Yandex has added a new transformative neural network to its autonomous cars that helps them plan their trajectories. The company’s developers said the neural network learns from data collected with the help of professional drivers to make the robots’ driving style as close to human as possible. The company believes this will enable cars to react more flexibly to challenging road situations and improve road safety.
To train the model, Yandex has assembled a team of highly qualified drivers who have undergone strict selection and training in emergency driving. During test drives, the cars’ sensors recorded each driver’s actions and road conditions. These data formed the basis of a special dataset, thanks to which the neural network learned to respond to already familiar scenarios and predict solutions for hypothetical situations.
The new planner is already being used on real city trips and is being tested in a simulator that simulates rare or complex situations. Yandex noted that its neural network planner shows impressive results – it finds safe routes even in the most difficult conditions, preventing accidents more often.
Source: Ferra
I am a professional journalist and content creator with extensive experience writing for news websites. I currently work as an author at Gadget Onus, where I specialize in covering hot news topics. My written pieces have been published on some of the biggest media outlets around the world, including The Guardian and BBC News.
