How Do Robotaxis Drive Without a Driver?

Daniel sees a video of a Waymo robotaxi driving through San Francisco with nobody in the front seat. He assumes it must be following GPS. It is doing something far more interesting. The car is constantly answering four questions: Where am I? What is around me? What might happen next? What should I do? It answers all four simultaneously, in real time, without a human involved. Before Waymo operates in a new area it builds extremely detailed maps -- lane markings, curbs, crosswalks, signs and signals. While driving, the car matches what its sensors are seeing against those maps to locate itself precisely. GPS helps, but the car is also recognizing the world around it. Three kinds of sensors feed the system. Cameras give it visual detail -- traffic lights, signs, lane markings, pedestrians and cyclists. Radar measures distance and speed and works well in challenging conditions. And LiDAR fires millions of laser pulses in different directions around the vehicle, measuring how long each one takes to return, and building a precise three-dimensional picture of everything nearby -- every vehicle, every pedestrian, every wall, updated continually. The software combines all of that to identify what is around the car and estimate what might happen next. A pedestrian approaching a curb. A car drifting toward another lane. The system considers multiple possible futures and uses those possibilities to choose a safe path forward. One of the hardest unsolved problems is the long tail -- all the rare and unusual situations that are difficult to anticipate and test. A traffic officer giving unusual directions. Debris in the road. An unpredictable driver. Engineers have to prepare the system not just for ordinary driving but for an enormous range of unusual situations. Waymo has now completed more than twenty million fully autonomous rides. What you will find in this episode: How detailed maps replace simple GPS navigation What cameras, radar and LiDAR each contribute -- and why all three are needed How the system predicts what might happen next rather than just reacting Why robotaxis operate in defined areas rather than anywhere in the world The long tail problem -- and why it is the hardest challenge in autonomous driving Daniel's closing line about watching a robotaxi handle a roundabout Short, current, and the kind of episode that makes every self-driving car you see feel completely different. Listen, wonder, and learn. Find us @smilewithDaniel everywhere. [topic:tech]