Titanium media news Recently, Nissan Motor Company announced the research and development progress of new driver assistance technologies. This technology can use a highly accurate, real-time vehicle surrounding environment information perception system to effectively reduce accidents and improve the level of automatic driving. After the technical launch, Takao Asami, senior vice president of Nissan Motor Company, and Tetsuya Iijima, director of the advanced technology research and development department of AD/ADAS, were also interviewed.
Nissan's "ground truth perception" technology combines information perceived by next-generation high-performance lidar (LIDAR), radar, and cameras to detect the shape and distance of targets in real time and accurately, as well as the spatial structure of the area around the vehicle. Using this information, the vehicle can analyze the current situation in an instant to make a judgment and automatically perform collision avoidance operations. At the same time, the technology can also detect deceleration traffic and road obstacles in the distance, and automatically change lanes. Importantly, in the absence of detailed map information, the technology can still provide drivers with strong driver assistance support.
Takao Asami said: "Nissan is the first company in the world to launch a number of advanced driver assistance technologies. Giving drivers confidence in the safety performance of vehicles is key to the future development of autonomous driving technology. Nissan's ground truth perception technology will play a crucial role in improving driver confidence, reducing traffic accidents, and driving the development of autonomous driving in the future. ”
In addition, Nissan Motor is working with leading companies to develop driver assistance technologies and share relevant experiences. Nissan Motors is working with Luminar to integrate next-generation lidar (LIDAR) into Nissan's driver assistance systems using leading technology. In addition, in the field of advanced collision control, it is also crucial to develop high-precision verification technology in a digital environment. Based on this, Nissan motor vehicle is cooperating with Applied Intuition, which has advanced simulation technology.
According to nissan Ambition 2030, Nissan is using next-generation lidar (LIDAR) to develop vehicle control technology to reduce the occurrence of traffic accidents. Nissan believes that in the coming era of autonomous driving, assisted driving technology that can avoid complex traffic accidents will be indispensable. Nissan plans to complete the development of ground truth perception technology in the mid-2020s. This technology will be implemented first on some new models and is planned to be applied to all new models by fiscal year 2030.
Timing planning for the deployment of Nissan's driver assistance systems
Takao Asami: "Nissan has sold more than 1 million vehicles equipped with driver assistance systems. ProPILOT super intelligent driving technology launched in 1996, the main application scenario is the highway, this technology requires the driver to hold the steering wheel, if it is separated from the steering wheel, it can not be operated. Introduced in 2019, ProPILOT 2.0 supports two-handed off-wheel driving, allowing drivers to drive with both hands off the wheel on the highway. ”
"This time we are introducing the automation technology of emergency avoidance operation, whether it is assisted driving or full automatic driving, which will greatly improve safety performance. Regarding future commercialization, I think there are multiple options for whether this technology will be suitable for autonomous vehicles, or for assisted driving vehicles, or both. But we hope to be able to complete the development of ground truth perception technology in the mid-2020s, which will be installed first on some new models and can be installed on all new models in fiscal 2030. At that point, this technology will significantly improve safety performance, so we are also considering applying it to driver assistance systems. ”
Which partner does Nissan LiDAR choose? How many lidars will be used?
Takao Asami: "Our partner is Luminar, of course, we have not identified Luminar as Nissan's mass production supplier, but as a co-development partner. ”
Tetsuya Iijima: "Nissan will only carry one lidar on the roof. Lidar detects the space around the vehicle and the vehicles on the road, that is, moving objects. The road structure and surrounding buildings are actually static, and the surrounding scene changes when the vehicle is moving. Lidar can sense changes in the scene and make three-dimensional measurements to communicate to the vehicle. Therefore, equipped with a lidar on the roof of the car, it is basically possible to grasp the surrounding space knots. ”
What solution does Nissan have to prevent traffic accidents?
Takao Asami: "At present, Nissan's driving assistance technology is L2 level, that is, the driver needs to drive. For vehicles that require a steering wheel in their hand, if the driver's hands are off the steering wheel, the system will issue a warning, and in the worst cases, the system will automatically judge the braking of the vehicle. For systems that allow the driver to take both hands off the steering wheel, the driver monitoring system also monitors whether the driver is looking ahead, and if the driver is not involved in driving, the system will stop such behavior. In the future, the road truth perception technology introduced this time can accurately identify the position of stationary objects, so that vehicles can avoid collisions with them during driving. ”
Tetsuya Iijima: "First of all, I will explain the identification of stationary objects. For cameras and radars that have now achieved mass production, the detection range is relatively limited, and stationary objects can be detected within a range of about 60 meters. But if it is a long distance, such as 100 meters, 150 meters or 300 meters, it is difficult for cameras and radar to recognize stationary objects. Therefore, when driving on high-speed roads, cameras and radar cannot detect stationary objects, resulting in accidents. The lidar introduced this time is designed to be able to spot stationary vehicles at a speed of 130 kilometers per hour and safely avoid them, so that it can capture stationary vehicles and objects 300 meters away. ”
How will Nissan reduce the cost of lidar?
Takao Asami: "Regarding cost, it is indeed a big challenge. We are currently developing technology to automate emergency avoidance operations and are already in the process of evaluating lidar performance and developing algorithms. On the issue of reducing costs, I think lidar may go through several upgrades until it is popularized in the market to significantly reduce costs. I think it is possible to achieve a cost reduction at this point in 2030. ”
Tetsuya Iijima: "Regarding the performance degradation of lidar in bad weather, because it is an optical sensor, if there is dirt or bird droppings in the front, it will not work properly." Therefore, it is necessary to install a wiper, washer or water washing device in front of it. However, for rainy weather, since lidar uses a very fine laser beam for scanning, it does not mean that a single water droplet falling vertically into space blocks all the beams each time, it fills the scene in front of it in several cycles, and can also be said to be a soft reproduction of the image behind the raindrop. Therefore, in terms of current performance levels, lidar is more powerful than radar and cameras in response to deteriorating performance in bad weather conditions. ”
How Nissan adapted to China's traffic environment
Takao Asami: "The point of whether or not there will be a collision with an object may be the same all over the world, but China's road conditions do have their own peculiarities." Nissan has set up an R&D base in Shanghai, and the team is currently putting considerable effort into conducting simulations based on China's existing traffic environment to verify whether the technology can be applied to China's road environment. This is very necessary, theory alone will not work, you must collect a lot of data. And I also think that it is necessary to conduct validation experiments on the spot, so I hope that the cutting-edge team in Shanghai will be able to develop in a pioneering form. ”
This article was first published on the Titanium Media APP, edited by | Wang Zhenchao