A chip smaller than a coin, perceiving everything in the world.
A chip smaller than a coin is scanning the world with an accuracy of 0.1 millimeters, supporting the "digital sensory system" of the intelligent era.
Jatech shipped 6 million millimeter - wave radar chips in 2024, and the shipment is expected to soar to 16 million in 2025, with the cumulative shipment exceeding 19 million. This Chinese enterprise's market share in China has jumped from 20% to 33%, and its chips have been embedded in the intelligent driving systems of dozens of mainstream automobile manufacturers worldwide.
Meanwhile, the 0.3 - millimeter - thick flexible tactile sensor of Hanwei Technology amazed visitors at the Shenzhen Sensor Exhibition. This "electronic skin" can accurately sense pressure distribution within 1 millisecond. Imperceptibly, sensor technology is moving from the background to the forefront, becoming the core cornerstone in fields such as artificial intelligence, new energy vehicles, and industrial Internet.
01 The Competition from Millimeters to Nanometers
This month, a patent application from Wuxi XinGanZhi revealed the latest breakthrough in sensor precision. By setting a silicon island membrane structure and a hollow buffer layer in the back cavity of the pressure sensor, both high sensitivity and high pressure resistance are achieved without a significant increase in cost. This kind of technological breakthrough that "has it all" is a microcosm of current sensor innovation.
In the field of measurement accuracy, laser distance - measuring sensors are breaking new limits. The space - borne laser rangefinder developed by Jiuziyang has been applied to the Gaofen - 7 satellite, improving the accuracy to the millimeter level. Its core frequency - locked laser uses photonic crystal fiber amplification technology, achieving a wavelength stability of ±0.0001nm. The IL - 3000 series sensors of Japan's Keyence have even achieved a measurement accuracy of 0.01μm, which is equivalent to one - hundred - thousandth of the diameter of a human hair. This accuracy level has been applied to the manufacturing line of Toshiba's memory chips, enabling real - time monitoring of nanoscale deformation of the lithography machine workbench. Behind this breakthrough is the in - depth coupling of the optical interference principle and digital signal processing algorithms - the phase demodulation technology improves the optical path difference resolution to λ/1000, and the AI noise reduction algorithm eliminates environmental vibration interference.
The field of vehicle - mounted perception has also witnessed a leap in accuracy. Jatech's millimeter - wave radar combines the unique RoP (Radar Packaging Technology) with 3D waveguide antennas, achieving a leap from "being able to detect" to "detecting accurately". Its angle calculation variance is optimized by 30% compared with traditional solutions, and it can still maintain point cloud stability at a distance of 70 meters.
Moreover, sensors are achieving functional integration while improving accuracy. Jatech's newly launched Dubhe TianShuXing series UWB chips can undertake multiple functions such as digital keys, in - cabin presence detection, and kick - to - open tailgate induction simultaneously through the 2T4R radar mode. One chip replaces the traditional multi - sensor system. This integrated breakthrough stems from the in - depth collaborative design of the radio frequency front - end and the digital baseband: the time - division multiplexing technology is used to separate radar echoes from UWB positioning signals, and the dynamic beamforming algorithm is used to achieve a spatial perception accuracy of ±5cm. After the intelligent cockpit system of the BMW iX model applies this chip, the gesture control response delay is shortened to 80ms, a four - fold improvement over the previous generation product.
Behind the accuracy revolution is the collaborative innovation of materials and processes. The penetration rate of MEMS processes has increased from 35% in 2018 to 58% in 2024, promoting the development of sensors towards miniaturization and integration. Chip - level sensors are replacing traditional bulky measuring devices, bringing laboratory - level accuracy to daily application scenarios.
02 The Two - way Explosion in New Energy Vehicles and Industrial Scenarios
On June 22, Siao Sensing announced on the interactive platform that its independently developed brake pressure sensor has officially entered the stage of customer targeting and will be applied to the AEBS (Advanced Emergency Braking System). Behind this news is the explosive growth of the demand for sensors in new energy vehicles.
