environment10:43 Largest Company in the World in 202511:25 From NVIDIA to Academia12:26 Meeting His Wife13:17 Difficulty of Getting into a PhD Program14:08 Moving to Los Angeles15:09 Multi-Robot Coordination Lab16:21 PhD work on Drones17:06 Virtual Reality and Drones18:02 Motion Planning for Multi-Robot Teams19:22 Main Challenges for Motion Planning on Drones20:35 Understanding the Physics of Multiple Drones22:13 Motion Planning for Drones25:46 Combining Planning and Learning 27:32 Drone Racing29:22 Applications of Large Swarms of Drones30:30 Where is my Package Delivering Drone?33:46 Bird-like Drones35:10 Advantages of Bird Designs36:10 How to Build a Drone Lab38:27 Demonstrations in the Drone Lab39:31 Difference between Drone Lab and Real World40:39 Drones in Warfare41:59 Drone Search And Rescue42:52 Larger Drones for Search and Rescue44:59 Civilian Applications of Drones46:35 Drone Lifetime and Battery Requirements 48:07 Connections to Material Science49:35 Current Research Projects50:36 Most Impactful Paper Ever Written51:42 Next Career Milestones52:15 Prerequisites to do Drone Research52:58 Value of H-Index Metric53:58 A Typical Day in Wolfgang's Life55:03 Weekly Events in the Research Groups55:32 Personal Definition of Success57:01 Mentors58:09 Counterbalance to Professorship01:00:08 Life Lessons LearnedReferences[1] https://www.sphericalinsights.com/reports/drone-marketEpisode LinksSpotify: https://open.spotify.com/episode/3wLmvai2GLfgcbyDlYtWm8?si=_hf8o__eQn-OjGlLMKsILAApple: https://podcasts.apple.com/us/podcast/wolfgang-h%C3%B6nig-multi-robot-coordination-drone-physics/id1812910570?i=1000708579725Wolfgangs Personal Website: https://whoenig.github.io/Wolfgangs Lab: https://imrclab.github.io/Lab Youtube Channel: https://www.youtube.com/channel/UCsFZxMGOnS8pC6DSeRVnDLgPodcast LinksYoutube: https://www.youtube.com/andreasortheyX/Twitter: https://x.com/andreas_ortheySpotify: https://creators.spotify.com/pod/profile/andreasorthey/LinkedIn: https://www.linkedin.com/in/aorthey/Support: https://buymeacoffee.com/aortheyPersonal Website: https://aorthey.comPodcast Website: https://aorthey.com/src/podcast

How do you build a robotics company? Join me in my latest podcast episode, where I talk with Sean Murray, co-founder and director of robotics engineering at Realtime Robotics.Our topics include:- His past as a chemical engineer- Motion planning on a chip- The challenges of starting a robotics company- The importance of early feedback in product development- Hiring and retaining the right people- How academic and industrial motion planning differs- The next 10 years of industrial roboticsPlease enjoy!Content00:37 From Montana to Berlin01:18 Born in Texas02:26 First Contact with Computers03:10 Chemical Engineering and Biotechnology04:10 Interests in Bioinformatics05:37 Reservoir Engineering 07:31 Challenges in Reservoir Engineering11:12 PhD in Computer Science13:05 Benefits of having a different background14:09 Goals in PhD15:33 Robot Motion Planning on a Chip18:53 Process of Starting a Robotics Company 20:29 Vision of First Product21:56 Not what the industry wanted22:57 First Customer23:38 How to acquire Funding24:50 Biggest Challenge to start Realtime Robotics25:50 On Culture and Hiring26:38 Philosophy of Hiring People28:02 Retaining Good People in a Startup Environment29:23 Building Technical Expertise and Retain it30:31 Prerequisites to work at Realtime Robotics31:40 Critical Skills in a Startup Environment32:35 Differences between Academic and Industrial Motion Planning35:22 Resolver for Workcell Optimization38:19 Significance of Early Feedback40:25 Current Challenges in the Robotics Industry41:27 Balancing Shortterm and Longterm Goals42:13 The Next 10 Years43:28 AI in Robotics44:44 Impact of Large Language Models in Robotics45:40 Advice for your younger Self46:39 Value of Pivoting Faster47:17 Automation and Fears of Unemployment 49:05 Which jobs will become Obsolete?49:32 Could Basic Universal Income Unemployment?51:35 Exercises to Balance out Sitting on a DeskEpisode LinksSpotify: https://open.spotify.com/episode/3lJBV0in1MDvWrmm61DyWi?si=_idjSqaoR-iWG9PDdjEFdQApple: https://podcasts.apple.com/us/podcast/sean-murray-how-to-build-a-robotics-company-robot/id1812910570?i=1000706674729Podcast LinksYoutube: https://www.youtube.com/andreasortheyX/Twitter: https://x.com/andreas_ortheySpotify: https://creators.spotify.com/pod/profile/andreasorthey/LinkedIn: https://www.linkedin.com/in/aorthey/Support: https://buymeacoffee.com/aortheyPersonal Website: https://aorthey.comPodcast Website: https://aorthey.com/src/podcast

