#97 – Sertac Karaman: Robots That Fly and Robots That Drive

Autonomous vehicles face significant hurdles when transitioning from controlled environments to human-populated spaces, necessitating advancements in both software and societal integration. Sertac Karaman, an MIT professor and co-founder of Optimus Ride, argues that scaling autonomous systems requires moving beyond simple navigation to mastering complex human-robot interactions and game-theoretic decision-making. While simulation remains a critical development tool, particularly for training perception algorithms, real-world deployment relies on iterative learning within geofenced environments to ensure safety and efficiency. Furthermore, the debate between camera-only systems and LiDAR-integrated approaches highlights a fundamental trade-off between computational intensity and hardware reliability. Ultimately, achieving mass deployment involves balancing technological innovation with public trust, as robots must learn to navigate unpredictable human behavior while operating in environments originally designed for human use rather than machine autonomy.