Nissan are demonstrating a new lidar system, set for launch the middle of this decade, which they say will let vehicles conduct high-speed emergency maneuvers without a human driver’s help or intervention. A Nissan Skyline sedan equipped with the system dodged errant vehicles; rolling tires, and road debris, and stopped for mannequins darting into the road—all by itself, and while cruising at up to 100 km/h.
The system also has a lidar function that enables the car to self-navigate in areas where there are no clearly-defined maps or road markings. This uses something called Dynamic SLAM, short for Simultaneous Localisation And Mapping. Tetsuya Iijima, the general manager in charge of driver-assist technologies at Nissan, says the system is uniquely capable of such high-speed, super-agile autonomy, and that other systems cover only routine driving under predictable conditions. The technology, Iijima says, is key to achieving “secure autonomous driving” that can safely stop a car in any emergent situation.
The lidar system, being developed with Luminar Technologies, increases the vertical field of view to above 25°, from around 10° in today’s lidar. It increases the detection range to 300 m ahead of the vehicle, compared with around 100 to 150 m in current systems. And it delivers higher resolutions, to a level as detailed as 0.05°; twice as fine as today’s 0.1°. Cost is a hurdle; today a single lidar sensor can cost $1,000. For realistic mainstream deployment, Iijima says that must drop below $300. Nevertheless, Nissan feel lidar is practically mandatory—like virtually everybody whose opinion matters; only Tesla CEO Elon Musk scorns it. Iijima says “Secure and safe autonomous driving must have this technology” as Nissan work toward a commercially sound L3 capability; “we want to perfect these emergency maneuvers first”.
DVN comment
ProPilot systems have been provided in Europe on Nissan and Infiniti models since 2017. This L2 automated driving system will evolve towards L3 with the next iteration of ProPilot, thanks to the integration of a high-resolution lidar. This decision demonstrates automakers’ consensus that L3 AD cannot be reasonably proposed without frontal lidar. This is the only technology capable of detecting small, potentially dangerous objects on the road. We’re working on an interview for our next DVN-L Newsletter with Tetsuya Iijima about his vision on autonomous driving technology development and market implementation.