Project background

China has a huge highway network with a total length of more than 5.35 million kilometers, and the length of expressways ranks among the highest in the world.

With the growing stock of infrastructure, highway maintenance management has entered a new era of "construction which is as important as maintenance".

With the increase of maintenance expenditure year by year, how to improve maintenance efficiency and reduce management cost has become an urgent problem to be solved.

Challenges

1. High-frequency inspection and labor costs: daily inspection work is heavy, high labor costs.



2. Data collection and management: Relying on manual data collection and recording has problems such as strong subjectivity and low efficiency, and there are personnel safety risks.

 

3. Limitations of annual testing: Although annual testing is reliable, it has high cost and low frequency, and cannot meet the real-time needs.

Policy orientation

The "14th Five-Year Plan" for the Development of Highway Maintenance Management emphasizes the promotion of digitization of facilities and the improvement of the digitization level of maintenance management, especially AI-based automated inspection and the quality management of maintenance projects through the Internet of Things.

Solutions

In view of such application scenarios, TWOWIN TECHNOLOGY and Lulan Software jointly developed a lightweight intelligent highway inspection solution.

The inspection equipment and system collect the road image data through the vehicle camera; The T930G edge computer is used for AI image analysis to identify road safety risks and diseases efficiently and accurately.

Upload images through the object storage HTTP(s) protocol interface; Upload metadata through MQTT protocol; At the same time, combined with 5G/4G communication and high-precision positioning technology, data is uploaded to the cloud in real time, providing a comprehensive and real-time assessment and display of road conditions.

In the case of poor 4G/5G network, the data will be cached and transmitted. The device has a built-in media server, and the mobile APP can access and play real-time inference videos through RTSP protocol.

Technological advantage

1. High efficiency: AI intelligent analysis improves inspection efficiency by 10 times;

2, High accuracy: detection rate >91.6%, false detection rate <5.4%, area calculation error <8.6%, to ensure data accuracy;

3, Real-time response: real-time detection, real-time reporting, second-level response;

4, High reliability: to provide a mechanism such as storage continuous transmission, not limited by the network environment, can ensure the safe and stable transmission of data;

5, Remote control: built-in VPN, mirroring deployment, remote upgrade and maintenance.

6. Rich algorithm model: Support multi-scenario and multi-version AI algorithm integrated management, including but not limited to: pavement scattered detection, pavement water icing detection, slope rock fall monitoring, weather observation along the way, road asset inventory, etc.

Customer revenue

1. Grasp the road condition with a map: realize real-time monitoring and evaluation of road conditions through daily inspection.

2. Inspection efficiency leap: greatly improve work efficiency, reduce labor costs and avoid security risks.

3. Real-time risk warning: through the on-board edge intelligent analysis, real-time detection of road risks, take preventive measures in advance.

4. Scientific decision support: based on the daily accumulation of effective data, to provide scientific decision-making basis for conservation management, reduce maintenance costs.

Implementation effect

The lightweight highway intelligent inspection program has been successfully implemented in 12 provinces and cities across the country, and has been widely recognized by customers.