Waymo's technology needs to test small changes over and over again, and at the same time create scenes that have never been seen before. This is the dream goal of autonomous driving technology, which is difficult to achieve in the real world, but in Carcraft, it is very easy to do .
Waymo, born out of the Google X Labs self-driving car project, has the world’s closest automated driving technology to large-scale commercial use. One of the important reasons is that the 10-year-old Waymo has accumulated 8 million miles of test mileage. This data follows As the deployment of vehicles increases and grows rapidly, more data can train more complete technologies.
One of the most important aspects of autopilot technology is testing. Testing under different road conditions, weather and environment factors, testing the environmental perception and recognition capabilities of different sensors under various conditions, and the judgment and decision-making capabilities of the control system. In order to obtain a more complete autonomous driving technology.
However, the test in the real world is still too slow. It took Waymo more than six years to complete the test range of 0-1 million miles. Even if the speed is increased now, the 1 million test mileage of the 7 million to 8 million is required. It takes a month, so is there any way to increase the speed of this test? That is Carcraft, the Matrix-like virtual reality world created by James Stout, a senior software engineer at Waymo.
Carcraft: Virtual World Test of Autonomous Driving Technology
James Stout joined Google in 2009, and in 2013 he joined Waymo, which was then Google’s X Labs self-driving project. Stout, who has a background in neuroscience and artificial intelligence,’s task is to teach a car to drive by itself. .
To accomplish this task, Scout and his colleagues started with a framework of ideas, for example, a four-way parking spot, which is a ubiquitous vehicle that is usually driven by humans, but this may involve a large number of vehicles. , And involves a complex decision-making process.
Many real-world inputs derived from the experience of Waymo cars can be obscured and diversified, multiplied, and then analyzed as edge cases. This result can be fed back to Carcraft and applied to actual Waymo tests, making the company's drivers more robust and capable.
Waymo's technology needs to test small changes over and over again, and at the same time create scenes that have never been seen before. This is the dream goal of autonomous driving technology, which is difficult to achieve in the real world, but in Carcraft, it is very easy to do .
If it fails such a test, Waymo’s real-world combination of lidar, sensors, and programming—CEO John Krafcik calls it a “driverâ€â€”can’t replicate the fact that the human brain has managed more than one. Thing of the century: driving a powerful four-wheeled machine through various environments.
Before being sent to the real world for testing, all technologies used in autonomous driving can be tested in a training room made of computer code. This is the role of Carcraft, and it can also test whether the driver mentioned by Krafcik Be competent for this job.
Currently, Carcraft's virtual fleet has 25,000 cars, which is much more than the vehicles tested in the real world. The vehicles of these virtual fleets run in Google's data center 24 hours a day. So far, these non-stop vehicle test mileage has reached an astonishing 5 billion miles.
Waymo's "driver" has an advantage over humans
According to Krafcik, the real driver of Waymo is: a non-physical robot pilot and navigator that can be installed on a variety of transportation platforms from passenger cars to semi-truck. Humans already know how to drive everything that can be driven, but humans have a poor record of safe driving: In the United States alone, nearly 40,000 people die in car-related accidents every year.
Except for professional racers, almost no one always simulates the fuzzy edge situations they may encounter when driving in their brains (even racers have to sleep when they are dreaming or wanting to drive). Self-driving cars are one of the most difficult problems computer scientists can solve, so Stout and his Waymo colleagues are using everything they know and amplifying it with huge computing power.
The Waymo autonomous driver will never feel tired, and its design better solves the cognitive difficulties of moving the vehicle from point A to point B. Scott said: "Cars are always paying attention. Human eyes evolved for hunting, but the eyes of self-driving cars have always been 360 degrees."
From just entering Google and working on Google search and Google earth, to seeing some early Google self-driving project cars, when the modified Toyota Prius was driving in the parking lot of the Google headquarters in Mountain View, Scott thought this was his dream. Now he has realized this dream, because Google allows employees to spend energy on some minor tasks, which is another successful example of Google’s 20% solution.
At the beginning, Google’s autonomous driving experimental fleet did not often run Scott’s focused simulation experiments. However, Google’s 20% solution provided Scott’s opportunity and sowed a seed for Carcraft. Up to now, this seed has been Growing into a towering tree, Carcraft has become one of Waymo's important technical capabilities.
Desktop style 12v and 24v series can be used in many different electronics. EMC, LVD, FCC, RoHS are available in our company. OEM and ODM are available, samples are free for testing, all our products have 2 years warranty.
Our products built with input/output overvoltage protection, input/output overcurrent protection, over temperature protection, over power protection and short circuit protection. You can send more details of this product, so that we can offer best service to you!
Led Adapter, Lcd Adapter,Speaker Power Adapter,Lcd Power Supply
Shenzhen Waweis Technology Co., Ltd. , https://www.laptopsasdapter.com