Sensors have become the "core senses" of intelligent vehicles. Compared with traditional fuel vehicles, the number of sensors per new energy vehicle reaches 300 - 500, and the market scale has an annual growth rate of up to 28%. This demand has directly driven the prosperity of the lidar market - the pre - installation volume of vehicle - mounted lidar in China will exceed 5 million units in 2025.
The AT128 lidar of Hesai Technology can output 1.53 million points per second and has been installed in models such as the Li L9. RoboSense has reduced the cost of lidar by 60% through chip - based technology and has received orders from automobile manufacturers such as BYD and XPeng. The annual compound growth rate of the lidar market is as high as 17.6%, far higher than the average level of the sensor industry.
As AEB regulations in China, the United States, Europe and other regions require the braking speed to be increased to over 120 kilometers per hour, radars need to have a longer detection range and the ability to identify weak targets. Jatech's dual - SoC cascade solution expands the height - measuring ability from the hundred - meter level to the full range of 350 meters through 64 virtual channels, and the detection distance of weak targets such as plastic cones is increased to over 100 meters.
The industrial field has also witnessed a sensor revolution. Data from Juyi Information Consulting shows that the networking rate of manufacturing equipment in China reached 55% in 2024, a significant increase from 38% in 2020. The networking of manufacturing equipment is the foundation for realizing the industrial Internet, and sensors are the key to realizing equipment networking. By installing various sensors such as temperature sensors, pressure sensors, and vibration sensors on equipment, real - time data on the operating status of the equipment can be collected, enabling remote monitoring, fault diagnosis, and predictive maintenance of the equipment. This not only improves the operating efficiency and reliability of the equipment but also reduces the operation and maintenance costs of enterprises. Therefore, the increase in the networking rate of manufacturing equipment directly drives the demand for industrial sensors.
In recent years, the global installation volume of industrial robots has continued to rise, reaching over 500,000 units for many consecutive years. From 2018 to 2023, the compound growth rate was 5%. According to the prediction of the International Federation of Robotics (IFR), by 2027, the installation volume of process robots will exceed 600,000 units. The widespread application of industrial robots has driven the growth of the demand for displacement measurement sensors. Displacement measurement sensors can accurately measure parameters such as the position, speed, and acceleration of robots, providing important basis for the precise control of robots. As industrial robots develop towards high precision, high speed, and intelligence, higher requirements are put forward for the accuracy and reliability of displacement measurement sensors. The demand for displacement measurement sensors increases by 22% annually, becoming an important growth point in the industrial sensor market.
03 The Birth of New Species
Sensor technology itself is constantly breaking through the boundaries of innovation. While traditional sensors are meeting the needs of large - scale applications through performance upgrades and cost optimization, "new species" of sensor technology based on new materials and new principles have quietly emerged.
The team of Du Cuifeng/Wang Yuan from the University of Science and Technology Beijing, in collaboration with the team of Academician Wang Zhonglin/Zhu Laipan from the Beijing Institute of Nanoenergy and Nanosystems, developed a Triboelectric Self - Powered Sensing Platform (TESS). It has been operating continuously for several months at a depth of 626 meters underground in a gold mine in Shandong. This technological breakthrough based on the triboelectric nanogenerator (TENG) solves the power supply problem for mine monitoring.
The ability to sense wind speed is achieved through a non - contact TENG - based horizontal turbine. Meanwhile, TESS consists of a unique set of TENGs and operates in a new working mode, balancing the advantages of the contact - separation and freestanding modes. With an optimized self - driven power management system, TESS achieves a charging power density of 16.36 mW per square meter; this energy is transmitted every 166 seconds to a sensor node (for measuring temperature, relative humidity, pressure, and the concentrations of carbon dioxide, nitrogen dioxide, and ammonia), a data processing unit, and a LoRa transmitter. This work has made a major breakthrough in developing a robust, low - cost, battery - free, and wireless triboelectric nanogenerator - based environmental monitoring platform.
Meanwhile, there is also good news in gas sensing technology. The "electronic nose" system developed by Hanwei Technology can quickly identify different odors within 5 seconds by integrating a highly sensitive gas sensor array and an AI algorithm, and has been applied in fields such as food safety and environmental monitoring.