Imagine tiny robots moving through your bloodstream, breaking down blood clots, repairing damaged cells, or delivering targeted drugs to tumor sites.This isn't science fiction---it is the real-world potential of microrobotics, a field on the verge of revolutionizing robotics research and transforming medicine as we know it.In episode #3 of my podcast, I sit down with leading expert Brad Nelson, Professor of Robotics and Intelligent Systems at ETH Zürich, to explore the fascinating world of robots at the smallest scales.Content00:38 Beginnings of Microrobotics: Manipulation at Small-Scale02:45 Grand Challenges in the Field: Small, Intelligent Machines03:27 Medical Applications of Microrobots04:08 Regulatory Challenges: Sterility05:08 Personal Motivations: From the Space Program to Microrobotics06:43 First Robotics Project: Assembly Workcells and Force Control09:00 PhD in Robotics: Microscopic Robots10:14 Curiosity and Forces at the Smallest Scales12:06 First Interest in Medical Robotics15:13 Two Leading Causes of Death: Heart Attacks and Ischemic Strokes16:09 Move Micro Robots: Magnetic Field Control18:11 From Blockage to Relief: CTs, Nanoparticles, and Microneedles21:25 Detection of Strokes: Face, Arm, and Speech Test23:06 Formation of Blockages23:55 Prevention of Blood Clots: Smoking, Diet, and Exercise25:28 Femtotools: Force Sensors at Small Scales27:19 Aeon Scientific and Nanoflex Robotics28:03 Magnetic Fields: Dangers and Weak Fields29:36 Effects of High-Frequency Magnetic Fields30:18 From Research to Product: Challenges and Lessons learned32:49 Building Exceptional Teams: Passion, Teamwork, and Uncertainty34:11 Ideal Research Student: Excitement, Collaboration, Hard Work35:39 Structure of Daily Work Days37:18 What is the Smallest Scale for Robots?39:40 The Limit: One Micrometer40:21 Alternatives to Magnetic Fields for Actuation41:11 Applications outside the Medical Field43:08 How realistic is the movie Transcendence?45:00 Current State on Transporting Drugs through the Body47:09 Micro-robots in the Brain: What can we already do?48:18 What Bradley is most proud of48:45 What do you need to break into Micro-Robotics?50:31 How to create a good interdisciplinary team?51:59 How do LLMs affect the field of Micro-Robotics?54:00 Current Research: Robotics Capsules and Teleoperations56:34 Time in United States Peace Corps: Botswana and Teaching Math58:56 Moving from Minnesota to Switzerland01:00:52 Why Robotics is AmazingEpisode LinksSpotify: https://open.spotify.com/episode/522KaV5Os08rHJUNTjaRhn?si=0jgsp-7HRcacRZlCxyCW7QApple: https://podcasts.apple.com/us/podcast/bradley-nelson-microrobotics-in-medicine-andreas-orthey-3/id1812910570?i=1000706674728Bradley Nelson Website: https://msrl.ethz.ch/the-lab/team/Brad_Nelson.htmlPodcast LinksYoutube: https://www.youtube.com/andreasortheyX/Twitter: https://x.com/andreas_ortheySpotify: https://creators.spotify.com/pod/profile/andreasorthey/LinkedIn: https://www.linkedin.com/in/aorthey/Support: https://buymeacoffee.com/aortheyPersonal Website: https://aorthey.comPodcast Website: https://aorthey.com/src/podcast

Steve LaValle is one of the pioneers of robot motion planning. I got the rare opportunity to talk with him in my podcast series on robotics.Topics included his work on the Planning algorithms book, development of the rapidly-exploring random tree algorithm, minimal sufficient robot brains, doing motion planning research with his son, and his new life in Oulu, Finland. Please enjoy!Content00:40 Early Childhood04:59 Getting into University07:27 Life at University10:13 Symbolic vs Visual Mathematics11:11 Getting into Phd Program12:12 How to get into MIT for students13:09 Advice from Phd Advisor14:17 First Paper16:30 Options after Phd17:47 Advice to Phd Students19:19 Postdoc in California21:46 Demo for Bill Gates22:52 Becoming Professor at Iowa State23:34 Developing RRT26:21 Reason of longterm success of RRT28:34 Sampling-based vs. Randomized30:30 Could RoboticsGPT deprecate motion planning?31:53 Value of motion planning research33:07 Information spaces34:10 Minimal sufficient robot brains35:37 Writing "Planning algorithms"40:33 Managing and organizing references42:08 Deprecated parts of the book43:16 Challenges in writing more books44:19 