Editor's Note: Sensors, as the "nerve endings of the information age," have penetrated every critical field of the socio-economic landscape. Since October 2025, China Electronics News has invited Guo Yuansheng, Deputy Director of the Science and Technology Committee of the Jiusan Society Central Committee and Executive Vice Chairman of the China Sensor and IoT Industry Alliance, to launch the column "Guo Yuansheng Explains Sensors." Focusing on eight major fields and scenarios: electric power, major equipment, intelligent manufacturing, smart agriculture, smart healthcare and big health, smart home appliances and consumer electronics, urban security, and low-altitude economy, it has sequentially published articles such as "Stationed at the Power Generation Side: Sensors Become the Cornerstone of Stable Operation of New Power Systems" and "Energy Storage Sensors Clarify Three Core Development Directions," which have attracted widespread attention and received rave reviews from readers. The current article focuses on intelligent sensors in the highway transportation sector, elaborating on their application prospects, industrial status, and future recommendations to build industry consensus and promote industrial development.
China is currently at a critical stage of transitioning highway transportation from "digitalization" to "smart transformation." Given that the national highway network mileage has exceeded 170,000 kilometers, and the proportion of key facilities such as bridges and tunnels exceeds 40%, the traditional manual monitoring model can no longer cope with multiple challenges such as structural aging, extreme weather, and traffic congestion. There is an urgent need for a comprehensive monitoring system centered on intelligent sensors to support proactive warning, precise maintenance, and vehicle-road coordination in highway transportation.
Focusing on core scenarios such as bridges, tunnels, and road surfaces, it is crucial to systematically analyze the application solutions of sensors in structural safety, environmental warning, and traffic efficiency improvement. Combined with technologies like BeiDou positioning and C-V2X communication, this provides actionable perception-layer solutions for the intelligent upgrade of highways.
I. Bridge Health Monitoring: Sensor Applications for Structural Safety
Core Monitoring Requirements
As critical nodes in highway transportation, the structural safety of bridges directly relates to the lives and property of passing vehicles and personnel. Early-built highway bridges in China have entered the "mid-to-late stage of service," with hidden dangers such as main girder deformation, bearing settlement, stress concentration, and abnormal vibration becoming increasingly prominent. According to data from the Ministry of Transport, in 2024, among registered highway bridges nationwide, Class III and Class IV bridges requiring regular monitoring accounted for about 12%. Among them, long-span box girder bridges and cable-stayed bridges, due to long-term bearing of vehicle loads, wind loads, and temperature changes, face particularly prominent issues such as excessive main girder deflection deformation and cable stress relaxation. Furthermore, extreme weather (such as heavy rain causing scouring of bridge pier foundations and low temperatures causing concrete cracking) further exacerbates the risk of bridge structural damage. Traditional manual inspections (such as periodic total station measurements and visual crack observation) have drawbacks like long cycles, low precision, and the inability to provide real-time warnings. Therefore, deploying intelligent sensors is necessary to achieve long-term real-time monitoring of bridge structural parameters, warn of structural damage 1-3 months in advance, provide precise data support for maintenance, and avoid major accidents such as bridge overload collapse.
Key Sensors and Technical Parameters
Addressing the core requirements of bridge health monitoring, different monitoring indicators require matching specific types of sensors. Their technical parameters must meet the environmental requirements for long-term outdoor service of highway bridges (e.g., waterproof rating ≥ IP68, temperature resistance range -40℃~85℃, fatigue life ≥ 1 million cycles). Details are as follows:
1. Displacement Sensors: Mainly used to monitor main girder deflection (vertical deformation) and bearing settlement (settlement at the connection between piers and bearings). They require high precision and low drift. Fiber Bragg Grating (FBG) displacement sensors are recommended, with precision ≤ 0.1mm (meeting the millimeter-level requirements for long-span bridge deflection monitoring), a range of 0-300mm (covering the normal deflection variation range of most highway bridges). Adopting an all-fiber structure with no electromagnetic interference, they are suitable for long-span box girder bridges in strong electromagnetic environments such as crossing rivers and valleys. Compared to traditional resistive displacement sensors, their fatigue life is increased to over 2 million cycles, avoiding increased maintenance costs caused by frequent sensor replacement.
2. Strain Sensors: Used to monitor stress concentration in key structures like main girders and piers, promptly detecting issues like concrete cracking and steel structure stress overload. Resistive strain sensors are selected, with precision ≤ 1με (με is microstrain, where 1με means 1 micrometer deformation per meter of material, capable of precisely capturing tiny concrete strains), a range of ±2000με (covering the strain range under normal service conditions of highway bridges; warning triggered when strain exceeds 1500με). They feature temperature compensation (compensation range -40℃~85℃, error ≤ ±5με), eliminating the impact of extreme temperatures on measurement precision, suitable for bridge monitoring in severe cold regions in the north or high-temperature and high-humidity environments in the south.
3. Vibration Sensors: Used to monitor the modal frequency of bridges (e.g., natural frequency, damping ratio) and identify bridge resonance risks. When the vehicle load frequency approaches the bridge's natural frequency, resonance can easily occur, causing structural damage. Piezoelectric vibration sensors are adopted, with a frequency range of 0.5-50Hz (covering the common vibration frequency range of highway bridges), sensitivity of 0.01g (g is gravitational acceleration, capable of capturing tiny vibration signals), and response time ≤ 10ms. When a change in the bridge's natural frequency exceeding 5% is detected (indicating decreased structural stiffness, possibly due to cracks or bearing aging), an alarm is automatically triggered. Meanwhile, the sensors support multi-axis measurement (X, Y, Z three axes), comprehensively capturing the bridge's vibration in lateral, longitudinal, and vertical directions under vehicle loads, suitable for long-span bridges like cable-stayed and suspension bridges that are easily affected by wind loads.
4. Tilt Sensors: Used to monitor the tilt angle of bridge piers, preventing the risk of tilt collapse caused by foundation scouring or uneven settlement. MEMS tilt sensors are selected, with precision ≤ 0.01° and a range of ±3° (the normal tilt angle of highway piers should be ≤ 0.5°, with emergency warning triggered if it exceeds 1°). With an IP68 waterproof rating, they can be directly installed on the side of piers. They feature low power consumption (operating current ≤ 10mA, supporting solar power supply, suitable for pier monitoring in remote areas without mains power). Their built-in filtering algorithms can eliminate vibration interference caused by vehicle passage, ensuring the accuracy of tilt angle measurements.
Typical Applications
A cross-river expressway bridge (main span 380-meter box girder bridge), to cope with structural safety risks brought by long-term river winds, vehicle loads, and flood season water level changes, has deployed a 200-unit multi-sensor networking system. Specific applications are as follows:
1. One FBG displacement sensor is installed every 20 meters at the bottom of the main girder to monitor main girder deflection changes in real time. The deflection warning threshold is set to 20mm (the design allowable deflection for this bridge is 25mm). In August 2025, the sensors detected that the deflection at the center of the main span reached 18mm, an increase of 3mm from the previous month. The system automatically pushed warning information to the highway maintenance department. Upon investigation, it was found that the bearing at this location had decreased support stiffness due to rubber aging. The maintenance team completed the bearing replacement 2 months in advance, avoiding the risk of main girder cracking caused by further deflection increase.
2. One MEMS tilt sensor is installed at the top and bottom of each pier to monitor the pier tilt angle. During the 2025 flood season, the foundation of Pier 3 on this bridge was scoured by river water, causing the tilt angle to increase from 0.2° to 0.6°. The sensor triggered a Level 2 warning (Level 1 warning is 0.5°, Level 2 warning is 0.8°). The maintenance department immediately riprap-reinforced the pier foundation to prevent the tilt angle from continuing to increase, ensuring the bridge's traffic safety during the flood season.
3. A total of 50 piezoelectric vibration sensors are deployed across the bridge to monitor the bridge's modal frequency in real time. When the natural frequency of the bridge was detected to drop from 3.2Hz to 3.0Hz during a certain period (a change rate of 6.25%, exceeding the 5% warning threshold), the system, combined with concurrent traffic flow data, found that a large number of heavy trucks were passing concentratedly during this period (single vehicle load exceeding 50 tons). It immediately linked with the expressway entrance weighing system to restrict heavy trucks from entering the bridge, while reminding the maintenance department to conduct stress testing on the main girder to avoid structural damage caused by resonance.
