Editor's Note: When smart vehicles achieve L2+ advanced driver-assistance systems (ADAS) on urban roads, when industrial robots accurately perform millimeter-level operations in factory workshops, and when medical devices capture human physiological signals in real time through non-invasive monitoring—the scenes behind all these are inseparable from sensors, the "perceptual nerves of the intelligent era." In recent years, the global sensor market has grown at an average annual rate of over 8%. As one of the world's largest sensor application markets, China's market size reached 460.86 billion RMB in 2025, according to the "China Sensor Market Research Report" released by CCID Consulting in June 2026 [1]. At the policy level, the "14th Five-Year Plan" explicitly lists sensors as a key development area for "new-generation information technology," and local governments have also introduced special support policies. For a time, the sensor industry seemed to stand at the cusp of a boom, with a surge in the number of enterprises and a flood of capital, presenting a prosperous scene.
However, beneath the surface of prosperity, deep-seated contradictions and problems in the industry are becoming increasingly prominent. According to relevant articles in "Economic Reference" and "China Electronics News," there are currently about 1,600 domestic specialized enterprises in sensitive components and sensors [2], with fewer than 20 enterprises having annual revenues exceeding 1 billion RMB [3]. The vast majority of enterprises are concentrated in the mid-to-low-end sensor fields such as temperature and pressure, resulting in severe product homogenization. In the high-end market, the localization rate of industrial-grade MEMS sensors is less than 15%. According to the "Greater China MEMS Industry 2025" report released by Yole Group, the high-end development of China's domestic MEMS sensor industry is highly insufficient: in 2024, the MEMS wafer output of China's domestic market (excluding China’s Taiwan region) accounted for only 2% of the global total [4]. Core components such as automotive-grade LiDAR and millimeter-wave radar still rely on imports.
More alarmingly, as AI (Artificial Intelligence) and IoT (Internet of Things) technologies evolve rapidly, sensors are upgrading from "single sensing devices" to "intelligent sensing terminals." However, domestic enterprises significantly lag in innovation capabilities for intelligent functions such as chip integration and edge computing, leading to the dilemma of "not knowing how to use, not easy to use, and unwilling to use" at the downstream application end. These problems do not exist in isolation but are intertwined and transmitted layer by layer, jointly restricting the transformation of China's sensor industry from "scale expansion" to "quality improvement."
This article categorizes these problems into three major types: "weak technological foundation," "suboptimal industrial ecosystem," and "pressure from the external environment." It deeply analyzes the manifestations and root causes of the ten core issues and proposes systematic breakthrough paths, aiming to provide a reference for the high-quality development of the industry.
Weak Technological Foundation: Still a Long Way from "Following" to "Leading" The competition in the sensor industry is essentially a competition of technological foundation and comprehensive strength. As an interdisciplinary field integrating physics, chemistry, biology, materials, microelectronics, software algorithms, and more, sensors have extremely high technological barriers. However, the domestic industry still has obvious shortcomings in key areas such as underlying technology, intelligent upgrading, high-end products, and testing verification, leading to the prominent problem of being "large but not strong."
Issue 1: External Dependence on Core Mechanism Technologies and Insufficient Underlying Innovation The core mechanism of a sensor is the "root technology" that determines its performance, such as the application innovation of basic physical/chemical principles like the piezoelectric effect, photoelectric effect, and Hall effect, as well as the R&D of new sensors based on new mechanisms. For a long time, domestic sensor research has mostly focused on the application integration of "from 1 to N," with a severe lack of investment in the underlying sensitive mechanism innovation of "from 0 to 1." Taking MEMS sensors as an example, for core manufacturing processes such as deep silicon etching, multi-layer heterogeneous integration and bonding technologies, domestic enterprises still mainly adopt mature foreign process routes or overall industry follow-up designs, with a scarcity of original micro-mechanical structures; core manufacturing processes such as deep silicon etching, heterogeneous integration, and bonding are highly dependent on overseas technological routes; in the field of flexible sensors, although the number of domestic papers on sensing mechanism research based on new materials such as carbon nanotubes and graphene leads the world, from the perspective of industrial implementation, there is a severe scarcity of industrialized achievements of new, independent, and original sensitive mechanism technologies in China. A large number of laboratory mechanism schemes are difficult to complete pilot scale-up and large-scale mass production, and there is a significant fault between basic research and commercial applications.