Unpublished RRT book44:28 Reading and favorite books45:50 Reading Ayn Rand46:53 On joining Oculus VR50:51 Virtual reality research52:24 Killer application for VR52:54 Facebook buys Oculus53:31 Control over time is most important54:20 Meeting Brett Leonard55:30 Huawei57:16 On being successful59:30 H-Index as performance indicator01:00:16 Metrics for academic career01:01:36 Time allocation01:02:28 Tired phase of career01:03:26 Advice to younger researchers01:04:40 Fascination with Finland01:07:46 Learning Finnish01:08:12 What Steve misses about the US01:08:55 Benefits from Saunas01:10:18 Future aspirations of lab01:11:21 Research with Alexander LaValle01:13:15 Robotics applications for daily life01:14:04 Gardner hype cycle of deep learning01:15:18 Value of material science for robotics01:16:12 Closing commentsEpisode LinksSpotify: https://open.spotify.com/episode/1tuJ6w35CCIkyXjsviFTtc?si=tKfYa5MXQXKFmnUivc5C4gApple: https://podcasts.apple.com/us/podcast/steven-m-lavalle-robotics-research-virtual-reality/id1812910570?i=1000706674834Podcast LinksYoutube: https://www.youtube.com/andreasortheyX/Twitter: https://x.com/andreas_ortheySpotify: https://creators.spotify.com/pod/profile/andreasorthey/LinkedIn: https://www.linkedin.com/in/aorthey/Support: https://buymeacoffee.com/aortheyPersonal Website: https://aorthey.comPodcast Website: https://aorthey.com/src/podcast

I recently talked with James Kuffner at the ICRA conference in Rotterdam in the Netherlands. Topics included getting into Stanford, his contributions to robot motion planning, robotics in japan, the google self-driving car project, Woven city, and his advice on becoming successful.Content00:00 Introduction01:24 Background02:38 Getting into robotics04:55 Programming at 12 years old05:52 Getting into Stanford07:30 Robotics at Stanford12:19 From computer vision to motion planning13:59 RRT-Connect algorithm20:31 Is research in motion planning still valuable?22:48 Will LLMs deprecate planning?26:23 Timeline for truly intelligent robots28:10 Breakthroughs required for AI32:30 Cloud robotics34:47 Going to Japan37:04 Technological progress and craftsmanship in Japan39:34 Work culture in Japan43:49 Professor at Carnegie Mellon University46:03 Collaborations at CMU47:25 Self-driving project at Google51:53 Coding mastery at google53:51 Code search tool55:09 Vi versus Emacs56:27 Full self-driving58:44 Who is ahead in self-driving technology?01:01:06 Achieving level 5 autonomy01:04:51 Behavior of robot cars and the median driver01:06:30 Impact of self-driving on labor01:10:37 Biggest problems for society01:13:36 Woven City by Toyota01:15:41 Job loss after Fukushima earthquake01:18:01 A city from scratch01:19:29 Climbing mount Fuji01:20:42 Inventor garage01:23:11 Startup incubator and beyond01:23:57 Success definition01:26:18 Daily habits01:27:59 Pulling all-nighters01:30:00 Advanced Shakespeare01:30:58 Advice for early career researchers01:31:59 Value of books01:32:31 Linus Pauling: Change fields every 7 years01:33:37 Next big project01:34:34 Influential books01:36:59 The value of biographies01:38:23 Personal goal of 50 books per year01:39:43 Biggest contribution to scienceEpisode LinksSpotify: https://open.spotify.com/episode/75jmWRsM7K2XkCiZr0zsNP?si=JauAqp1FQ5Ol19UPr1HHugApple: https://podcasts.apple.com/us/podcast/james-kuffner-toyota-woven-city-robotics-in-japan-google/id1812910570?i=1000706674948Podcast LinksYoutube: https://www.youtube.com/andreasortheyX/Twitter: https://x.com/andreas_ortheySpotify: https://creators.spotify.com/pod/profile/andreasorthey/LinkedIn: https://www.linkedin.com/in/aorthey/Support: https://buymeacoffee.com/aortheyPersonal Website: https://aorthey.comPodcast Website: https://aorthey.com/src/podcastErrata49:16 James wants to add that the CMU car went 7.32 miles and not about 17 miles in the first DARPA grand challenge.1:24:13 The success quote is from Ralph Waldo Emerson and not E. Pugh. The original quote is: "What is success? To laugh often and much; to win the respect of intelligent people and the affection of children; to earn the appreciation of honest critics and endure the betrayal of false friends; to appreciate the beauty; to find the best in others; to leave the world a bit better, whether by a healthy child, a garden patch Or a redeemed social condition; to know even one life has breathed easier because you have lived. This is to have succeeded!" -- Ralph Waldo Emerson