While Google, Tesla, Uber and many other companies are fighting for the top place in the race for self driving cars, Facebook may come ahead as they focus on Augmented/Virtual Reality (AR/VR) technologies, that indirectly will solve a lot of problems in self driving car development.

In a recent article:

SLAM: Bringing art to life through technology

They showcase a lot of focused work in simultaneous localization and mapping area (SLAM), this is an ongoing challenge in computer vision and robotics research.

SLAM can be used to localize people (and cars or any other mobile robots) mainly using their phone camera, accelerometer/gyro and GPS (where available), while mapping in realtime new areas.

They already have pretty optimized algorithms that can achieve fast recognition of objects and their 3D structure:

An algorithm tailored to each device

Our SLAM library integrates features from multiple systems (ORB-SLAM, SVO, and LSD SLAM), but what really sets it apart is the library's performance optimization, down to the very last instruction. Having a SLAM system capable of running at 60 Hz on mobile devices is hard: every 16 milliseconds, your phone has to capture an image, find hundreds of interesting key points, match them with the same points in the previous frame, and then use trigonometry to determine where each of these points is in 3D space. Since there is so much going on, we needed to do lots of fine-grained optimization and rethinking​ of how these algorithms operate.

Hopefully they will open source some of their work in this area.

At OSSDC we will also be focusing on getting SLAM and 3D Reconstruction using stereo or multi camera setup, to help detect and avoid obstacle from very close to far away (needed to predict optimum motion path).

To learn more and contribute, I invite you to join the discussions at http://ossdc.org/

For fresh news about self driving cars, artificial intelligence and robotics, follow me on Twitter at @gtarobotics