From Embodied and Physical AI to Autonomous Drone Swarms

Explore the evolution of artificial intelligence from digital interfaces to physical-world applications in this comprehensive podcast episode featuring Dr. Denis Garagić, CTO at Palladyne AI. Discover how embodied and physical AI systems are revolutionizing robotics, manufacturing, and autonomous operations by moving beyond traditional screen-based interactions to real-world problem-solving. Learn about the fundamental differences between embodied AI and conventional automation, examining how adaptive, embedded systems are replacing rigid, pre-programmed robots in industrial settings. Understand the critical role of data processing, edge computing, and multi-modal perception in enabling true autonomy in physical environments. Examine the limitations of large language models in industrial and physical applications, and discover why hybrid autonomy approaches often deliver superior performance in production environments. Delve into the mechanics of autonomous drone swarms, including decentralized decision-making processes, edge-based planning systems, and self-organizing UAV fleet coordination. Explore practical applications across automotive, aerospace, defense, and construction industries, as well as public safety implementations. Address the operational, organizational, and safety challenges inherent in deploying embodied AI systems in production environments, including workforce preparation, verification and validation processes, and regulatory compliance considerations. Gain insights into the future trajectory of embodied and physical AI, covering world-models, runtime memory systems, simulation-to-reality transfer techniques, and emerging industry developments that will shape the next generation of intelligent physical systems.