Russian Scientists Teach Robots a New Way to Navigate
The new approach makes autonomous technologies affordable for mid-sized businesses without sacrificing reliability or precision.
Researchers at the Moscow Institute of Physics and Technology have unveiled a navigation system for autonomous forklifts that relies on ceiling-mounted markers. The solution lowers the cost barrier for deploying driverless technologies in mid-sized companies while maintaining accuracy and operational reliability.
A “Visual Cortex” Reads Ceiling Markers
Expensive sensors and complex algorithms are no longer required. The idea behind the system is both simple and effective. A navigation camera, effectively the robot’s “visual cortex,” reads circular optical markers mounted on the ceiling and encoded in a specific way. To determine its exact position, the machine only needs to see two such markers.
The key advantage of the development lies in its affordability. The hardware-and-software package Bespilotny Pogruzchik (Autonomous Forklift) is several times cheaper than existing alternatives while matching them in performance. Bespilotny Pogruzchik receives tasks from a warehouse management system, can transport loads of up to two tons, and features multi-level safety protection.
It Will Politely Ask You to Step Aside
Reliability is ensured through additional cameras that monitor movement and enable precise positioning near storage racks. Personnel safety is handled by a combination of a mechanical bumper, lidars, and intelligent computer vision.
The project demonstrates that breakthrough technologies do not have to be expensive to be effective. Pilot deployments have already been successfully completed at logistics centers operated by VkusVill, X5 Group, and SberLogistics, and the first commercial contract is now being carried out for the company NizhBel.
