⚙️
🌟 Najlepsze dopasowanie
Wpisz szukaną frazę, aby zobaczyć propozycje AI.

Over 20 cameras will inspect a truck in a single pass. A new solution from Poznań scientists

Over 20 cameras will inspect a truck in a single pass. A new solution from Poznań scientists
Photo: Łukasiewicz-PIT

How can you check the technical condition of a truck’s exterior without stopping the vehicle or performing time-consuming inspections? Scientists from the Łukasiewicz – Poznań Institute of Technology have developed the TruckVision 360 inspection gate, which uses over 20 cameras to record the appearance of a passing vehicle. The system creates detailed photographic documentation, allowing for fleet condition monitoring and the identification of changes occurring during operation. The solution is currently undergoing final tests before its planned implementation.

 

Trucks cover hundreds of kilometers every day, transporting multi-ton loads and operating in diverse road conditions. Intensive use promotes the formation of damage – from minor scratches and body dents to progressive corrosion or damage to bodywork components. For transport companies, especially those managing large vehicle fleets, regular condition checks pose a significant organizational challenge. In many cases, inspections are performed by the driver, and the results depend on their attentiveness, available time, and the method of documenting observed irregularities. Automating the inspection process can be the answer to these challenges.

The objective eye of the camera

Scientists from the Łukasiewicz – Poznań Institute of Technology have developed the TruckVision 360 inspection gate, which enables the recording of a vehicle’s exterior condition as it passes through. The device is equipped with a system of over 20 cameras positioned to capture the front, rear, both sides, and the top of the truck. The captured images are then processed by algorithms developed at the Institute, which combine them into extensive panoramic images of individual vehicle surfaces.

One of the most important advantages of our solution is the ability to perform an inspection without the need to stop the vehicle. The driver passes through the gate at a low speed, and the system automatically records its appearance. This allows us to obtain repeatable photographic documentation without involving the driver in additional inspection-related tasks – explains Sebastian Szymczyk from Łukasiewicz – PIT.

The system also automatically recognizes vehicle license plates, assigning the collected images to them along with the date and time of the pass. The information is sent to a dedicated application that can be used by authorized personnel responsible for fleet management. In practice, this means the ability to build a visual history for every vehicle. By comparing documentation from successive passes, the user can notice changes in the bodywork’s appearance, identify new damage, and narrow down the period in which it may have occurred. It is particularly useful to perform inspections before starting a route and after the vehicle returns to the base. This not only streamlines fleet control but also provides documentation material helpful in clarifying the circumstances of damage.

Artificial intelligence to support inspection

The next stage in the development of TruckVision 360 will be the use of artificial intelligence for the automatic analysis of recorded images. Scientists are working on a solution that will compare the vehicle’s photographic documentation from successive passes and identify potential changes in its appearance. Ultimately, the system is intended to support the detection of new damage, reducing the need for manual review of collected photos.

We want the gate not only to document the vehicle’s condition but also to help interpret the collected data. Thanks to the use of artificial intelligence algorithms, it will be possible to automatically indicate areas where changes have appeared. This is particularly important for large fleets, where hundreds of passes can be recorded daily – explains Sebastian Szymczyk.

Not just the bodywork. Scientists are also working on tire inspection

TruckVision 360 is also being developed towards the automatic assessment of tire condition. One of the solutions being developed is a system for measuring tire tread depth as the vehicle passes through the gate. For this purpose, the structure of the station includes special ramps equipped with measuring sensors. Their task will be to acquire data allowing for the determination of tread wear without the need for traditional manual measurement. The solution remains at the research and development stage and requires further testing, but its ultimate application could significantly expand the capabilities of the entire system. Regular tread depth checks allow not only for assessing tire wear but also for identifying irregularities that may indicate problems with tire pressure or suspension geometry.

Technology for large fleets

The TruckVision 360 gate is being created primarily for companies managing fleets of trucks and buses. Automating the process of documenting vehicle conditions can reduce the time needed for inspections, facilitate damage identification, and streamline the management of fleet operational history.

We want to create a solution that will be a real support for transport companies. Automatic recording of vehicle conditions allows for organizing the inspection process and provides access to documentation from successive passes. In the future, thanks to the development of image analysis and the tread measurement system, we want to further expand the range of information that can be obtained during a single pass through the gate – concludes Sebastian Szymczyk.

Currently, TruckVision 360 is undergoing final tests preparing the solution for implementation. In parallel, work is continuing on additional functionalities, including automatic image analysis and tread depth measurement.

The TruckVision 360 project is being carried out by the Łukasiewicz – Poznań Institute of Technology as part of the Łukasiewicz Center subsidy.