Technical Upgrade Directions
During the "15th Five-Year Plan" period, bridge health monitoring sensors will upgrade towards "high precision, self-diagnosis, and AI integration" to further improve monitoring efficiency and warning accuracy:
1. Displacement Sensor Upgrade: FBG displacement sensors will integrate dual-parameter compensation algorithms for temperature and humidity, improving precision from 0.1mm to 0.05mm. This can distinguish between thermal expansion and contraction of the main girder caused by temperature changes and structural deformation, avoiding false alarms. Meanwhile, the sensors will support self-diagnostic functions; when a fiber optic link breaks or experiences loss, they will automatically push fault information to the monitoring platform, shortening maintenance response time (from the original 24 hours to within 2 hours).
2. Vibration Sensor Upgrade: Piezoelectric vibration sensors will integrate the BeiDou-3 high-precision positioning module (positioning precision ≤ 0.5m), achieving precise matching of vibration data with the bridge's spatial location, capable of pinpointing vibration anomalies to specific girder segments or piers. Combined with AI algorithms, by analyzing historical vibration data and bridge damage cases, a "vibration frequency-structural damage" correlation model can be established to predict the development trend of bridge damage (e.g., predicting that main girder cracks will expand to a dangerous range within 3 months), formulating maintenance plans in advance, and realizing the shift from "passive warning" to "active prediction."
3. Sensor Networking Upgrade: Adopting 5G-A (5th Generation Mobile Communication Advanced) technology to replace traditional LoRaWAN transmission, achieving millisecond-level uploading of sensor data (transmission delay ≤ 10ms) and supporting synchronous data collection across all bridge sensors, avoiding multi-parameter analysis errors caused by transmission delays. Meanwhile, the high bandwidth characteristics of 5G-A can support sensors equipped with micro-cameras, realizing synchronous "vibration + image" monitoring. When vibration sensors issue a warning, the cameras automatically capture images of the abnormal parts, assisting maintenance personnel in quickly judging the damage type and improving decision-making efficiency.
II. Tunnel Safety Monitoring: Sensor Applications for Environment and Structure
Core Monitoring Requirements
As a special enclosed scenario in highway transportation, tunnel safety monitoring must balance two dimensions: structural health and traffic environment. Structurally, long-term exposure to surrounding rock pressure, groundwater infiltration, and vehicle vibration makes it prone to issues like lining cracking, rock bolt loosening, and invert settlement. Statistics show that in 2025, about 15.3% of national highway tunnels had varying degrees of lining cracks, with the crack incidence rate in mountainous tunnels reaching as high as 22.5% due to complex geological conditions. Environmentally, poor ventilation and dim lighting in tunnels can easily lead to excessive carbon monoxide concentration, reduced visibility (e.g., smoke, dust), and slippery road surfaces (e.g., water leakage, icing). These not only affect driving safety but may also trigger major accidents like fires and personnel asphyxiation. Therefore, intelligent sensors are needed to achieve "structure + environment" collaborative monitoring, providing real-time warnings for risks such as lining damage, fire hazards, and abnormal air quality, ensuring 24-hour safe passage through tunnels.
Key Sensors and Technical Parameters
Addressing the dual monitoring requirements, two types of sensors for structural and environmental monitoring must be deployed. Their technical parameters must meet the service requirements of damp, dusty, and electromagnetically complex tunnel environments. Details are as follows:
1. Structural Monitoring Sensors
(1) Crack Sensors: Used to monitor the width changes of tunnel lining cracks and promptly detect structural damage. Vibrating wire crack sensors are recommended, with a measurement range of 0-5mm (covering common crack widths in highway tunnel linings), precision ≤ 0.01mm, and temperature compensation function (error ≤ 0.02mm within the -20℃~60℃ range), eliminating the impact of temperature fluctuations in the tunnel on measurement results. The sensors adopt a stainless steel shell with an IP68 protection rating, can be directly embedded in lining cracks, withstand groundwater infiltration and dust erosion, and are suitable for mountainous tunnels with water seepage.
(2) Rock Bolt Stress Sensors: Used to monitor the stress state of surrounding rock support rock bolts in tunnels, preventing surrounding rock instability caused by rock bolt loosening. FBG rock bolt stress sensors are selected, with a measurement range of 0-200MPa (the design stress of highway tunnel rock bolts is generally ≤ 150MPa), precision ≤ 1MPa, and resolution of 0.5MPa. They can capture sudden changes in rock bolt stress in real time (e.g., warning triggered if stress suddenly increases by more than 10MPa). Adopting all-fiber transmission with strong anti-electromagnetic interference capabilities, they are suitable for electromagnetic environments in tunnels with high-voltage cables and communication equipment, avoiding measurement errors caused by signal interference.
(3) Settlement Sensors: Used to monitor the settlement of the tunnel invert (bottom structure), preventing road surface cracking and vehicle bumping caused by invert settlement. Laser settlement sensors are adopted, with a measurement range of 0-100mm, precision ≤ 0.1mm, and response time ≤ 1s, enabling continuous monitoring of invert settlement changes. The sensors feature a self-calibration function, automatically calibrating once every 24 hours to avoid zero drift caused by long-term use, suitable for segmented settlement monitoring of long-distance tunnels (e.g., highway tunnels exceeding 3km in length).
2. Environmental Monitoring Sensors
(1) Carbon Monoxide (CO) Sensors: Used to monitor CO concentration in tunnels, avoiding personnel poisoning caused by vehicle exhaust accumulation. Electrochemical CO sensors are selected, with a measurement range of 0-500ppm (the national standard upper limit for CO concentration in tunnels is 150ppm), precision ≤ ±5ppm, and response time ≤ 3s. When the concentration exceeds 100ppm, a Level 1 warning is triggered (reminding to turn on ventilation equipment); when it exceeds 150ppm, a Level 2 warning is triggered (linking with tunnel entrance signal lights to restrict vehicle entry). The sensors feature an anti-dust interference design, with built-in filter membranes blocking dust from entering the sensor probe, extending the service life to over 2 years (traditional sensors have a life of about 1 year).
(2) Visibility Sensors: Used to monitor visibility reduction caused by smoke and dust in tunnels, warning of fire and driving safety risks. Laser scattering visibility sensors are adopted, with a measurement range of 10-500m (highway tunnel visibility must be ≥ 50m to ensure safe driving), precision ≤ ±5m, and response time ≤ 2s. When visibility drops below 100m, emergency lighting and fog lights in the tunnel are automatically turned on; when it drops below 50m, the traffic control system is linked to close the tunnel. The sensors are waterproof (IP67), can withstand water dripping in the tunnel, and are suitable for tunnels in rainy southern regions.
(3) Road Surface Condition Sensors: Used to monitor slippery and icy road surfaces in tunnels, reminding drivers to slow down. Capacitive road surface condition sensors are selected, capable of identifying four states: dry, wet, waterlogged, and icy. Measurement precision: ice thickness ≤ 0.5mm (error ±0.1mm), water depth ≤ 5mm (error ±0.5mm), response time ≤ 1s. The sensor surface uses wear-resistant ceramic material, capable of withstanding vehicle tire rolling (life ≥ 1 million rollings). Installed on both sides of the road surface in the tunnel to avoid occupying traffic lanes, they are suitable for expressway tunnels.
(4) Fire Sensors: Used to monitor fire hazards in tunnels and achieve early warning. A combination of infrared flame sensors and smoke sensors is adopted. The infrared flame sensor has a response time ≤ 0.5s, can detect flame radiation with wavelengths of 700-1000nm, and has a detection distance of 0-10m. The smoke sensor uses the photoelectric smoke principle, with a measurement range of 0-10%obs/m (optical smoke concentration unit), precision ≤ ±0.5%obs/m, and response time ≤ 3s. The two work in linkage; when both flame and smoke are detected simultaneously, a fire alarm is triggered to avoid false alarms from a single sensor.
Typical Applications
A long mountainous expressway tunnel (total length 4.2km), to cope with unstable surrounding rock, water seepage, and vehicle exhaust emissions, has deployed a 300-unit structure + environment sensor networking system. Specific applications are as follows:
1. Structural Monitoring: One vibrating wire crack sensor is installed every 50m in the tunnel lining, with a total of 80 units installed. In November 2025, the sensors detected that the lining crack width at K2+100 increased from 0.2mm to 0.8mm (exceeding the 0.5mm warning threshold). The system immediately pushed warning information to the tunnel management station. Upon investigation, it was found that the surrounding rock pressure in this section increased due to continuous rainfall, causing structural cracks in the lining. The maintenance team promptly grouted and reinforced this section of the lining, avoiding further crack expansion leading to lining detachment.