This tendency of "valuing application over foundation" keeps domestic enterprises in a "following" position in technological iteration. When foreign enterprises launch high-performance sensors based on new mechanisms, domestic enterprises often need to spend 2-3 years on reverse engineering, by which time foreign enterprises have already occupied market dominance through patent barriers and economies of scale. Although there are numerous domestic enterprises, the "collective voicelessness" in core mechanism technologies makes it difficult for the entire industry to break free from the dilemma of "low-end lock-in."
Issue 2: Lagging Upgrade of Smart Sensors and Insufficient AI Adaptation Capability As AI technology evolves rapidly, sensors are upgrading from "single data collectors" to "intelligent sensing terminals." Traditional sensors only output raw data, while smart sensors need to integrate MCUs (Microcontroller Unit) and AI algorithm modules, possessing functions such as data preprocessing, edge computing, and self-diagnosis, to achieve the integration of "sensing-computing-analyzing-decision-making." For example, smart industrial sensors can perform real-time anomaly detection on collected vibration and temperature data, directly sending warning signals to the control system; smart vehicle sensors can improve environmental perception accuracy through multi-modal data fusion, reducing the computing load of autonomous driving systems.
However, the development of domestic smart sensors significantly lags behind market demand. From the perspective of hardware integration, over 80% (the vast majority) of smart sensors produced by domestic enterprises adopt a simple hardware integration combination scheme of "sensor chip + externally purchased MCU," while leading foreign enterprises have achieved single-chip integration of "sensor + dedicated AI chip," reducing the volume by 40% and power consumption by 50%. From the perspective of algorithm capability, domestic enterprises' smart sensors mostly rely on general AI algorithm models, with insufficient optimization for specific scenarios, resulting in data processing accuracy 10%-15% lower than foreign products and response speeds more than 20% slower. In practical applications, there is a certain gap in data processing accuracy compared to mature foreign products, and the overall response speed is also relatively lagging.
This "insufficient intelligence" directly leads to the "three nos" dilemma at the downstream application end: "not knowing how to use"—the interface protocols and data formats of smart sensors are not unified, requiring application enterprises to develop additional adaptation software, which accounts for over 30% of the total system integration cost [5][6]; "not easy to use"—insufficient algorithm accuracy leads to a high false alarm rate. An automotive parts enterprise reported that when using domestic smart pressure sensors, the false alarm rate of the braking system reached 0.5%, while imported products were only 0.05%, showing more stability and an obvious gap; "unwilling to use"—the comprehensive performance of domestic smart sensors has a gap with imported products, but the price is only 10%-20% lower, with no obvious cost-performance advantage. For example, a NEV (New Energy Vehicle) manufacturer once tried to use domestic smart battery sensors, but because the algorithm could not accurately predict the battery SOC (State of Charge), it ultimately had to switch back to imported products.
Issue 3: Insufficient Supply of High-End Products and a Long Way to Go for Import Substitution The sensor market presents an obvious "pyramid" structure: the top of the pyramid consists of high-end sensors such as industrial-grade, automotive-grade, and medical-grade sensors, which have high technological barriers and high added value. Their global market share is less than 20%, but their profit share exceeds 60%; the base of the pyramid consists of mid-to-low-end sensors for consumer electronics, which have low technological thresholds and fierce competition, with a profit share of less than 20%. Domestic enterprises have long crowded into the base market, resulting in a severe shortage of high-end product supply.
In the industrial field, for laser displacement sensors used in precision manufacturing, the measurement accuracy of domestic products is mostly above ±5μm, while the accuracy of products from Germany's SICK and Japan's KEYENCE has reached ±0.1μm. According to data from the "2025 China Laser Displacement Sensor Market Share and Industry Competition Pattern Analysis Report" by Boxun Zhishang, in 2024, the market share of domestic laser displacement sensor manufacturers in the domestic market has increased from less than 10% in 2015 to 38% [7]; in the automotive field, according to data from the "2026 China Intelligent Driving Industry Chain Map and Investment Layout Analysis" by the China Commerce Industry Research Institute, the market size of China's millimeter-wave radar in 2024 was about 9.8 billion RMB, the localization rate of corner radars has exceeded 55%, and the share of forward radars has risen to 13.3% [8]; for automotive-grade millimeter-wave radar, although domestic enterprises have achieved mass production, they are 1-2 technological generations behind the products of Robert Bosch GmbH and Continental in key indicators such as detection distance and angular resolution; in the medical field, optical sensors used for non-invasive blood glucose monitoring are still in the laboratory stage in China, and the market is completely monopolized by Abbott and Dexcom in the United States.
