Driverless Car Perspectives
Morning, Everyone. It's a great honor to speak to you today. My name is Olivia, major in Pattern Recognition and Intelligent System, mainly research driverless car. So, today, I'm here to tell you a little bit about how self-driving cars worked and the future of driverless cars.
Do you know that driving accidents are the no.1 cause of death? And do you realize that more than 90% of those are due to human error? Do you know that almost 200,000 people die on the roads in China each year, at a cost of over 1.5 billion dollars? Through driverless cars, we can avoid most of the car accident and regain the time we waste on the daily commute.
Now, I'll talk about how our self-driving car worked and what the future may become with these cars.
Firstly, let's talk about how our car sees the world. So, this is our vehicle, on the roof of the car, is a laser range finder, by which it magically can see everything around it. The rotating laser scanner like this, we called it 64-beam laser, can generate a detailed 3D map of the surrounding world instantly, accurate down to 2 centimeters. The car then combines the laser instruments with high-resolution maps of the world, producing different types of data models that allow it to drive itself while avoiding obstacles. Several computers inside the car's trunk perform split-second measurements and calculations, processing data from the sensors, and make decisions about every aspect of driving.
This image shows the 3d-data captured by the lidar. Our car managed 300,000 miles of testing autonomous travel without an accident, which far exceeded human driving. Our car has proven that the future is there, or at least, a short drive away. Now, the next question is what it will become if we release the car to decide which route to take?
Current navigation applications such as Google Map focus on optimizing for a single user - the driver - when crunching numbers to find the shortest or the fastest path. If the drivers are human, this is, of course, fine if they don't follow the apps' recommend. But once we replace the driver with the machines, they will definitely listen and follow the recommendations. If we can control all of the cars, then our navigation algorithm could optimize for the network flow, instead of a simple distance for a single driver, and will ease traffic jams, reduce speeding tickets and car accidents, waste less energy because of instead efficiency.
Soon, we may not be in the driver's seat, if the laser and computer take control, driverless car let us "just have fun" while the car does all the work. We no longer waste our time on the daily commute, we can do anything in the car, like dealing with official business, playing games or even having a lunch. Given how easily humans are distracted, someone died from a car-related accident every 15 minutes. We let software code take the wheel, some worry travel become so safety and so convenient.
After my presentation, I hope you can understand how driverless cars worked and the importance and the inexorable trend of the development of driverless cars. And here, is my reference. Thanks for your listening, and any questions, please?