The average sales growth rate of domestic sensors

The average sales growth rate of domestic sensors in the next five years will reach 31%

statistics show that the value of chips in each car soared from $250 in 2000 to $350 in 2017. With the development of intelligence, electronization and automatic driving, the chips used in cars continue to grow. The importance of chips for the automotive industry is increasing. Sensors, memories, MCU, communication chips... Have been integrated into the automotive industry. This is not the same as the correction point, and has become an indispensable part. What is the current development of automotive chips? What trends are showing? How can Chinese chip enterprises seize development opportunities? This newspaper plans a series of reports on the "core" of smart cars to explore its development context

with the development of science and technology, simple human-computer interaction has been difficult to meet people's needs. In order to save driving time and cost and improve traffic safety, autonomous driving technology came into being. Engineers use sensors to continuously collect information about the surroundings of vehicles and traffic conditions, and transmit it to the cloud through car couplets. They use big data and artificial intelligence to analyze the data, and then send control commands back to the vehicle to execute commands. This is a simple data process of intelligent driving. It can be seen that the sensor in the starting position of the data flow plays a decisive role, so how does it support the whole system

with the blessing of automatic driving, the importance of on-board sensing has increased.

automatic driving requires sensing the surrounding environment. How to successfully collect data into the auto drive system in different environments is the basic function of the sensor

there are three types of on-board environmental sensors: one is conventional sensors, such as yaw rate sensor, acceleration sensor, wheel speed sensor, etc; One is the sensing sensor, which is installed on the car body, such as radar, camera and ultrasonic wave; There is also a class of virtual sensors, representative of which, such as digital maps and wireless communications, belong to this category

at present, sensing sensors are popular in the market. It is understood that among many sensing sensor devices, millimeter wave radar (forward) can achieve the maximum range of sensing, followed by forward-looking camera. However, not all cameras can perceive a wide range of environment. They can be divided into single eye, double eye, three eye and other categories according to the medium and long-distance scenes. The perceived objects include lane lines, traffic signs, obstacles, pedestrians and so on. In addition, it will also be affected by changes in external conditions such as light and weather. On the other hand, the camera has high requirements for the software and hardware of the processor, and the relevant technical barriers have become the focus of major manufacturers to break through

the camera needs the support of image sensors for information acquisition. At present, there are many kinds of image sensors in the market, and their functions are also different. Italy France semiconductor and adasky, an Israeli far infrared imaging technology (FIR) startup, announced a technical cooperation agreement to produce a customized chip using ST's proprietary 28 nm fd-soi (silicon on fully depleted insulator) process, and then integrate the chip into adasky infrared thermal imager to make an image sensor called viper. It is understood that viper can solve the most difficult problem of image sensors at present: how to recognize roads and surrounding environment in bad weather and dim light

with the help of the camera, Viper uses the fir micro bolometer sensor to detect the temperature of the target object, and uses the unique algorithm based on convolution neural network to classify the obstacles, so as to overcome the shortcomings and weaknesses of the sensor in some light or weather conditions. According to Fabio marchio, vice president of Italian French semiconductor department and general manager of automotive digital products department, viper's powerful infrared vision sensor and high-value algorithm add a high-value complementary information layer to the existing sensors. "Driving conditions are still one of the main concerns of autonomous driving. How to solve the challenging difficulties brought by formal conditions is exactly the direction we are working hard." FabI Jinan testing machine factory cupping machine new standard o marchio said

with the growth of the market, international giants have entered the market.

affected by the rapid development of the market, major international intelligent automobile and sensor enterprises have focused on the field of automotive intelligent sensors. Google was the first company to study smart car technology. It uses sensors to obtain 3D street view maps. Through its environmental sensing system scheme, combined with 360 degree scanning lidar sensors, it can complete the information collection of the surrounding environment. Google calculates the physical information such as the distance, speed and angle between the smart car and the surrounding target objects through the top mounted lidar sensor. According to expert analysis, this part of the lidar is one of the reasons why it is difficult to industrialize the smart car. In addition to the high cost, the performance and stability of the car will be affected to a certain extent in bad weather such as fog, rainstorm and blizzard

the sensors used in smart cars developed by continental Germany are another type of Representative - multi-sensor data fusion technology. The technical scheme is to sense the environmental information around the vehicle through low-cost environmental sensing sensors such as millimeter wave radar and camera. In the whole vehicle, there is a smart antenna module, which can communicate the information of the whole vehicle with the outside world

Bosch is also one of the largest sensor manufacturers in the world. Its products mainly cover millimeter wave radar. At present, the technological development is concentrated on 24GHz and 77GHz. 24GHz is mainly used in the rear of cars to realize short-range detection, and 77GHz is used in the front and side. "Bosch is the world's largest supplier of MEMS sensors. The future car service is a major direction, and there are many different challenges, but Bosch is ready to explore product applications and provide more suitable solutions to Bosch's use and design creators." Leopold beer, President of Bosch Asia Pacific, said at the bma400 new product launch

