While other companies like Google, Uber, Ford, etc. keep their self-driving cars project heavily guarded. The Chinese search engine giant has announced that its self-driving platform is now available for free. Anyone who wants to be a part of self-driving car’s revolution can download the software for and use it for free.
In April 2017, Company announced that it’ll open source its self-driving software, the company kept its promise and finally released the software.
Baidu’s Apollo software provides an Autonomous Driving Platform to anyone who wants to create a self-driving car.
The Apollo platform comes with a massive collaboration from different institutions and companies from across the China. Users can download Apollo from its Official Github repository. Some statistics suggest that becoming the first company to open source autonomous driving platform will make Baidu a leading company in this field.
The name of the project is inspired by NASA’s 1960’s to 1970’s Apollo missions. In fact, project’s README.md starts with the famous quote by J.F. Kennedy:
“We choose to go to the moon in this decade and do the other things, not because they are easy, but because they are hard.”
The Apollo projects came with 7 big components that include Intelligent and Precise Control, Open Data Platform, HD Map and Localization, DuerOS, Simulation, Accurate Percepion and Security. let’s go deep down into 4 out of 7 major components provided by Apollo.
Intelligent and Precise Control:
The vehicle uses Intelligent and Precise Control to provide intelligently optimized driving plans by using experienced drivers as its deep learning reference. It provides intelligently optimized driving plans by factoring real-time traffic, speed limits etc. The control accuracy can reach down to 10 cm.
HD Map and Localization:
Based on GPS, IMU, HD map, and a variety of sensor inputs, Apollo is one of the few platforms that produce HD mapping data on a large scale. It uses deep learning and artificial intelligence technology to produce HD mapping data. The localization system gives centimeter-level accuracy to the vehicle, which can be customized with software and hardware.
DuerOS provides a complete voice control of the vehicle by using speech recognition system. The DuerOS provides control of navigation, virtual Q&A assistant, personalized audio content recommendations, etc.
According to Apollo developers, Apollo provides a complete security framework which separates the vehicle’s networks from outside networks. It’s deep learning system make sure that only trustworthy commands are executed.
The company hopes that by the year 2020 self-driving cars won’t be running on just simple urban road conditions. Baidu believes by the year 2020 self-driving cars will reach Highways and Open City Roads.
Do you like this decision made by Baidu? Share your views in comments.