2. Environmental Monitoring (CO Concentration): One electrochemical CO sensor is installed every 200m in the tunnel, with a total of 20 units installed. During the morning rush hour in January 2026, the sensors detected that the CO concentration in the middle section of the tunnel rose to 120ppm. The system automatically turned on 3 axial flow fans in the tunnel to strengthen ventilation. After 10 minutes, the CO concentration dropped to 80ppm, returning to normal levels and avoiding further concentration increases affecting the health of personnel in vehicles.
3. Fire Warning: One set of infrared flame + smoke sensors is installed every 100m at the top of the tunnel, with a total of 40 sets installed. In September 2025, a truck in the tunnel caught a small fire due to tire spontaneous combustion. The sensors at K1+500 detected the flame signal within 0.3s of the flame appearing, while the smoke concentration rose to 3%obs/m. The system immediately triggered a fire alarm, linked with the tunnel entrance to turn off signal lights, and notified the fire department to handle the situation on-site. Due to the timely warning, the fire caused no casualties or major traffic congestion.
Technical Upgrade Directions
Tunnel safety monitoring sensors will upgrade towards "multi-parameter fusion, low power consumption, and intelligent linkage" to further improve the comprehensiveness of tunnel monitoring and emergency response efficiency:
1. Environmental Sensor Fusion Upgrade: Integrating CO sensors, visibility sensors, and road surface condition sensors into the same monitoring node to form a "tunnel environment multi-parameter fusion sensor," reducing the number of sensor installations (from independent installation of 3 types of sensors to 1 integrated sensor) and lowering deployment costs. The integrated sensors support edge computing functions, capable of locally fusing and analyzing CO concentration, visibility, and road surface condition data. For example, when "excessive CO concentration + reduced visibility" is detected, it automatically judges it as exhaust accumulation caused by vehicle congestion and prioritizes turning on ventilation equipment without uploading to the cloud for analysis, reducing response time from the original 10s to 3s.
2. Structural Sensor Low-Power Upgrade: Vibrating wire crack sensors and FBG rock bolt stress sensors will adopt energy harvesting technology (e.g., harvesting vehicle vibration energy and light energy in the tunnel) to achieve passive power supply, breaking free from the replacement cycle limitations of traditional battery power (traditional battery life is about 3 years). Meanwhile, the sensor sleep current drops to ≤ 1μA, and operating current to ≤ 10mA. They enter sleep mode when there is no abnormal data, waking up once every 5 minutes to collect data, and collecting in real time when anomalies occur, further reducing energy consumption and suitable for tunnels in remote mountainous areas without mains power.
3. Fire Monitoring Intelligent Linkage Upgrade: Infrared flame + smoke sensors will integrate an AI image recognition module. When a flame signal is detected, it automatically captures on-site images and uploads them to the tunnel management platform. Through AI algorithms, it identifies the fire type (e.g., vehicle spontaneous combustion, cargo burning) and links with the tunnel's emergency broadcast system to play targeted escape guidance (e.g., "Vehicle spontaneous combustion 200m ahead, please evacuate from the right emergency passage"). Meanwhile, the sensors link with fire hydrants and sprinkler systems in the tunnel; when a fire is confirmed, it automatically turns on the 2 nearest sprinkler sets to the fire point for initial fire extinguishing, buying time for fire rescue.
III. Road Surface Condition Monitoring: Sensor Applications for Traffic Efficiency and Safety
Core Monitoring Requirements
As the basic carrier of highway transportation, road surface conditions directly affect driving safety and traffic efficiency. Given the continuous growth in highway traffic volume, especially the increasing proportion of heavy trucks (heavy trucks account for over 30% on some expressways), road surfaces are prone to diseases such as rutting (permanent deformation at wheel tracks), cracking (transverse and longitudinal cracks), and potholes (depressions formed by road surface material falling off). According to 2025 data from the Ministry of Transport, the national expressway road surface disease incidence rate is about 18.2%, of which sections with rutting depth exceeding 15mm account for about 5.1%, easily causing vehicle skidding in high-temperature seasons. In winter, the ice and snow coverage rate on road surfaces in northern regions can reach over 30%, increasing the traffic accident rate by 2-3 times. Real-time monitoring of road surface conditions through intelligent sensors, timely warning of disease risks, and supporting precise maintenance (e.g., milling and repaving for rutted sections, spreading de-icing agents on icy roads) can reduce traffic congestion and accidents caused by road surface issues.
Key Sensors and Technical Parameters
Addressing the core requirements of road surface condition monitoring, sensors suitable for the harsh outdoor road environment (crush-resistant, high/low temperature resistant, UV resistant) must be matched. Technical parameters must meet the accuracy and real-time requirements for road surface disease identification (e.g., rutting depth measurement precision ≤ 1mm, crack width precision ≤ 0.5mm, response time ≤ 5s). Details are as follows:
1. Rutting Sensors: Used to monitor road surface rutting depth and warn of vehicle skidding risks in high-temperature seasons. Laser rutting sensors are recommended, with a measurement range of 0-50mm (the warning threshold for expressway rutting depth is 10mm, and the extreme threshold is 15mm), precision ≤ 0.5mm, and resolution of 0.1mm. They can continuously measure the longitudinal profile of the road wheel track. Sensors are installed at the bottom of highway maintenance vehicles or fixed on gantries, with a scanning frequency ≥ 100Hz (capable of collecting 100 data points per meter of road surface), ensuring measurement accuracy during high-speed movement. The shell uses wear-resistant aluminum alloy material, with crush resistance ≥ 20t, suitable for busy expressway sections with daily traffic exceeding 10,000 vehicles.
2. Crack Sensors: Used to monitor the width and length of transverse and longitudinal cracks on road surfaces, promptly detecting structural damage. Visual crack sensors are selected, integrating high-definition cameras (resolution ≥ 20 megapixels) and AI recognition algorithms, capable of identifying road cracks with widths of 0.2-50mm and lengths ≥ 50mm, with a recognition accuracy ≥ 95%. The sensors feature anti-glare functions (eliminating image interference caused by direct sunlight and headlight illumination), operating temperatures of -30℃~70℃, and can adapt to severe cold in the north and high temperatures in the south. Supporting wireless transmission (4G/5G), they upload crack images and recognition results every 30 minutes, suitable for segmented monitoring of long-distance trunk highways.
3. Pothole Sensors: Used to monitor the depth and area of road potholes, warning of tire blowout risks. Ultrasonic pothole sensors are adopted, with a measurement range of 0-100mm (immediate warning required if pothole depth ≥ 20mm), depth precision ≤ 1mm, area measurement precision ≤ 0.01m² (about the size of a basketball), and response time ≤ 2s. Sensors are installed 5cm below the road surface (using a shallow-buried design), with the surface covered by a wear-resistant rubber layer, crush-resistant life ≥ 1 million times (equivalent to over 10 years of use on a highway with 20,000 daily vehicles). When a pothole depth exceeding 20mm is detected, it automatically pushes the pothole location (combined with BeiDou positioning, precision ≤ 1m) and size to the highway management department for quick repair.
4. Road Surface Icing/Snow Accumulation Sensors: Used to monitor winter road icing and snow accumulation conditions, supporting de-icing and snow removal operation decisions. Dielectric constant icing sensors are selected, capable of distinguishing four states: dry, wet, snow, and ice. Measurement of icing thickness 0-10mm (precision ≤ 0.5mm), snow accumulation thickness 0-50mm (precision ≤ 1mm), response time ≤ 1s. The sensors feature a self-heating function (heating power ≤ 5W), can work normally in environments below -20℃, avoiding false alarms caused by the sensor itself icing up. Installed on highway shoulders or central reserves, with spacing ≤ 500m (spacing ≤ 200m for severely icing sections in the north), forming a grid monitoring network to provide real-time feedback on road ice and snow distribution.