Behind the insufficient supply of high-end products is the gap of domestic enterprises in core indicators such as "reliability, consistency, and long life." According to public industrial analysis by Sensor Expert Network, the MTBF (Mean Time Between Failures) of foreign high-end industrial sensors can generally reach over 100,000 hours, while the MTBF of mainstream domestic high-end industrial sensors is about 50,000 hours, showing an obvious gap in long-term operating condition reliability [9]; automotive-grade sensors need to pass wide-temperature tests from -40℃ to 125℃, and the performance degradation rate of products from some domestic enterprises exceeds 20% under extreme temperatures, making it difficult to meet automotive safety standards. This status of "usable but not easy to use" has led to the long-term monopoly of the high-end market by European, American, and Japanese enterprises, and the road to import substitution remains long.
Issue 4: Weak Testing and Verification Capability and Difficulty in Guaranteeing Reliability The performance indicators of sensors need to be guaranteed through strict testing and verification, and the testing of high-end sensors involves multi-dimensional environmental simulations such as mechanics, temperature, humidity, and electromagnetic compatibility, which places extremely high demands on testing equipment and technical capabilities. Currently, there are obvious shortcomings in domestic sensor testing and verification capabilities: from the equipment side, high-end testing equipment such as MEMS dynamic characteristic testers and automotive electronic reliability testing systems are over 90% dependent on imports, and they are expensive. A single unit can cost tens of thousands, hundreds of thousands or even tens of millions of RMB, which is difficult for small and medium-sized enterprises to afford; from the system side, China has not yet formed a full-chain testing standard covering "materials-chips-modules-complete machines." Most enterprises can only refer to foreign standards for self-testing, and the authority and recognition of the testing results are insufficient.
The weakness in testing and verification capabilities directly affects product reliability and market trust. For example, a new product developed by a MEMS pressure sensor enterprise, due to the lack of professional temperature cycling testing equipment, was only subjected to simple high and low-temperature tests before being launched to the market. As a result, a large number of failure problems caused by temperature drift occurred in customer applications, with a failure rate as high as 8%. This not only caused economic losses of millions of RMB but also damaged the reputation of domestic sensors. In addition, insufficient testing capabilities also lead to prolonged product iteration cycles. The testing cycle for a sensor from R&D to mass production in foreign enterprises is about 6 months, while domestic enterprises need an average of 10-12 months, missing market opportunities.
Suboptimal Industrial Ecosystem: Endogenous Momentum Needs Continuous Activation The healthy development of the sensor industry cannot be separated from the support of a sound industrial ecosystem. However, the domestic sensor industry still has many shortcomings in areas such as the standard system, production-application docking, talent cultivation, financing support, and intellectual property protection, leading to prominent problems of "difficult innovation, difficult transformation, and difficult profitability," making it difficult to effectively release endogenous momentum.
Issue 5: Non-unified Standard System and Hindered Industry Collaboration There are numerous categories of sensors, involving thousands of varieties and over 20,000 forms of physical, chemical, and biological quantities. Different application scenarios have vastly different requirements for the technical indicators, interface protocols, and data formats of sensors. Currently, the domestic sensor standard system has problems such as "multi-headed formulation, cross-repetition, and missing lags": at the national standard level, about 180 sensor standards have been released, but they only cover about 85% of mainstream sensor categories [10]. There are only 9 current standards for smart sensors, and standards for emerging fields such as flexible sensors are basically blank; at the industry standard level, different industries such as machinery, automotive, and electronics formulate their own standards, leading to inconsistent testing methods and certification requirements for the same type of sensor in different industries. Enterprises need to undergo repeated certifications, significantly increasing production costs by 20%-30%.
Non-unified standards directly restrict industry collaboration and large-scale application. According to the "2025-2030 China Smart Sensor Market In-Depth Panoramic Research and Investment Prospect Analysis Report" released by Zhongyan Puhua Industry Research Institute, there are currently up to 12 domestic sensor interface protocols, leading to a 20%-30% increase in system integration costs [11]. For example, there are more than ten protocols such as LoRa, NB-IoT, and ZigBee for sensor data interfaces in the IoT field. When building IoT platforms, application enterprises need to develop dedicated gateways for sensors with different interfaces, increasing system integration costs by over 50%; in the industrial internet field, the data formats of smart sensors from different brands are not unified, making it impossible for factories to achieve interconnection and interoperability between equipment. The "data island" problem is prominent. An automotive factory reported that due to incompatible sensor data, the equipment utilization rate of the production line decreased by 15%. In addition, the lack of standards also brings difficulties to market regulation. Some enterprises use inferior materials to produce sensors to reduce costs and pass off fake for genuine. According to data from the State Administration for Market Regulation, the unqualified rate of household combustible gas detectors sampled in 2024 reached 8.9% [12], disrupting market order.
