Amazon’s Scale Model Self-driving Car to Teach Developers ML

Amazon’s Scale Model Self-driving Car to Teach Developers ML
The Siliconreview
29 November, 2018

The new invention by Amazon will help developers to understand ML (Machine learning) better. This week, the company launched AWS DeepRacer, a self-driving 1/18th-scale race car. The car is designed to teach developers about ML through a learning technique called RL, reinforcement learning.

RL works when an “agent” is allowed to act on a trial-and-error basis within an interactive environment. The process works by collecting feedbacks from its actions to learn overtime as to maximize some type of score or to achieve a predetermined goal.

The process of RL is different from other ML learning methods like supervised learning. RL can get started without any labeled training data and can make short-term decisions and at the same time, it can optimize itself a long-term goal.

The new autonomous race car paved the way for developers to carry out an experiment with RL by learning through self- driving.

Theprice of the self-driving car is $399 but currently being offeredfor $249.

The self-driving car is powered by Intel Atom Processor, and is equipped with a radio-controlled system and has four wheels. It has anUbuntu’sOS 16.04 LTS, ROS (Robot Operating System) and Intel OpenVino computer vision toolkit.

Apart from that, the car is sported with a 4MP (1080p resolution) camera, a Wi-Fi connectivity of 802.11ac, multiple USB ports and its battery power last upto two hours.