Typical Applications
A northern expressway (total length 200km), to cope with summer rutting and winter icing issues, has deployed a 500-unit road surface condition sensor networking system. Specific applications are as follows:
1. Rutting Monitoring: One laser rutting sensor is installed every 4km on the expressway main line (50 units installed in total). During the high-temperature period in July 2025 (temperature ≥ 35℃), the sensors detected that the rutting depth in the K80-K82 section reached 12mm (exceeding the 10mm warning threshold). The system immediately pushed warning information to the highway maintenance department. The maintenance team milled and repaved this section of the road during the low-traffic night period. After repair, the rutting depth dropped to 3mm, avoiding vehicle skidding accidents during the high-temperature period.
2. Pothole Monitoring: Ultrasonic pothole sensors are shallow-buried installed in vehicle deceleration sections such as expressway interchange ramps and toll stations (pothole high-incidence areas) (100 units installed in total). After a rainstorm in September 2025, the sensor at K120 detected a pothole with a depth of 25mm and an area of 0.03m². The system immediately locked the pothole location (accurate to the right lane at K120+150m) and notified the maintenance vehicle to carry asphalt mixture for repair. From pothole discovery to repair completion took only 2 hours, avoiding traffic congestion caused by subsequent vehicles blowing tires.
3. Winter Ice and Snow Monitoring: One road surface icing/snow accumulation sensor is installed every 200m along the entire expressway (350 units installed in total). During the cold wave in December 2025, the sensors detected icing on the road surface in the K50-K70 section, with an icing thickness of 1-3mm. The system immediately linked with the de-icing agent spreading vehicles of the highway management department, carrying out precise spreading according to the icing distribution fed back by the sensors (spreading only on icing sections, not on non-icing sections). Compared to the traditional full-section spreading method, de-icing agent usage was reduced by 30%, while ensuring that icing sections resumed normal traffic within 2 hours, lowering the winter traffic accident rate.
Technical Upgrade Directions
Road surface condition monitoring sensors will upgrade towards "high precision, self-repair, and vehicle-road coordination" to further improve the intelligence level of road monitoring and its support for traffic efficiency:
1. Rutting Sensor Upgrade: Laser rutting sensors will integrate 3D laser scanning technology, not only measuring rutting depth but also obtaining road surface texture depth (affecting tire grip). Measurement precision is improved to 0.1mm, texture depth measurement range 0-5mm (precision ≤ 0.05mm). Combined with AI algorithms, by analyzing the correlation data between rutting depth and texture depth, the remaining service life of the road surface can be predicted (e.g., predicting that under current traffic volume, the rutting depth of a certain section will reach the 15mm extreme threshold in 1 year), formulating maintenance plans in advance, and realizing the shift from "disease repair" to "preventive maintenance."
2. Pothole Sensor Self-Repair Upgrade: Ultrasonic pothole sensors will be equipped with micro self-repair capsules (containing fast-curing asphalt material). When a pothole depth ≤ 10mm is detected, the self-repair capsule automatically ruptures, releasing asphalt material to fill the pothole, completing initial repair within 2 hours (strength reaching 80% of the original road surface). This can temporarily guarantee vehicle passage, buying time for subsequent formal repairs. The sensors also feature a damage alarm function; after the self-repair capsule is used, they automatically remind the maintenance department to replenish capsules, suitable for emergency repair of highways in remote areas.
3. Ice and Snow Monitoring Vehicle-Road Coordination Upgrade: Road surface icing/snow accumulation sensors will achieve real-time data interoperability with vehicle-mounted terminals. When sensors detect road icing, in addition to notifying the highway management department, they will push icing information to the vehicle navigation systems of passing vehicles via C-V2X communication technology, reminding drivers "Road icing 500m ahead, recommended to slow down to below 60km/h." Meanwhile, the sensors link with expressway variable speed limit signs, automatically adjusting the speed limit for icing sections (e.g., from 120km/h down to 80km/h), forming a closed loop of "sensor monitoring - vehicle warning - traffic control" to further improve winter highway traffic safety.
IV. Sensor Applications in Vehicle-Road Coordination: Core Support for Intelligent Traffic
Core Monitoring Requirements
Vehicle-road coordination is the core direction of highway intelligent upgrade during the "15th Five-Year Plan" period, aiming to achieve "human-vehicle-road-cloud" collaborative decision-making through real-time data interaction between roadside sensors and vehicle-mounted terminals, improving highway traffic efficiency and safety levels. Currently, China's highway transportation faces issues such as normalized congestion (the average congestion duration during peak hours on expressways around major national cities reaches 1.5 hours) and a relatively high accident rate (in 2025 expressway traffic accidents, accidents caused by driver blind spots and sudden road condition changes accounted for over 41%). The traditional "vehicle-end independent perception" model has drawbacks such as limited perception range (e.g., unable to know in advance about a traffic accident 500m ahead) and delayed decision-making. Therefore, multiple types of sensors need to be deployed on the roadside to achieve comprehensive perception of traffic flow, vehicle status, and road environment, and transmit data in real time to vehicle-mounted terminals and cloud platforms, supporting vehicle autonomous driving assistance (e.g., emergency braking warning, lane departure correction) and intelligent traffic flow regulation (e.g., dynamically adjusting speed limits, tidal lane switching), promoting the shift of highway transportation from "passive management" to "active service."
Key Sensors and Technical Parameters
Vehicle-road coordination places extremely high demands on sensor perception precision, response speed, and data transmission capabilities, requiring millisecond-level response, centimeter-level positioning, and simultaneous multi-target identification. Details are as follows:
1. Millimeter-Wave Radar: Used for roadside real-time monitoring of vehicle position, speed, and driving direction, supporting vehicle collision warning. 77GHz millimeter-wave radar is selected, with a detection distance of 0-300m (high-speed scenarios require covering the forward distance of at least 2 lanes), distance precision ≤ 0.1m, speed precision ≤ 0.5km/h, and angle precision ≤ 1°. It can simultaneously track ≥ 100 targets (e.g., vehicles, pedestrians, non-motorized vehicles), with a response time ≤ 10ms. When detecting that a vehicle ahead is braking hard (deceleration ≥ 5m/s²) or a vehicle in an adjacent lane is changing lanes with a collision risk, it immediately pushes warning information to surrounding vehicles. It possesses anti-severe weather capabilities (can penetrate rain, fog, snow) and is suitable for complex traffic scenarios like expressways and urban expressways.
2. LiDAR (Light Detection and Ranging): Used to construct roadside 3D traffic scenarios, achieving high-precision vehicle profile recognition and lane line detection. A 128-line solid-state LiDAR is adopted, with a detection distance of 0-200m, point cloud density ≥ 1 million points/second, and positioning precision ≤ 2cm. It can identify vehicle types (e.g., sedans, trucks, buses), vehicle dimensions (error ≤ 5cm), and lane line deviation (precision ≤ 1cm). Supporting dynamic obstacle classification (recognition accuracy ≥ 98%), it can distinguish stationary vehicles from construction cones, avoiding misjudgment. Operating temperature is -30℃~65℃, protection rating is IP67, suitable for long-term outdoor roadside deployment.
3. High-Definition Cameras: Used for roadside traffic flow statistics, violation behavior identification (e.g., running red lights, not driving in guided lanes), and event detection (e.g., traffic accidents, vehicle breakdowns). 8-megapixel high-definition intelligent cameras are selected, with a frame rate ≥ 30fps, resolution of 3840×2160, supporting AI video analysis algorithms. They can achieve traffic flow statistics of ≥ 200 vehicles per second (precision ≤ ±2%) and identify over 10 types of traffic violations (accuracy ≥ 95%). Featuring wide dynamic range (capable of handling strong light and backlight scenarios) and low-illumination imaging capabilities (can clearly identify vehicle license plates in nighttime environments without streetlights), the event detection response time is ≤ 2s. When a traffic accident is detected, it automatically triggers the traffic control system to adjust signal lights or issue warning information.
4. BeiDou High-Precision Positioning Base Stations: Adopting BeiDou-3 dual-mode positioning base stations, positioning precision is ≤ 2cm horizontally and ≤ 5cm vertically, differential data update rate ≤ 1Hz, and coverage radius ≥ 5km. Supporting multi-constellation compatibility (BeiDou, GPS, GLONASS), they can maintain positioning continuity in signal-shielded areas like tunnels and viaducts (through inertial navigation assistance, maintaining positioning precision ≤ 10cm for 10 minutes after signal interruption). After networking with roadside sensors, they can achieve precise matching of vehicle positions and road condition data. For example, when a vehicle reaches K100+500m, it can accurately push information about road icing and construction at that location.