Issue 6: Unsmooth Production-Application Connection and Prominent "Supply-Demand Mismatch" The technological R&D of sensors must be deeply bound to downstream application scenarios to achieve "promoting research through application and promoting production through research." However, there is a severe "supply-demand mismatch" in the domestic sensor industry: from the supply side, most enterprises lack an in-depth understanding of downstream scenarios and blindly follow the trend in R&D directions. For example, in recent years, LiDAR has become a hotspot, and hundreds of LiDAR enterprises have emerged in China. However, the industry generally presents the characteristics of converging technological routes and highly homogenized core performance parameters. Products are generally concentrated in similar indicators such as detection distance, number of lines, and resolution, failing to carry out differentiated designs for different scenarios such as autonomous driving and industrial inspection; from the demand side, downstream application enterprises have a "trust barrier" to domestic sensors, especially in fields with high safety requirements such as automotive and aerospace. Even if the performance of domestic sensors is close to imported products, enterprises still tend to choose foreign brands, leading to "high trial-and-error costs and long verification cycles" for domestic sensors. Industry data shows that the AEC-Q100 certification and complete vehicle verification cycle for domestic automotive-grade MEMS sensors is generally as long as 1-2 years [13]; while imported products, having already obtained international mutual recognition such as FAA/EASA and AEC-Q100, can shorten the domestic verification cycle to about 6 months.
This "disconnection between production and application" makes it difficult to transform scientific research achievements. According to data from the Ministry of Science and Technology, the laboratory technology transformation rate in the domestic sensor field is less than 20% [14], while the advanced level abroad exceeds 36% [15]. For example, a new type of gas sensor developed by a university performed 30% better than imported products in laboratory tests, but due to the lack of cooperation with home appliance enterprises, it was unable to optimize power consumption and costs according to actual application scenarios, and ultimately could only stay at the sample stage; an industrial robot enterprise needed a high-precision force sensor. Although domestic enterprises could produce it, they could not provide certification reports meeting international standards, and the enterprise had to spend 3 times the price to import German products.
Issue 7: Shortage of Compound Talents and Insufficient Industry-University-Research Collaboration The development of the sensor industry requires a large number of compound talents who "understand technology, understand processes, and understand the market," but the current domestic talent supply is severely insufficient. From the cultivation side, sensor-related majors in universities are mostly affiliated with electronics and mechanical colleges. The curriculum focuses on theoretical teaching and lacks practical links combined with industrial practice. According to industry research and enterprise practice, sensor major graduates need an average of 1-2 years to work independently after entering enterprises; from the introduction side, the domestic supply gap for high-end talents such as MEMS process experts and sensor algorithm engineers exceeds 100,000 people [16]. Enterprises have to introduce them from abroad at high salaries, with annual salaries several times higher than similar domestic personnel, further increasing costs.
Behind the talent shortage is the imperfect industry-university-research collaborative innovation mechanism. Currently, the cooperation between universities, research institutes, and enterprises mostly stays at shallow levels such as "project cooperation" and "technology transfer," lacking long-term and stable interest-sharing mechanisms. For example, university research teams focus more on paper publication and patent applications, while enterprises focus on product performance and market benefits. The inconsistent goals make it difficult to deepen cooperation; after some scientific research achievements are transferred, enterprises find it difficult to promote industrial implementation due to the lack of subsequent technical support; according to widespread industry reflection, the industrialization success rate of scientific research achievements in the sensor field is low, and the problem of transformation failure is prominent, with a failure rate exceeding 40%. In addition, the opening rate of instruments and equipment in research institutes to enterprises is insufficient, less than 30% [17], making it difficult for enterprises to carry out pilot tests and verification, restricting the efficiency of technology transformation.
Issue 8: Single Financing Channels and High Financial Pressure for the Development of SMEs Sensor R&D has the characteristics of "high investment, long cycle, and slow return": it takes 3-5 years from R&D to mass production for a mid-to-high-end sensor, with R&D investment often reaching tens of millions of RMB, while market returns often begin to appear 2-3 years after production. Currently, the financing channels for domestic sensor enterprises are still mainly bank loans and equity financing, with prominent problems of "difficult and expensive financing."
