Lars Reger, Senior Vice President and Chief Technology Officer of NXP: Autonomous driving technology will be first implemented in China
The development of autonomous driving technology has entered an accelerated phase, and some even claim that 2019 is the first year of autonomous driving, which has attracted widespread attention from the industry. A few days ago, the reporter interviewed Lars Reger, senior vice president and chief technology officer of NXP, and conducted an in-depth analysis of the current state of autonomous driving and the development trend of future technologies.
Autonomous driving technology has reached L3 + level
With the issuance of a series of road test licenses, vehicle road tests have been implemented, basic data has been collected, and autonomous driving technology has also made considerable progress. It is generally believed that the current autonomous driving technology is between L3-L4 (L3 +), that is, it has a relatively weak "highly autonomous driving" capability. Under certain special circumstances, autonomous vehicles can be used without driver intervention. Drive normally.
Lars Reger, senior vice president and chief technology officer of NXP, acknowledged this in an interview. He pointed out, for example, that the current L3 + vehicles can already achieve autonomous driving on the highway. The driving system can be taken over from the high speed, and the driver can let go of his hands and feet to let the car proceed automatically. When the speed is high, the vehicle will send a notification to the driver to ask the driver to take over.
"In the next year or two, such a scenario will become popular because autonomous driving is easier to achieve on the highway. The traffic here is smooth, there are no pedestrians, and there are no strange roadblocks. To achieve L4 or even L5 levels, that is, Fully autonomous driving will take time. Because of that level of autonomous driving, vehicles will face very complex road conditions, especially under complex traffic conditions on urban roads. To achieve full road conditions and all-weather automatic driving has a very large Difficult, "Lars Reger said.
Although it will take some time for L5 autonomous driving, more and more people are still optimistic about the development prospects of autonomous driving. Regarding this, Lars Reger said: "In the past two or three years, we have seen relatively large changes in this industry. More and more vehicle companies have begun to pay attention to and invest in the development of autonomous driving technology. For example, we hope that Enzhi will participate in its development. In the construction and design of the entire vehicle. Of course, this is not to help them do mechanical design, but to help them think about how to intelligently and autonomously drive vehicles in the future to optimize the system. "
Four types of chips for autonomous driving
The chip is one of the key components for the realization of autonomous driving technology. Attracted by the huge market prospects, more and more manufacturers have accelerated their development layout in this field.
According to Lars Reger, there are currently four major types of chip technology development and development that are important for autonomous driving. First, the realization of autonomous driving must have accurate sensing capabilities, that is, sensor technology; second, there must be a strong ability to think and judge, that is, big data analysis and artificial intelligence on the car side for learning, reasoning and decision-making; third, Fast networking capabilities. If the brain on the car side is too small to cope with it, it must have the ability to connect to the cloud and get corresponding opinions; fourth, stable execution capabilities. The actuator needs to issue instructions to other systems. These are equally important and mutually reinforcing. Without such a system, autonomous driving would not be formed.
NXP is committed to the development of autonomous driving semiconductor products and solutions, including the Bluebox platform, which includes a series of chip products, such as the S32 processor, and high-performance AI accelerators. "We want to simulate human driving habits into the development process. AI helps us to train and learn. Safety is the primary indicator of automatic driving. We need to rely on data and we also need to learn and train Introducing innovative ideas and conducting a series of autonomous and creative training and learning. The combination of these two can form a relatively complete system. In addition, we need to communicate more with Tier 1 and OEM manufacturers. Together, we will explore the best solution. In this process, we will understand the future development strategies of vehicle manufacturers, understand the level at which they want to achieve autonomous driving, their time schedule, and their corresponding software readiness and complexity. In this way, we help automakers to upgrade their products in terms of functional safety and AI acceleration. We began to invest in the field of autonomous driving after we merged with Freescale in 2015. We plan to acquire Marvell's Bluetooth and WiFi businesses. , And also independently developed a variety of technologies including UWB, these are all self-driving And technical reserves. "Lars Reger said.
The industry expects that the Chinese market may be the fastest-growing region for autonomous driving technology. Lars Reger believes that Chinese consumers ’acceptance of new technologies and the social and legal environment are more suitable for the implementation of autonomous driving technology, because they are more optimistic about the development of autonomous driving in China. In addition to sensors, electronic circuits and chips, there are many social factors that will affect the development of autonomous driving. Therefore, I am very convinced that the first fully autonomous vehicle will appear in China and not elsewhere. "
Editor-in-chief: Chen Bingxin