Typical Applications
An expressway demonstration section (total length 30km), to build a vehicle-road coordination benchmark, has deployed 100 roadside sensors (including millimeter-wave radar, LiDAR, high-definition cameras) and 5 BeiDou high-precision positioning base stations. Specific applications are as follows:
1. Collision Warning: Millimeter-wave radar is deployed at the ramp entrances of expressway interchanges (collision high-incidence areas). When detecting that a mainline vehicle is traveling at 80km/h and a ramp vehicle is merging at 40km/h, with a distance of less than 50m and a collision trajectory, the radar immediately pushes "collision warning" information to the vehicle-mounted terminals of both vehicles via C-V2X technology. Upon receiving the information, the mainline vehicle automatically triggers emergency braking assistance (deceleration 3m/s²), and the ramp vehicle pauses merging, avoiding a collision accident. Within half a year of operation in this demonstration section, the collision accident rate at ramp entrances dropped by 80%.
2. Traffic Flow Regulation: High-definition cameras and LiDAR are deployed every 3km on the expressway main line to statistically monitor traffic flow in each lane in real time (e.g., left lane traffic flow 200 vehicles/hour, right lane traffic flow 300 vehicles/hour). When detecting that traffic flow in a section exceeds 4,000 vehicles/hour (80% of saturated flow), the system automatically adjusts variable speed limit signs, reducing the speed limit for this section from 120km/h to 100km/h, while reminding drivers via vehicle navigation "Heavy traffic flow 3km ahead, recommended to slow down." Through flow regulation, the average congestion duration during peak hours in this demonstration section was shortened from 1.2 hours to 0.5 hours, and traffic efficiency improved by 58%.
3. Emergency Event Handling: High-definition cameras are deployed along the entire expressway. When detecting a sedan stopped in the emergency lane at K15 due to a tire blowout (event detection response time 1.5s), the system immediately initiates the emergency process: first, accurately locking the event location via BeiDou positioning (emergency lane at K15+200m) and pushing "Vehicle broken down in the emergency lane ahead, please avoid" information to passing vehicles within 5km behind; second, linking with the expressway management center to notify the clearance vehicle to handle it; third, adjusting the variable message board 2km ahead to display "Vehicle broken down at K15, pay attention to slow down." From event detection to the clearance vehicle arriving on-site took only 15 minutes, avoiding secondary accidents caused by the broken-down vehicle.
Technical Upgrade Directions
Vehicle-road coordination sensors will upgrade towards "ultra-fusion, low cost, and wide coverage" to promote the popularization of vehicle-road coordination from demonstration sections to national expressways:
1. Sensor Ultra-Fusion Upgrade: Integrating millimeter-wave radar, LiDAR, and high-definition cameras into the same roadside unit (RSU) to form a "multi-modal perception fusion terminal." Through sensor data complementation (millimeter-wave radar excels at measuring speed, LiDAR at measuring profiles, and cameras at measuring images), target recognition accuracy is improved to 99.9%, reducing single-sensor misjudgment (e.g., millimeter-wave radar may misjudge roadside trees as vehicles in rainy days, which can be corrected by combining camera images). The fusion terminal supports edge computing, capable of completing over 80% of traffic event analysis (e.g., traffic accidents, congestion) locally without uploading to the cloud, reducing data transmission bandwidth requirements (bandwidth occupation reduced by 60%) and shortening response time from 10ms to 5ms.
2. Low-Cost Upgrade: Through solid-state design (eliminating mechanical rotating parts) and domestic chip substitution, the cost of LiDAR drops from the current 10,000 RMB/unit to below 2,000 RMB/unit, meeting large-scale deployment needs. High-definition cameras adopt an AI chip integration solution, embedding video analysis algorithms directly inside the camera without the need for additional server deployment, reducing the cost of a single camera by 30%. BeiDou high-precision positioning base stations reduce independent station construction costs by 40% through shared communication tower deployment, promoting the coverage of vehicle-road coordination sensors on over 90% of national expressways.
3. Vehicle-Road-Cloud Integration Upgrade: Roadside sensors, vehicle-mounted sensors, and cloud platforms achieve deep collaboration. For example, when roadside LiDAR detects a construction area 200m ahead, it uploads data such as the 3D coordinates and construction time of the construction area to the cloud. The cloud then synchronizes the data to the vehicle-mounted LiDAR of passing vehicles. The vehicle system, combined with its own position data, automatically plans a driving route to bypass the construction area. Meanwhile, the cloud platform analyzes sensor data from national expressways to form a "traffic heat map - accident risk map - maintenance demand map," providing macro decision support for highway planning and emergency dispatch, realizing the upgrade from "single highway intelligentization" to "national highway intelligent network."
V. Highway Meteorological Environment Monitoring: Warning Sensor Applications Under Extreme Weather
Core Monitoring Requirements
Highway transportation is greatly affected by meteorological conditions. Extreme weather (such as heavy rain, heavy snow, dense fog, high temperature, low temperature icing) is an important factor leading to highway congestion and soaring accident rates. According to 2025 data from the Ministry of Transport, expressway traffic accidents caused by extreme weather nationwide account for 35.5%, of which the accident rate under heavy snow weather increases by 2.8 times compared to normal days, and the visibility reduction caused by dense fog easily leads to multi-vehicle rear-end collisions. Traditional highway meteorological monitoring mainly relies on meteorological stations (spacing usually ≥ 5km), which have problems such as insufficient monitoring density and inaccurate perception of local micro-meteorology (e.g., crosswinds at tunnel entrances, bridge deck icing), unable to meet the "minute-level, meter-level" extreme weather warning needs. Therefore, high-density meteorological sensors need to be deployed along the highway to achieve real-time monitoring of meteorological parameters such as precipitation, visibility, road surface temperature, wind direction and speed, and icing/snow accumulation. Combined with warning algorithms, road meteorological warning information is pushed in a timely manner to support traffic control decisions (e.g., road closure, speed limits) and public travel guidance, reducing the impact of extreme weather on highway transportation.
Key Sensors and Technical Parameters
1. Visibility Sensors: The measurement principle adopts the aerosol forward scattering method, with a measurement range of 5m-10km and resolution ≤ 1m. By detecting the scattered light intensity of aerosol particles in the atmosphere through infrared pulse light, it is converted into Meteorological Optical Range (MOR). They can be densely deployed in fog-prone sections, automatically triggering fog warnings when visibility drops below 200 meters.
2. Wind Speed and Direction Sensors: Adopting the ultrasonic time-difference method, wind speed measurement range is 0～60m/s (±0.1m/s), resolution ≤ 0.01m/s; wind direction measurement range is 0～360° (±2°), resolution 1°. The probe cover features a hidden design to avoid interference from rain and snow accumulation, capable of precisely capturing crosswinds in sections like tunnel entrances and canyons. When the crosswind speed exceeds 10m/s, a strong wind warning is triggered to restrict the passage of large vehicles.
3. Road Surface Condition Sensors: Adopting infrared remote sensing measurement technology, they can identify 6 states of the road surface: dry, wet, waterlogged, snow-covered, icy, and ice-water mixture, with detection accuracy ≥ 95%. Road surface temperature measurement range is -40℃ to +60℃ (±0.3℃), water depth 0-10mm, ice thickness 0-2mm, snow thickness 0-2mm, and slipperiness resolution 0.01 (0.01 is extremely slippery, 0.82 is strong friction). Non-contact measurement, compatible with concrete, asphalt, and other road materials. When the road surface temperature is below 0℃ and humidity exceeds 85%, a road surface icing warning is automatically triggered.
4. Piezoelectric Rainfall Sensors: Adopting the PVDF piezoelectric film rain-sensing principle, embedded AI neural networks distinguish raindrop signals. Measurement range is 0-4mm/min (error ≤ ±4%), resolution 0.01mm, capable of distinguishing precipitation types like rain and snow. It can effectively avoid false triggers caused by interference from gravel, dust, vibration, etc. When 1-hour rainfall exceeds 30mm (heavy rain warning standard), it links to trigger a heavy rain warning.