From the perspective of bank loans, sensor enterprises are mostly asset-light enterprises, with over 90% of enterprises having fixed assets accounting for less than 30%; Most domestic manufacturers follow the Fabless model; IDM players are few and small-scale, so enterprises generally hold a low share of fixed assets, lacking fixed assets that can be mortgaged, and the uncertainty of R&D projects is high, so banks have a low willingness to lend. According to data from the CBIRC, the approval rate of bank loans for sensor enterprises is only 45%, 20 percentage points lower than the average level of the manufacturing industry. As typical technology-based SMEs, their difficulty in getting bank loans approved is even greater, and the loan level is significantly lower than that of traditional manufacturing enterprises; from the perspective of equity financing, capital is more inclined to invest in projects that can see results in the short term, and is relatively cautious about investment in fields with long cycles and high R&D investments like sensors. In 2025, equity financing for China’s sensor sector accounted for merely 8% of total semiconductor equity financing, compared with over 20% in mature overseas markets. SMEs are even the "hardest hit areas" for financing difficulties. The head of a MEMS sensor start-up said that in the 3 years since the enterprise was established, it has only obtained one round of angel round financing, with an amount of less than 10 million RMB. Subsequent R&D funds mainly rely on the founder's self-raising, leading to the forced suspension of some core technology R&D. Single financing channels make it difficult for domestic enterprises to increase R&D investment. In 2025, the average R&D investment ratio of domestic sensor enterprises was only 8%, while leading foreign enterprises generally exceeded 15% [16].
Issue 9: Weak Intellectual Property Protection and Suppressed Innovation Enthusiasm Intellectual property is the core competitiveness of sensor enterprises, but there are still many problems in the intellectual property protection of the domestic sensor industry: from the perspective of infringement, patent infringement in the mid-to-low-end sensor field is prominent, and counterfeit products are relatively common. According to research and statistics by the CNIPA, in 2025, the number of patent infringement disputes in the industry increased significantly, with a year-on-year increase of 35%. Some enterprises copy foreign product designs through "reverse engineering" and even directly steal the patent technology of domestic enterprises to seize the market at low prices; from the perspective of rights protection costs, patent litigation has a long cycle and high costs. Enterprise rights protection often takes 2-3 years and costs millions of RMB, while the infringement compensation amount is generally low, with an average compensation of less than 5 million RMB, which is difficult to make up for the losses of enterprises.
Weak intellectual property protection directly suppresses the innovation enthusiasm of enterprises. Some enterprises, fearing that their technology will be copied, are unwilling to invest in core technology R&D and instead choose "low-level redundant construction," leading to intensified homogeneous competition and price wars in the industry. In 2025, this phenomenon triggered price wars in the mid-to-low-end sensor market, causing the industry's average profit margin to drop to below 5%, and the profit margin was greatly compressed; some possessing innovation capabilities of SMEs, unable to bear the high rights protection costs, had to be forced to shrink business and give up some market shares. For example, a new type of humidity sensor developed by an enterprise obtained multiple patents. Less than half a year after it was launched, multiple counterfeit products appeared on the market, with prices only 60% of the genuine products. Although the enterprise filed a lawsuit, due to the difficulty in collecting infringement evidence, no judgment has been made so far, and its market share has dropped from 30% to 15%.
Pressure from the External Environment: Intensified Globalization Risks and Challenges In the context of globalization, the supply chain of the sensor industry is highly dependent on international collaboration. However, in recent years, factors such as geopolitical conflicts and the rise of trade protectionism have led to a decline in the stability of international supply chains, and the external risks faced by domestic enterprises continue to intensify.
Issue 10: Unstable International Supply Chain and Hindered Globalization Layout Domestic sensor enterprises are still highly dependent on imports in fields such as key raw materials, core equipment, and high-end chips: in terms of raw materials, high-purity single-crystal silicon wafers and piezoelectric ceramic materials required for manufacturing MEMS sensors are over 80% from enterprises such as Shin-Etsu in Japan and MEMC in the United States [18][19]; in terms of core equipment, MEMS lithography machines mainly rely on overseas manufacturers such as ASML in the Netherlands and Nikon in Japan, with the overall domestic self-sufficiency rate of less than 10% [20]; in terms of high-end chips, the self-sufficiency rate of dedicated AI chips required for smart sensors is less than 10% [21], mainly imported from NVIDIA and Texas Instruments in the United States.
