A Russian System Prevents Forklift-Pedestrian Collisions Without Radio Tags
Russia has developed a system that can detect pedestrians directly in video footage rather than relying on radio tags – and can stop warehouse equipment before a collision occurs.
Russian company KASTA has tested Antinaezd (Collision Prevention), an AI camera-based system that does not require workers to wear personal RFID tags. Its purpose is to prevent forklifts from striking people, including in areas with limited visibility. The algorithm analyzes video feeds in real time, detects people, estimates their distance and approach trajectory, and alerts the driver when a collision risk arises. Depending on the system configuration and its integration with a particular vehicle, it can also apply the brakes automatically.
The key difference from conventional solutions is that workers do not have to wear tags. Traditional systems rely on radio-frequency identification: if a person is not carrying a tag, the equipment effectively cannot detect them. KASTA takes a different approach. Its camera looks for the person, not an electronic identifier. This is particularly important for contractors, visitors and temporary workers, who cannot always be issued tags in advance.
Outlook and Potential
One advantage of the technology is that it can be integrated with equipment already in operation, without requiring companies to wait until they renew their vehicle fleets. Connecting the system to digital fleet-management platforms could offer another benefit: companies would be able to collect data on near misses, identify high-risk areas in warehouses and adjust vehicle routes accordingly. The most immediate opportunity is to equip existing forklift fleets with AI cameras at warehouses, logistics terminals and industrial sites.
A hybrid approach could offer further benefits in the future by combining video analytics with ultra-wideband (UWB) tags and other sensors. Cameras can compensate for blind spots in radio-based systems, while radio-frequency devices can help in areas where visibility is limited. Together, these technologies could significantly improve the overall reliability of pedestrian protection.
The most likely export market is the Commonwealth of Independent States (CIS), where demand is growing for solutions that modernize warehouse operations. International competition, however, is strong: major manufacturers of industrial vehicles already offer similar technologies.
From Tags to Computer Vision
Russian industrial companies have been developing collision-prevention systems for several years. In 2022, Karelsky Okatysh (Karelian Pellet), a mining company, deployed a UWB-tag-based system for haul trucks and shift buses, with radio-frequency identification at the center of the approach. In subsequent years, the focus shifted toward video analytics. At Kola Mining and Metallurgical Company (Kola MMC), researchers tested thermal imaging cameras connected to neural networks. In 2025, the system was extended to reach stackers at the same facility, demonstrating the need to adapt algorithms to different types of vehicles.
Other competing products have also entered the market. In May 2025, Kar-For introduced its AI Motion system, followed in August 2026 by an upgraded Camkord camera. KASTA is therefore entering an established market, where competition is no longer about the concept itself but about reliability, ease of integration and total cost of ownership.
New Safety Standards
Testing Antinaezd without personal tags marks an important step in industrial safety. The technology could protect everyone in an area where industrial vehicles operate, not just workers wearing tags. The next few years are likely to be a period of pilot operation. If the system proves reliable and maintains a low false-alarm rate, its adoption could expand across warehouses and distribution centers. Further integration with corporate occupational safety and incident-analysis systems may follow.
Some challenges remain. Camera performance depends on lighting conditions and clean optics, while automatic braking must not put the load or the vehicle itself at risk. Integrating the system with different forklift models will also require individual configuration.
The ultimate measure of success will be real-world results: how much the system reduces the number of dangerous close encounters and helps prevent injuries. If the figures are convincing, the technology could become a standard safety feature in Russian warehouses and eventually gain recognition in international markets.