5. Multi-Element Integrated Meteorological Sensors: Adopting integrated principles such as diode junction voltage method (temperature), capacitive method (humidity), and piezoresistive method (pressure). Technical parameters: air temperature -40-60℃ (±0.3℃, resolution 0.01℃); air humidity 0-100%RH (±3%RH, resolution 0.1%RH); atmospheric pressure 300-1100hpa (±0.25%, resolution 0.1hpa). They can simultaneously monitor basic meteorological elements such as temperature, humidity, and pressure, providing data support for comprehensive judgment of extreme weather. For example, under high-temperature weather (temperature ≥ 35℃), it can warn of road surface softening risks.
Typical Applications
1. Extreme Meteorological Monitoring of a Super Large Bridge in Guizhou: Located in a canyon area, instantaneous wind force can reach level 10, winter road surface icing thickness reaches 5-10mm, and frequent dense fog causes a sharp drop in visibility. 12 integrated meteorological monitoring stations (including ultrasonic wind speed and direction sensors, infrared road surface condition sensors, and visibility meters) are deployed, adopting a "4G + Ethernet + BeiDou short message" multi-modal transmission solution. Application results: Strong wind warning accuracy is 100%, road surface icing is warned 25 minutes in advance, and the traffic accident rate under extreme weather drops by 85%. Through precise positioning of icing sections, de-icing agent usage is reduced by 40%, and road closure duration is shortened by 60%.
2. Ice and Snow Warning on the Handan Section of the Beijing-Hong Kong-Macau Expressway in Hebei: Winter low temperatures easily cause road icing, and traditional manual patrol warnings lag behind. Remote-sensing road surface condition sensors and temperature sensors are deployed along the route, with solar power and mains power dual-mode ensuring continuous data transmission. Road icing can be warned 2 hours in advance, and the accident rate drops by 63% compared to the same period in previous years. Linking with maintenance departments achieves "point-to-point" snow melting and de-icing, improving operational efficiency by 50%.
Technical Upgrade Directions
1. Sensor Miniaturization and High-Density Networking: Developing smaller and lower-power micro-meteorological sensors to achieve high-density deployment at 500-meter intervals in risk sections such as bridges, tunnel entrances/exits, and sharp curves, improving micro-meteorological monitoring precision.
2. Optimization of Multi-Parameter Coupled Warning Algorithms: Breaking through the limitations of single-parameter warnings, constructing a "temperature-humidity-road surface condition-wind force" multi-parameter coupled model. For example, when the road surface temperature is below 0℃, humidity > 80%, and accompanied by snowfall, the icing warning level is automatically elevated.
3. Intelligent Energy and Self-Maintenance Technology: Promoting solar + supercapacitor power supply solutions to solve the power supply problem in remote sections. Developing sensor self-cleaning functions (e.g., ultrasonic de-icing, dustproof coatings) to reduce manual maintenance costs under extreme weather.
4. Extension of Vehicle-Road Coordination Meteorological Services: Integrating highway meteorological sensor data into the Internet of Vehicles platform, pushing "lane-level" meteorological warnings to drivers via OBU (On-Board Units) or navigation APPs (e.g., "Left lane icing 300 meters ahead"), realizing the extension from "road warning" to "vehicle warning."
5. Strengthening Extreme Weather Prediction Capabilities: Combining AI meteorological prediction models, utilizing historical meteorological data and real-time sensor data to achieve 1-hour short-term prediction of extreme weather such as heavy rain and heavy snow, leaving longer emergency response time for traffic control departments.
VI. Highway Transportation Sensor Networking and Data Application
Future Networking Optimization Directions
The core advantage of comprehensive highway transportation monitoring lies not only in the precise point perception of various sensors but also in the networking linkage and data interoperability of multiple devices, scenarios, and dimensions. Traditional highway sensors are mostly deployed in single points and fragments, suffering from data silos, coverage blind spots, transmission lag, and rigid networking, unable to adapt to the development needs of smart highway comprehensive perception, real-time warning, and collaborative decision-making. Therefore, relying on a new generation of communication and information technologies, it is necessary to comprehensively optimize the sensor networking architecture, build a highly reliable, flexible, fully covered, and intelligent perception network, break down multi-source data transmission barriers, maximize the value of monitoring data, and provide core network and data support for highway intelligent control, vehicle-road coordination implementation, and proactive traffic maintenance.
1. Space-Air-Ground Integrated Networking: Introducing low-altitude UAVs and satellite remote sensing technology to form a "space-air-ground" three-dimensional monitoring network with ground sensors. UAVs equipped with high-definition cameras and LiDAR regularly patrol remote sections (supplementing the insufficient coverage of ground sensors); satellite remote sensing data is used to obtain large-scale meteorological and geological data along the highway (e.g., regional heavy rain warnings, mountain landslide hazards), complementing ground sensor data to improve monitoring comprehensiveness.
2. Intelligent Self-Healing Networking: Developing sensor network self-healing algorithms. When a sensor or communication node fails (e.g., roadside RSU damage), it automatically switches communication paths (e.g., forwarding data through adjacent RSUs) to ensure uninterrupted data transmission. Meanwhile, it supports dynamic addition/exit of sensors to the network (e.g., temporarily deployed construction area sensors can quickly access the existing network), improving networking flexibility.
3. Deepening Edge-Cloud Collaboration: Promoting deep collaboration between edge computing nodes and cloud platforms. Deploying some AI models (e.g., long-term traffic flow prediction, bridge life assessment) on the cloud, while edge nodes handle real-time data collection and short-term analysis. The cloud is responsible for long-term trend prediction and macro decision-making, forming a collaborative model of "fast response at the edge, deep analysis on the cloud" to enhance data application value.
4. Bidirectional Data Interaction between Vehicle and Network: Achieving bidirectional data transmission between the highway sensor network and vehicle-mounted terminals. The sensor network pushes real-time road conditions, meteorology, and accident warning information to vehicle-mounted terminals; vehicle-mounted terminals feed back vehicle status (e.g., braking status, fault information) to the sensor network, forming a "road-to-vehicle - vehicle-to-road" data closed loop, supporting higher-level vehicle-road coordination applications (e.g., cooperative obstacle avoidance for autonomous vehicles).
Platform Data Correlation Characteristics
Under the highway transportation sensor networking system, massive monitoring data covers multiple dimensions such as bridge structure, tunnel environment, road surface condition, traffic flow, meteorological environment, and vehicle operation. Various data do not exist independently but possess strong correlation, coupling, and complementarity. Relying on a unified smart highway data platform, multi-source heterogeneous data can be fused, aggregated, correlated, classified, and linked, breaking the monitoring limitations of traditional single data dimensions and realizing the upgrade from "single-point data monitoring" to "comprehensive data empowerment," comprehensively improving the refinement, intelligence, and scientific level of highway transportation control. Its core correlation application characteristics are mainly reflected in the following three aspects:
1. Multi-Source Data Fusion and Correlation to Achieve Full-Scenario Situational Awareness: The platform can integrate various sensor data such as structural monitoring, environmental monitoring, traffic monitoring, and meteorological monitoring, breaking down scenario data barriers. Fusing and correlating heterogeneous data such as bridge deformation, tunnel gas concentration, road icing, road traffic flow, and rain/snow weather can accurately judge potential risks that cannot be identified by single data. For example, combining low-temperature meteorological data with road surface condition data to predict road icing risks; combining tunnel traffic flow data with air quality data to dynamically adjust ventilation and lighting equipment, achieving full-scenario, three-dimensional highway operation situational awareness.
2. Spatiotemporal Data Linkage and Correlation to Support Dynamic Warning and Precise Maintenance: All sensor-collected data are attached with time and spatial coordinate attributes. The platform can serially correlate data based on spatiotemporal dimensions to build a full-lifecycle data archive for highway facilities. By comparing the changing patterns of monitoring data in different periods and sections, it can accurately identify issues such as facility aging, performance degradation, and sudden road condition changes, realizing the shift from passive repair to proactive prediction and precise maintenance. Meanwhile, relying on spatiotemporal data linkage, the occurrence location and impact scope of accidents, diseases, and disasters can be quickly located, providing precise data support for emergency handling, command dispatch, and traffic guidance.