The intensification of geopolitical conflicts has led to an increased risk of "supply interruption" in the international supply chain. For example, in 2024, a certain country implemented new export controls on semiconductor equipment to China, making it difficult for some domestic MEMS enterprises to obtain the latest key manufacturing equipment such as high-end etching, and the high-end capacity expansion plan was forced to be postponed; in April 2025, Japan officially restricted export controls on semiconductor materials and related items [22]. Affected by this, the safety stock cycle of key imported raw materials for domestic sensor enterprises generally dropped to 3-6 months [23], and production faced the risk of interruption. In addition, the rise of trade protectionism has also increased the difficulty of globalization layout for domestic enterprises. Some countries restrict domestic sensors from entering their markets on the grounds of "national security." According to official data from the China Electronic Components Industry Association (CECA), from January to October 2025, China's export volume of sensitive components and sensors increased by 23.14% year-on-year, while exports to the European and American markets decreased by more than 15%. For example, on January 1, 2025, the United States officially implemented the new "Section 301" tariff rules, raising the most-favored-nation additional tariff rate for core industrial robot components such as collision sensors, torque feedback modules, and safety control units from 15% to 25%. According to statistics from the General Administration of Customs of China, in the first quarter of 2025, China's exports of industrial robot components to the United States decreased by 11% year-on-year, among which the export decline of collision sensors reached 14.7% [24].
Breakthrough Paths: Multi-dimensional Collaboration to Promote High-Quality Industrial Development In response to the ten major problems in the above three categories, the breakthrough of China's sensor industry needs to make coordinated efforts from three dimensions: technology, ecology, and external environment. Through "making up for shortcomings, strengthening advantages, and preventing risks," it promotes the transformation of the industry from "scale expansion" to "quality improvement," truly achieving autonomous and controllable high-quality development.
Aiming at "Weak Technological Foundation": Strengthen Underlying Innovation and Intelligent Upgrading First, increase the research on core mechanism technologies. It is recommended to establish a national special fund for basic sensor research, with an annual investment of no less than 2 billion RMB, focusing on supporting original research "from 0 to 1" such as new sensing mechanisms, new material applications, and interdisciplinary integration. The goal is to increase the proportion of independently developed new MEMS structural designs to 20% by 2030, and raise the transformation rate of original mechanism technologies for flexible sensors to 15%.
Second, accelerate the breakthrough of smart sensor technologies. Promote the R&D of single-chip integration of "sensor + AI chip," and provide subsidies of up to 10 million RMB per product for enterprises that achieve integration; build an "open-source platform for smart sensor algorithms," and collaborate with universities to develop 100+ industry-specific algorithm models. The goal is to narrow the gap in data processing accuracy between domestic smart sensors and foreign products to within 5% by 2028, and increase the response speed by15%.
Third, enhance the supply capacity of high-end products. Implement the "Special Action for High-End Sensors," and reward products that pass AEC-Q100 automotive certification and ISO 13485 medical certification with 8%-12% of their sales. The goal is to increase the localization rate of industrial-grade MEMS sensors to 30% by 2030, and break through 50% for the localization rate of automotive-grade millimeter-wave radar.
Fourth, improve the testing and verification system. Build 2-3 national-level sensor testing and verification centers in the Yangtze River Delta and the Pearl River Delta, equipped with over 100 sets of high-end testing equipment, providing full-chain testing services from materials to complete machines. The goal is to achieve an 80% coverage rate of high-end sensor testing standards by 2027, and shorten the product iteration cycle of enterprises to within 8 months.
Aiming at "Suboptimal Industrial Ecosystem": Build a Collaborative and Efficient Development Environment First, unify the standard system. The MIIT should take the lead in establishing the National Sensor Standardization Technical Committee, and formulate more than 50 national standards for emerging fields such as smart sensors and flexible sensors within 3 years, promoting the interoperability of industry standards in automotive, industrial, and other sectors. The goal is to achieve a 90% standard coverage rate for mainstream sensor categories by 2028, and reduce the repeated certification costs for enterprises by 40%.
Second, deepen production-application docking. Establish a "sensor production-application docking platform," release 100+ key downstream application demands annually, and organize more than 50 "enterprise-user" matchmaking meetings; carry out "domestic sensor demonstration applications" in the automotive and aerospace fields, and provide subsidies of 10%-15% of the procurement amount for application enterprises. The goal is to shorten the trial-and-error cycle for domestic sensors to within 1 year by 2027, and increase the laboratory technology transformation rate to 40%.
Third, strengthen talent cultivation and industry-university-research collaboration. Support 20 universities in establishing undergraduate majors in "Sensor Engineering," build 50 school-enterprise joint internship bases, and cultivate more than 10,000 compound talents annually; promote the formation of 20 innovation consortia by enterprises, universities, and research institutes, binding cooperation through "technology equity + profit sharing." The goal is to reduce the failure rate of industrialization of scientific research achievements to below 20% by 2028.