The system uses a multi - sensor collaborative working mode. Each sensor has high sensitivity to specific gases, and the AI algorithm is used to analyze multi - dimensional data in real - time, breaking through the limitations of traditional single - gas detection. For example, in the food safety scenario, the "electronic nose" can judge the freshness or spoilage of food by identifying the gas components volatilized from the food; in environmental monitoring, it can quickly locate the source of harmful gases or odors in the air, providing data support for pollution early warning.
This bionic sensing technology mimics the complex recognition ability of biological olfaction, can adapt to various gas - mixing scenarios, and has higher detection efficiency and practicality compared with traditional sensors, promoting the development of gas sensing technology towards intelligence and multi - functionality.
04 Where Will Sensors Go?
According to the Yole report, the global MEMS (Micro - Electro - Mechanical System) industry reached a critical turning point in 2024. With the "inventory effect" subsiding in the second quarter, the annual revenue reached $15.4 billion (a 5% year - on - year increase), and the shipment volume exceeded 31 billion units. Market research shows that as the terminal demand in consumer electronics, automotive electronics and other fields recovers, coupled with the drive of new technologies such as AI and the Internet of Things, the industry growth rate is expected to further increase in 2025. It is predicted that the compound annual growth rate from 2024 to 2030 will reach 3.7%, and the market scale will reach $19.2 billion in 2030, with the sales volume increasing to 35 billion units.
While the MEMS industry is achieving steady growth through traditional application scenarios, quantum sensing technology is brewing a breakthrough in the laboratory.
Quantum sensors use quantum phenomena. Compared with traditional sensors, their sensitivity is greatly improved, opening up various new applications, including electric vehicles (EVs), non - GPS navigation, medical imaging, and communication. Industry experts call this the "Second Quantum Revolution".
Experts believe that while quantum mechanics has great application potential in the field of computing, it may also completely change the sensing industry.
The sensitivity of the quantum state makes it possible to detect tiny changes, enabling measurement with unprecedented accuracy. Quantum sensors can measure various physical properties such as current, electric field, magnetic field, light, linear acceleration, angular acceleration, and time with higher sensitivity compared with traditional sensors.
One example is the tunnel magnetoresistance (TMR) sensor, which can be produced in millions of chips. It is currently sold in the automotive field for remote current sensing. Although biomagnetic imaging using an optically pumped magnetometer (OPM) is still in the early stage of development, it has shown good potential, and startups such as Cerca Magnetics have sold early products and prototypes to research centers.
In addition, there are desktop atomic clocks, which have been used in research and international time standards for many years. In the medical field, quantum sensors can simply measure the natural magnetic field of the heart, providing much more data than current electrocardiogram (ECG) machines. While ECG devices measure through electrodes directly attached to the skin, quantum sensors can be integrated into clothes, mattresses, and other items.
Moreover, the tentacles of technological innovation have quietly extended into living organisms - this leap from "external environmental monitoring" to "internal biological perception" lays a technological foundation for the in - depth integration of brain - computer interfaces and sensors, and makes the real - time capture of neuronal activity move from scientific imagination to clinical possibility. The integration of brain - computer interfaces and sensors opens up new possibilities. The implanted device of Neuralink integrates a micro laser distance - measuring module to monitor neuronal activity in real - time with an accuracy of 10 nanometers. This cross - field technological integration expands sensors from external environmental monitoring to internal signal collection of living organisms.
From nanoscale measurement of 0.01 micrometers to precise manipulation of the quantum state, from precision instruments in the laboratory to mass production of automotive - grade chips, that chip smaller than a coin is not only the "digital eyes" of intelligent vehicles, the "tactile nerves" of industrial robots, but also the "quantum probe" to explore the mysteries of life. When the accuracy of sensors breaks through the nanoscale threshold, what we measure is no longer just physical distance, but also the cognitive boundary from silicon - based to carbon - based, from machines to life - the cornerstone of the Internet of Everything will eventually move from "connection" to "perception".
This article is from the WeChat public account "Semiconductor Industry Insights" (ID: ICViews). Author: Fang Yuan. It is published by 36Kr with authorization.