Baidu Research affirms that its autonomous driving project is in similitude with the Apollo missions of NASA in the 1960s and 1970s. In fact, the readme in their Github repo starts with the famous quote by John. 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.”
Their goal of developing fully autonomous cars by 2020 is equally ambitious. However, it is not new. The dream for completely autonomous cars is decades old. To remind you, it has been ten years since DARPA unveiled the Urban Challenge in 2007. With a massive increase in computing power and AI capabilities, this goal is all the more realizable.aidu Research has released its Open Source Autonomous Driving platform called Apollo. It believes that open sourcing poses a win-win situation; community development is faster for testing and deployment, therefore, pacing the project, forbye more participants implies more travel data. Armed with massive collaboration with other institutions, Apollo is ready to transform the Autonomous Driving Industry.
Features of Apollo
Developers of Apollo have built a platform capable of supporting versatile tasks like intelligent control, precise sensing, and localization and cloud ecosystem in which the vehicle and the cloud are unified.
Apollo runs entirely on ROS (Robot Operating System) and supports C++ and Python; however, Python 3 support is currently unavailable. It provides localization and basic controls. As for the hardware, it supports GPS and IMUs (for locomotion and orientation).
The platform is clearly in its early days. The crucial elements of autonomous driving like perception and motion planning are still lacking. Furthermore, support for basic sensors like LiDAR, RADAR, Camera is still under development.