Fourth, broaden financing channels. Establish a national-level sensor industry fund with a total scale of 50 billion RMB, focusing on investing in the technological R&D of SMEs; encourage banks to develop "intellectual property pledge loans," and provide interest rate discounts of 30 basis points below the LPR for sensor enterprises. The goal is to increase the approval rate of bank loans for sensor enterprises to 70% by 2027, and raise the average R&D investment ratio of the industry to 12%.
Fifth, strengthen intellectual property protection. Establish a rapid rights protection center for intellectual property in the sensor industry, shortening the handling cycle of patent infringement disputes to within 3 months; increase the punishment for infringement, imposing fines of over 5 million RMB for malicious infringement. The goal is to reduce the number of patent infringement dispute cases in the industry by 20% year-on-year by 2028.
Aiming at "Pressure from the External Environment": Enhance the Autonomous and Controllable Capability of the Supply Chain First, build an autonomous supply chain system. Support domestic enterprises in R&D of key materials and equipment such as high-purity single-crystal silicon wafers and deep silicon etching machines, and provide one-time rewards of 20%-30% for products that achieve import substitution. The goal is to increase the self-sufficiency rate of key raw materials for sensors to 50% by 2030, and break through 30% for the self-sufficiency rate of core equipment.
Second, diversify international cooperation. Encourage enterprises to acquire advanced technologies through technological cooperation and overseas mergers and acquisitions, and provide a 5% subsidy for projects with M&A amounts exceeding 100 million USD; strengthen cooperation with countries along the "Belt and Road," and jointly build more than 5 sensor production bases. The goal is to increase the proportion of sensor exports to emerging markets to 40% by 2030.
Conclusion The high-quality development of the sensor industry is a key support for China to achieve the strategies of "industrial digitalization," "digital industrialization," "manufacturing powerhouse," and "cyber powerhouse." Facing the challenges of "weak technological foundation, suboptimal industrial ecosystem, and pressure from the external environment," the whole industry needs to lay out underlying technologies with a more long-term vision, improve the industrial ecosystem with a more coordinated mechanism, and cope with global competition with a more open attitude. Only in this way can China's sensor industry truly break through the "low-end lock-in" and take the initiative in the global competition in the intelligent era.
References: [1] China Sensor and IoT Industry Alliance (CCID Consulting "China Sensor Market Research Report" https://news.sohu.com/a/1043559720_378413) [2] Guo Yuansheng. Challenges and Suggestions Facing the Development of China's Sensor Industry [J]. Economic Reference, 2024 (27): Total Issue 2645. [3] Guo Yuansheng. How Can Sensor Technology Empower Intelligent Manufacturing? [N]. China Electronics News, 2026-01-16. [4] Yole Group "Greater China MEMS Industry 2025" (Greater China MEMS Industry - 2025 Edition) "Greater China MEMS Industry - 2025 Edition" - MEMS Sensor Market Report - WeiMi: Professional MEMS Market Research Media - Hosted by MEMS Consulting [5] Guo Yuansheng. How Can Sensors Empower the Petrochemical Industry? [N/OL]. China Electronics News, 2026-04-23. http://news.qq.com/rain/a/20260423A04XZ600 [6] Peng Zhao. National Standard for Intelligent Agent Interconnection Released: Why May It Still Fail to Connect to the Physical World Even with Unified Interfaces? [EB/OL]. (2026-06-30)[2026-08-07]. https://36kr.com/p/3875446086840577. [7] Boxun Zhishang. 2025 China Laser Displacement Sensor Market Share Report and Industry Competition Pattern Analysis Report [R/OL]. (2025)[2026-07-07].https://www.docin.com/touch_new/preview_new.do?id=5018773624. [8] China Commerce Industry Research Institute. 2026 China Intelligent Driving Industry Chain Map and Investment Layout Analysis [EB/OL]. 2026-06-26 [2026-07-07].https://xueqiu.com/2294082574/397019001. [9] Sensor Expert Network. Technical Bottlenecks of Pressure Transmitter Sensors Restrict the Upgrading of Industrial Intelligence [EB/OL]. 