Lin Yahui, an analyst at Jibang Tuo Industrial Research Institute, said that with the designation of dedicated frequency bands by countries, 79ghz is expected to replace 24GHz. In addition to millimeter wave radar, Bosch also has mature products in ultrasonic radar. Bosch ultrasonic radar has the advantages of low cost, strong penetration and small attenuation, and there is still room for progress in the development of miniaturization technology

China's automotive sensor industry needs to master the core technology as soon as possible

sensor fusion is bound to be a trend of future development. Automatic driving continuously collects environmental information outside the vehicle through sensors, judges moving and stationary objects in the surrounding environment, detects and tracks moving objects, and then judges whether the target will pose a threat through artificial intelligence. The amount of information collected by the whole system is huge, which can not be supported by one sensor or another. Therefore, it is necessary to fuse different sensors, optimize the weighted value of each sensor, improve the accuracy of output target data and reduce the error caused by the sensor itself. "Each sensor has its own advantages, but as for a single sensor, it is still limited by the interference of factors such as weather and vehicle high-speed driving, which makes sensor fusion and combined vehicle coupling (v2x) become the key direction of automatic driving technology development." Lin Yahui said

in addition, solving the problem that multiple sensors are difficult to fully meet the vehicle's perception of the surrounding environment information in unstructured roads will also be a focus of sensor development. Solving this problem in the market requires intelligent algorithms such as machine learning or neural networks, which will bring technical barriers to computing speed and information storage

"Sensing fusion refers to that the processor can collect and calculate a variety of signals at the same time, so as to meet the complex environmental perception needs, help the vehicle make safer control instructions; combined with the car Union, it helps the vehicle send heading information with the surrounding vehicles, pedestrians, infrastructure, etc. in real time, and help other vehicles plan their paths. The above two technologies will be an important auxiliary to realize automatic driving, and effectively improve the safety of the auto drive system." Lin Yahui said that plastic materials and products in the EU market may be removed from shelves or notified because they do not meet the requirements of the new regulation

at present, millimeter wave radar is the best in China's on-board sensors, and the manufacturers include Yidao, centritec, zhuotai, etc. In the vehicle lens part, some companies also carry out technical research and development, such as Suzhou Zhihua, meiruisi, Zhejiang Haikang and Beijing Jingwei Hengrun. But when it comes to overall strength, China's on-board sensor technology is still backward

liuzhaobin, former chief engineer of the General Administration of quality supervision, inspection and quarantine, proposed to rapidly improve the standardization level of sensors in China. He said that China's sensor industry is generally at the middle and low-end level. In order to transform the IOT and sensor industry from the middle and low-end to the middle and high-end, it is necessary to vigorously enhance the synergy of the industrial value chain and strive to grasp the pricing power of corresponding products

"China's automotive sensors have always been highly dependent on imports, and the development of related industries lags behind European and American countries, and has not yet formed an independent industry. For the sensors produced in China at present, there are still problems such as low sensing accuracy and weak anti-interference. Coupled with the lack of primary suppliers in China, it is difficult for sensor manufacturers to systematically enter large-scale vehicle factories, and can only basically meet the needs of small batches and low-end vehicles." Lin Yahui said

with China becoming the world's largest automotive market, the overall scale of sensors is bound to expand. "In the next five years, the average sales growth rate of domestic sensor market will reach 31%. Automotive electronics will be the main driving force for the rapid growth of sensor demand in China." Luo Baihui, chief analyst of Jinmo industrial control, said

the huge market scale does have the opportunity to drive the development of domestic automotive sensors. "In particular, China does not lag behind European and American countries in terms of algorithms and data in the development of autonomous driving. In order to better integrate software and hardware, international autonomous driving solution providers have also begun to seek Chinese partners, so sensor practitioners should actively expand their cooperative networks with solution providers and enter the supply system of vehicle manufacturers in a sub-system way." Lin Yahui said

on the other hand, the Sino US trade friction is dominated by automotive electronics, and automotive sensors are the main part of the limitation scope of the US tax list. Trendforce Jibang consulting believes that China's exports to the United States are small, and the impact of trade friction can be almost ignored. In the short term, it may even be conducive to Chinese enterprises to obtain space and opportunities for independent scientific and technological innovation and enhance the strength of independent intellectual property rights

"I haven't seen the impact yet, and it may take some time. According to what I know, at present, European and Chinese devices still account for a large proportion in this sensor market. There are many kinds of sensors, and there are many Sensor Companies in the United States." Qin Liang, marketing manager of Italian French semiconductor micro motor product department, said

but in the long run, when the industry enters the stage of white hot competition