3. Business Data Closed-Loop Correlation to Empower Comprehensive Smart Control Upgrade: The data platform opens up the correlation channels of perception data, operation and maintenance data, control data, and vehicle interaction data, forming a complete business closed loop of "data collection - analysis and judgment - instruction issuance - implementation - data feedback." On the one hand, sensor monitoring data guides highway maintenance, traffic control, and emergency warning businesses; on the other hand, business execution results and vehicle feedback data are reversely incorporated into the platform analysis system to continuously optimize monitoring models, warning algorithms, and control strategies, constantly improving the intelligent operation level of highway transportation, and providing solid data support for vehicle-road coordination, smart road networks, and comprehensive traffic governance.
VII. Industrial Status and Future Trends of Highway Transportation Sensors
Industrial Development Status
1. Market Size and Growth
In recent years, given the acceleration of highway intelligent construction, the market size of highway transportation sensors has continued to expand. According to data from the China Intelligent Transportation Industry Alliance, the market size of China's highway transportation sensors reached 15.87 billion RMB in 2025, a year-on-year growth of 23.4%, of which the expressway field accounts for over 60% (mainly used for bridge health monitoring and vehicle-road coordination), the ordinary national and provincial highways and urban highways field accounts for about 30%, and the rural highways field accounts for about 10% (limited by construction costs, currently mainly meteorological and road surface condition sensors). It is expected that from 2026 to 2030, the market will maintain a compound growth rate of 18%-20%, and the scale is expected to exceed40 billion RMB by 2030. Core driving factors include: the implementation of the "15th Five-Year Plan" highway intelligentization special plan (planning to add 32,000 km of intelligent expressways), the national promotion of vehicle-road coordination demonstration projects, and the frequent occurrence of extreme weather driving up the demand for meteorological sensors.
2. Industry Chain Layout
The highway transportation sensor industry chain has formed a complete layout of "upstream core components - midstream sensor manufacturing - downstream system integration and application":
(1) Upstream Core Components: Mainly involving chips, sensitive materials, and packaging components. Domestic enterprises have gradually achieved breakthroughs in the field of MEMS chips (e.g., pressure sensor chips, acceleration sensor chips) (e.g., SMIC (Semiconductor Manufacturing International Corporation) and GoerTek Microelectronics have mass-produced MEMS pressure chips for bridge stress sensors). However, high-precision LiDAR chips (128 lines and above) and millimeter-wave radar RF chips still rely on imports (e.g., Texas Instruments and Infineon account for over 70% of the market share). In terms of sensitive materials, domestic enterprises have achieved self-supply in the fields of piezoelectric ceramics (used for rainfall sensors) and infrared detection materials (used for road icing sensors) (e.g., Shenzhen Sunlord Electronics, Shanghai Institute of Ceramics). However, special materials such as high-temperature strain gauges (used for bridge strain monitoring) still need to be imported.
(2) Midstream Sensor Manufacturing: Divided into general-purpose sensors and dedicated sensors. The general-purpose sensor market (e.g., temperature, humidity, rainfall sensors) is fiercely competitive. Domestic enterprises occupy over 60% of the market share relying on cost advantages, and product prices have declined year by year (e.g., ordinary rainfall sensors dropped from 2,000 RMB/unit in 2020 to 800 RMB/unit in 2024). Dedicated sensors (e.g., strain sensors for bridge health monitoring, 77GHz millimeter-wave radar for vehicle-road coordination) have high technical barriers and are currently jointly occupied by domestic and foreign enterprises. Foreign enterprises account for about 40% in the high-end bridge sensor field (mainly supplying large super-large bridge projects). Domestic enterprises are rising rapidly in the field of vehicle-road coordination millimeter-wave radar and LiDAR, with the market share increasing from 15% in 2020 to 35% in 2024.
(3) Downstream System Integration and Application: Mainly including intelligent transportation system integrators, highway construction and maintenance enterprises, and traffic management departments. Domestic leading integrators dominate the system integration of vehicle-road coordination sensor networks relying on technological advantages such as 5G and edge computing. Traditional highway maintenance enterprises are gradually laying out sensor data applications, developing highway maintenance management platforms, and realizing the transformation from "manual patrol" to "data-driven maintenance." Traffic management departments promote the application of sensors in traffic control (e.g., dynamic speed limits, tidal lanes) through government procurement. In 2025, the sensor procurement budget of provincial traffic management departments nationwide increased by 31% year-on-year.
3. Regional Development Pattern
The industry presents a pattern of "agglomeration in the east, follow-up in the central and west":
(1) Eastern Region: Centered on the Yangtze River Delta and the Pearl River Delta, gathering over 60% of highway transportation sensor enterprises. The Yangtze River Delta region (Shanghai, Jiangsu, Zhejiang) focuses on high-end sensor R&D and system integration. Relying on the foundation of the automobile and electronic information industries, it forms a full industry chain ecosystem of "sensors - vehicle-road coordination - intelligent transportation." The Pearl River Delta region (Shenzhen, Guangzhou) focuses on general-purpose sensor manufacturing and IoT communication, achieving large-scale sensor production relying on supply chain advantages.
(2) Central and Western Regions: Represented by Beijing, Hubei, and Sichuan, relying on universities and research institutes, they have formed characteristics in the field of dedicated sensor R&D (e.g., bridge stress sensors, tunnel environment sensors). Meanwhile, leveraging policies such as the Western Development and the Yangtze River Economic Belt, they promote the application of sensors in highway construction in the central and western regions.
Existing Problems and Challenges
1. Technical Bottlenecks to be Broken Through
(1) Low Localization Rate of High-End Sensors: The localization rate of core products such as high-precision LiDAR (128 lines and above), millimeter-wave radar RF chips, and high-temperature strain gauges is less than 30%. Reliance on imports leads to high costs (e.g., the price of imported 128-line LiDAR is about 15,000 RMB/unit, which is 3 times that of domestic 64-line LiDAR), limiting large-scale deployment.
(2) Difficulty in Multi-Sensor Data Fusion: Data formats of different types of sensors (e.g., LiDAR and high-definition cameras) and sensors from different manufacturers are not unified, lacking unified data fusion standards, leading to severe data "silo" phenomena (e.g., an expressway simultaneously deploys meteorological sensors from 3 manufacturers, and the data cannot be directly integrated and analyzed).
(3) Insufficient Adaptability to Extreme Environments: The performance of some sensors decreases under extreme conditions. For example, low temperatures in northern winters (below -30℃) shorten the battery life of battery-powered sensors by 50%. High-temperature and high-humidity environments in the south accelerate the aging of sensor packaging (service life shortened from 5 years to 3 years). Salt spray corrosion in coastal areas increases the damage rate of sensor metal components.
2. Contradiction Between Cost and Cost-Effectiveness
(1) Excessive Cost of Dedicated Sensors: Dedicated products such as FBG sensors for bridge health monitoring and 77GHz millimeter-wave radar for vehicle-road coordination have high prices (a single FBG sensor is about 5,000 RMB, a single 77GHz millimeter-wave radar is about 8,000 RMB), leading to a per-kilometer cost of 500,000-800,000 RMB for expressway intelligent transformation, which is unaffordable for ordinary national and provincial highways (per-kilometer transformation budget is usually ≤ 200,000 RMB).
(2) High Operation and Maintenance Costs: Sensors require regular maintenance after deployment (e.g., cleaning, calibration, replacing faulty equipment). According to estimates, the annual operation and maintenance cost of expressway sensors is about 15%-20% of the initial procurement cost (e.g., the annual operation and maintenance cost of sensors for a 30km smart expressway reaches 2 million RMB). Long-term operation and maintenance pressure is high, especially the maintenance difficulty in remote sections is higher.
3. Imperfect Standard System
(1) Lack of Unified Technical Standards: Currently, highway transportation sensors lack unified performance testing standards (e.g., precision testing methods for visibility sensors, target recognition accuracy testing standards for millimeter-wave radar). Product performance varies greatly among different manufacturers (e.g., visibility sensors with the same nominal measurement range of 0-10km may have actual errors ranging from ±2% to ±10%), affecting engineering application effects.
(2) Lack of Data Security Standards: Security standards in the process of sensor data transmission, storage, and sharing are imperfect, posing risks of data leakage and tampering (e.g., in 2025, meteorological sensor data of an expressway in a certain province was maliciously tampered with, leading to erroneous heavy rain warnings and affecting traffic control decisions). Meanwhile, the lack of cross-departmental data sharing security specifications restricts the release of data value.