2026-05-04 [2026-07-07]. https://www.sensorexpert.com.cn/article/453893.html. [10] Standardization Administration of China. National Standard Information Public Service Platform [EB/OL]. 2026-07-08[2026-07-08]. https://std.samr.gov.cn/. [11] Zhongyan Puhua Industry Research Institute. 2025-2030 China Smart Sensor Market In-Depth Panoramic Research and Investment Prospect Analysis Report [R]. Shenzhen: Zhongyan Puhua, 2025. [12] State Administration for Market Regulation. Notice on the 2024 National Supervision and Sampling of Product Quality of 32 Kinds of Products Including Household Electric Washing Machines [EB/OL]. 2025-01-02 [2026-07-25]. https://www.samr.gov.cn/zw/zfxxgk/fdzdgknr/zljds/art/2025/art_f77a794ba80344cfb1018b47603b5903.html. [13] Tianxia Factory Industry Research Institute. China MEMS Sensors 2026 - Dual-Wheel Domestic Breakthrough from Consumer Electronics Acoustics to Industrial Inertia [R]. Beijing: Tianxia Factory Industry Research Institute, 2026. https://faxiangongchang.com/reports/china-mems-sensors-2026 [14] China National Intellectual Property Administration. Patent Transformation and Application: Speeding Up Without Dismounting [EB/OL]. (2026-03-27)[2026-08-07]. https://www.cnipa.gov.cn/art/2026/3/27/art_55_205535.html. [15] European Patent Office. New EPO study: European patents preferred tool for the commercialisation of inventions developed by Europe's universities and public research organisations[EB/OL]. (2020-11-24)[2026-08-07]. https://www.epo.org/en/news-events/news/new-epo-study-european-patents-preferred-tool-commercialisation-inventions. [16] Guo Yuansheng. From Catching Up to Leading: The "14th Five-Year Plan" Answer Sheet and "15th Five-Year Plan" Blueprint of China's Sensor Industry [J]. China Economic Report, 2025 (22). [17] Ministry of Science and Technology of the People's Republic of China, Ministry of Finance of the People's Republic of China. Notice of the General Office of the Ministry of Science and Technology and the General Office of the Ministry of Finance on the Publication of the Evaluation and Assessment Results of the Open Sharing of Major Scientific Research Infrastructure and Large Scientific Research Instruments in Central Universities and Research Institutes in 2025 [EB/OL]. 2025-10-10 [2026-07-29]. https://www.most.gov.cn/xxgk/xinxifenlei/fdzdgknr/qtwj/qtwj2025/202510/t20251010_194850.html. [18] Sensor Expert Network. 11 Major Choke Point Technologies Most Urgently Needing Breakthroughs in China's Sensors [EB/OL]. 2024-05-11. https://www.sensorexpert.com.cn/article/370103.html. [19] Sina News. As a Key Link in the AI Computing Power Industry Chain, What Challenges Will MEMS Technology Face in Localized Manufacturing in China? [EB/OL]. 2025-12-03.https://news.sina.cn/bignews/insight/1970-01-01/detail-infznqyq9655164.d.html?vt=4. [20] Sina Finance "ASML Lithography Machine Chokes the Throat, US-Japan-Netherlands Monopolize 90% of Equipment, China's Etching Machine Has Entered the 28nm Battlefield", 2026-07-10, https://finance.sina.com.cn/roll/2026-07-10/doc-inihifxw7063259.shtml [21] Zimu Research. Overview of the Progress of Autonomous Supply Chain in China's Semiconductor Industry (December 2025) [R/OL]. (2025-12-31)[2026-08-03]. https://www.xinsanbanbao.com/20251231/news/n98415.html. [22] Caixin. Japan Adds Export Controls on Semiconductor Equipment, China Firmly Opposes [EB/OL]. (2025-04-03)[2026-08-05]. https://www.caixin.com/2025-04-03/102305871.html. [23] Hongliu Strategy. Export Controls on Key Materials to Japan and Benefiting A-Share Companies - Based on the Production Reduction Transmission Timetable and Business Purity Screening of Hongliu Strategy (Complete Version) [EB/OL]. (2026-06-07)[2026-08-05]. https://xueqiu.com/6513538491/393326066 [24] SHC Sensor. The U.S. imposes an additional 25% tariff on Chinese industrial robot components[EB/OL]. (2026-08-03)[2026-08-05]. https://www.shcsensor.com/news/xingyexinwen/meiduihuagongyejiqirenlingbujianjiazheng25_guanshui.html.
Author | Guo Yuansheng, Deputy Director of the Science and Technology Committee of the Central Committee of the Jiusan Society, Executive Vice Chairman of the China Sensor and IoT Industry Alliance Editor | Yang Pengyue Art Editor | Ma Liya