This is old news, but I don't think it was ever posted here.
Gasgoo:
“Based on the previous technical R&D and the iteration of road tests for the first half of 2019, the new sensing system that is composed of 10 cameras will help Baidu's vehicles to conduct the end-to-end closed-loop autonomous driving on urban roads sans high-beam rotatable lidar, said Wang Liang.
Baidu revealed that the Apollo Lite has been tested on multiple road sections in Beijing and reached the world-leading level in terms of number of vehicles tested, cumulative testing mileage and vehicle's performance in closed-loop autonomous driving on urban roads.”
VentureBeat:
It's not 100% clear to me whether Apollo Lite is merely a means of achieving “true redundancy” or a system intended for commercialization.
Gasgoo:
“Based on the previous technical R&D and the iteration of road tests for the first half of 2019, the new sensing system that is composed of 10 cameras will help Baidu's vehicles to conduct the end-to-end closed-loop autonomous driving on urban roads sans high-beam rotatable lidar, said Wang Liang.
Baidu revealed that the Apollo Lite has been tested on multiple road sections in Beijing and reached the world-leading level in terms of number of vehicles tested, cumulative testing mileage and vehicle's performance in closed-loop autonomous driving on urban roads.”
“According to Baidu, Apollo Lite can process “vast” amounts of data generated by 10 cameras to detect objects up to 700 feet away while delivering real-time, 360-degree sensing of the environment. In tests on public roads in Beijing, vehicles relying on Apollo Lite managed to drive sans lidar sensors, which measure the distance to target objects by illuminating them with laser light and measuring the reflected pulses. (Lidar forms the foundation of a number of autonomous car systems, including those from Uber and Lyft.)
A strictly vision-based approach to autonomous driving is one advocated by Intel’s Mobileye, which is developing a custom accelerator processor chip — EyeQ5 — that offers 360-degree coverage courtesy proprietary algorithms, cameras, and ultrasonic. Similarly, driverless semi truck startup TuSimple says its camera-based technology (which employs lidar largely for redundancy) has a 1,000-meter detection range.
A strictly vision-based approach to autonomous driving is one advocated by Intel’s Mobileye, which is developing a custom accelerator processor chip — EyeQ5 — that offers 360-degree coverage courtesy proprietary algorithms, cameras, and ultrasonic. Similarly, driverless semi truck startup TuSimple says its camera-based technology (which employs lidar largely for redundancy) has a 1,000-meter detection range.
A robust vision-based system is critical to the safety of autonomous driving, especially in high-speed situations where real-time sensing is critical,” said Apollo’s technical committee head Liang Wang. “Apollo Lite further strengthens Baidu’s sensor fusion based level 4 autonomous driving system that leverages the capabilities of camera, lidar, and radar to achieve the ‘true redundancy’ necessary for a safe and fully autonomous driving experience.”
The VentureBeat article is somewhat inaccurate or at least unclear. Mobileye is an advocate of a “true redundancy” approach in the sense of developing a camera-only system independently from a radar and lidar system.
It's not 100% clear to me whether Apollo Lite is merely a means of achieving “true redundancy” or a system intended for commercialization.