4. Development Gap Between Domestic and Foreign Countries
(1) Technological R&D Gap: Foreign enterprises started early in high-end sensor R&D, possessing over 20 years of technical accumulation, and lead domestic enterprises by 3-5 years in aspects such as LiDAR point cloud processing algorithms and millimeter-wave radar anti-interference technology. Although domestic enterprises have caught up in the field of general-purpose sensors, they still need breakthroughs in basic fields such as core chip design and special material R&D.
(2) Industrial Ecology Gap: Foreign countries have formed a complete ecology of "sensor R&D - chip manufacturing - system integration - application implementation." For example, Germany's Bosch cooperates with Mercedes-Benz and BMW to promote the application of vehicle-road coordination sensors on expressways. The domestic ecology is still in a scattered state, with insufficient collaboration among sensor enterprises, automakers, and traffic departments, leading to a slower pace of technology implementation (e.g., in domestic vehicle-road coordination demonstration sections, compatibility issues between sensors and vehicle-mounted terminals still need to be run in).
(3) Standard Formulation Gap: The International Organization for Standardization (ISO) has released multiple highway transportation sensor standards (e.g., ISO 21219-2 for performance testing of road meteorological sensors). Foreign enterprises actively participate in standard formulation and hold the right to speak. Domestic related standards (e.g., "Technical Specifications for Highway Transportation Sensors") are still being formulated, with fewer enterprises participating in international standard formulation and insufficient international competitiveness.
“15th Five-Year Plan” and Medium-to-Long-Term Development Trends
1. Technical Breakthrough Directions
(1) Localization of Core Components: The state will increase support for the R&D of high-precision sensor chips and sensitive materials, promoting the domestic substitution of LiDAR chips and millimeter-wave radar RF chips, with the goal of increasing the localization rate of high-end sensors to over 50% by 2027. Meanwhile, developing new sensitive materials (e.g., low-temperature-resistant battery materials, salt-spray-resistant packaging materials) to improve sensor adaptability in extreme environments (e.g., sensor battery life extended to over 8 years in a -40℃ low-temperature environment).
(2) Multi-Sensor Fusion Technology Upgrade: Formulating unified data fusion standards for highway transportation sensors, developing AI-based multi-source data fusion algorithms (e.g., using Transformer models to fuse LiDAR, camera, and millimeter-wave radar data), achieving seamless integration of data from different types and manufacturers of sensors, and improving data accuracy (e.g., target recognition accuracy improved from 95% to over 99%).
(3) Development of Low-Power and Long-Life Technologies: Promoting energy harvesting technology (e.g., solar, vibration energy harvesting) to solve the power supply problem for sensors in remote sections. Adopting new packaging technologies (e.g., nano-coating packaging) to improve sensor corrosion and aging resistance, with the goal of extending the average service life of sensors from 5 years to 10 years by 2030 and reducing operation and maintenance costs by 50%.
2. Market Expansion Trends
(1) Extension from Expressways to Ordinary Highways: Given the decline in costs (e.g., the price of domestic LiDAR drops to below 2,000 RMB/unit), highway transportation sensors will gradually cover ordinary national and provincial highways and rural highways, focusing on deploying meteorological sensors and road surface condition sensors to solve the problems of insufficient extreme weather warnings and untimely discovery of road diseases in rural highways (it is expected that from 2026 to 2030, the compound growth rate of the sensor market size for ordinary national and provincial highways will reach over 25%).
(2) Penetration from New Construction to Existing Transformation: During the "15th Five-Year Plan" period, the country will launch the intelligent transformation of existing expressways (planned transformation mileage reaches 50,000 km), promoting the installation of sensors in existing bridges, tunnels, and road surfaces (e.g., installing strain sensors on existing bridges, installing smoke sensors in existing tunnels). The existing transformation market will become a new growth point (the proportion of existing transformation is about 32% in 2025, expected to increase to 50% in 2030).
(3) Increase in Cross-Boundary Fusion Applications: Highway transportation sensors will cross-boundary fuse with fields such as the low-altitude economy and smart logistics. For example, using highway meteorological sensors to provide meteorological data support for low-altitude UAV delivery, and using road surface sensors to provide road condition information for smart logistics vehicles, expanding sensor application scenarios (it is expected that the market size proportion of cross-boundary applications will reach 20% in 2030).
3. Industrial Ecology Upgrade
(1) Deepening Industry-University-Research-Application Collaboration: The government will promote cooperation among universities, research institutes, and enterprises (e.g., establishing joint laboratories for highway transportation sensors) to accelerate the transformation of technological achievements. Meanwhile, encouraging automakers and traffic departments to participate in sensor R&D to form a closed loop of "demand traction - R&D - application."
(2) Accelerating Industry Chain Integration: Leading enterprises will integrate the industry chain through mergers and acquisitions, cooperation, etc. For example, Huawei acquires domestic LiDAR enterprises to improve the vehicle-road coordination sensor layout. Midstream sensor manufacturing enterprises will extend to upstream core components to reduce reliance on imported components and enhance industry chain resilience.
(3) Accelerating Internationalization Development: Domestic enterprises will actively participate in international standard formulation (e.g., promoting the inclusion of China's highway transportation sensor standards into the ISO system) and expand overseas markets (e.g., highway construction projects in Southeast Asia and the Middle East). Relying on cost advantages and technological cost-effectiveness, they will increase international market share (it is expected that the overseas market share of domestic enterprises will increase from 5% in 2025 to 20% in 2030).
4. Policy Support and Guidance
(1) Continuous Introduction of Special Policies: During the "15th Five-Year Plan" period, the state will issue the "Special Plan for Highway Intelligentization Development," clarifying the deployment goals for highway transportation sensors (e.g., by 2027, the expressway sensor coverage rate reaches 100%, and the ordinary national and provincial highway coverage rate reaches 60%), and providing financial subsidies (e.g., providing 30%-50% financial subsidies for sensor deployment on rural highways) to promote sensor popularization.
(2) Gradual Improvement of the Standard System: The State Administration for Market Regulation will join hands with the Ministry of Transport to formulate a series of standards such as "General Technical Conditions for Highway Transportation Sensors" and "Data Security Specifications for Highway Transportation Sensors," unifying sensor performance indicators, testing methods, and data formats to regulate market order.
(3) Establishment of Data Sharing Mechanisms: The government will promote the establishment of a highway sensor data sharing platform among departments such as transportation, meteorology, and public security, breaking down data barriers and realizing cross-departmental and cross-regional data sharing (e.g., fusing meteorological data from meteorological departments with road condition data from traffic departments to improve the accuracy of extreme weather warnings) and releasing data value.
Conclusion
As the core support for smart transportation construction, highway transportation sensors are transforming from single monitoring tools into full-chain enablers of "perception - networking - decision-making - service." From guarding the structural safety of bridges and tunnels, precise perception of road surface conditions, to early warning of extreme meteorology, efficient linkage of vehicle-road coordination, and then to the data closed loop of sensor networking and the gradual maturity of the industrial ecology, their applications have been deeply integrated into the full-lifecycle management of highway transportation.
Currently, although the highway transportation sensor industry faces challenges such as insufficient localization of high-end technologies, high cost and operation/maintenance pressure, and an imperfect standard system, with the continuous intensification of national policies, gradual breakthroughs in core technologies, and deepening of industry chain collaboration during the "15th Five-Year Plan" period, it will realize the leap from "partial intelligence" to "comprehensive intelligence" in the future. It will not only support the intelligent upgrade of expressways and ordinary national/provincial highways but also extend to rural highways and cross-boundary services for the low-altitude economy, providing a solid technological foundation for improving highway traffic efficiency, ensuring traffic safety, and promoting the construction of a transportation powerhouse.
Facing the future, highway transportation sensors need to further strengthen the development logic of "demand-oriented, technology-centric, and application-targeted," continuously breaking through key technical bottlenecks, reducing deployment and operation/maintenance costs, and improving the standard system, ultimately achieving deep collaboration of "human - vehicle - road - environment" and making highway transportation safer, more efficient, and more intelligent.
Author | Guo Yuansheng, Deputy Director of the Science and Technology Committee of the Jiusan Society Central Committee and Executive Vice Chairman of the China Sensor and IoT Industry Alliance
Editor | Yang Pengyue Art Editor | Ma Liya Supervisor | Zhao Chen