So he demanded that the driver assistance software be as safe as possible before public release? paving the way for full self driving 6-7 years later? is this a bad thing?
If he demanded it was as safe as possible, he wouldn’t have refused to add lidar or radar capabilities.
For a multi-ton metal projectile that drives itself (a car), you want multiple data sources to draw a consensus from. Relying on one data source is a point of failure, and that’s not acceptable when you have the potential to kill not only the driver but others outside the car.
Exactly. Also lidar is important in instances where you need millimeter precision. Its useful for calibrating camera systems in self driving cars but in order to drive safely you don’t need that level of detail about the world around the car. It makes no difference if a car or pedestrian is 72 or 73 inches away.
I thought the “needs lidar” debate was settled years ago? Lidar cannot read signs. It is also prohibitively expensive to put in vehicles. If you’re going to drive with a neural network you need as much training data as possible, which means as many sensors in as many vehicles as possible.
If your cameras detect something the lidar does not, you trust the cameras, every time. Lidar can very easily misinterperet the world. It works great for simple robots who need to know where walls are and don’t need to specifially identify animals, people, obstacles, speed bumps, construction zones, etc.
Theres also the simple fact that humans can drive just fine without having evolved a lidar sensor.
If your cameras detect something the lidar does not, you trust the cameras
Yes, but if the lidar sees something the cameras doesn’t, you trust the lidar.
