× close
KIMM researcher Han-Min Lee explains the research results regarding the newly developed technology. Credit: Korea Institute of Mechanical Materials (KIMM)
Real-time off-road environment recognition technology that detects foreign objects such as dust, mud, snow, and rain during off-road automatic driving of construction machinery, agricultural machinery, UGVs (unmanned ground vehicles), etc., and removes these sensor signals on the vehicle. It was first developed in South Korea.
These new technologies are expected to be applied to industrial machinery such as excavators, dump trucks, search vehicles, and military self-driving vehicles in the future, providing a safe working environment for workers.
A research team led by Chief Researcher Lee Han-min of the Industrial Machinery DX Division, Virtual Engineering Platform Research Division, Korea Institute of Mechanical Materials Research has developed off-road environment recognition technology for driving in the following off-road environments. We have transferred technologies for mountainous areas, waterfronts, and snowy areas to related companies, including sensor protection and cleaning technology, sensor signal correction technology, and drivable area recognition technology.
Among the off-road environment recognition technologies newly developed by KIMM, the “sensor protection/cleaning module” technology can be used to spray detergent or wipe off muddy water and mud that may come into contact with sensors during off-road autonomous driving. . Use the wiper to remove dirt in real time and remove dirt almost completely.
Furthermore, “sensor signal correction” technology that removes minute foreign objects such as dust, snow, and rain generated while driving allows for more stable off-road autonomous driving even under non-structural environmental conditions. It becomes possible. Like bad weather.
Furthermore, by using the “drivable area estimation technology” developed by KIMM, it is possible to detect not only general obstacles, but also steep slopes, potholes, uneven roads, etc., and avoid these obstacles. By automatically identifying alternative routes for vehicles, machinery and vehicles can be prevented from running. Protects from collision with other objects.
Furthermore, KIMM has also developed “driving control technology” that controls vehicle driving in real time by selecting only the necessary functions from the various technologies listed above.
Until now, there has been no sensor protection technology suitable for off-road environments where vehicles are covered with dirt and mud, or technology that removes sensor signals caused by foreign objects such as dust, snow, and rain in real time. The LiDAR or camera sensor signal contains matter.
Additionally, there was a lack of real-time drivable area estimation technology that could recognize not only uneven obstacles such as trees and rocks, but also hollow obstacles such as cliffs and holes.
Meanwhile, the newly developed off-road environment recognition technology has improved processing speed by more than 1.5 times while maintaining important performance indicators such as sensor dirt recovery rate, sensor noise removal accuracy, and off-road driving area estimation accuracy. We will realize technology that is equal to or better than existing technology, paving the way for the practical application of off-road autonomous driving control.
“These are technologies to solve the problem of environmental awareness, which can be a dangerous obstacle during off-road autonomous driving,” said Hanmin Li, chief researcher at KIMM.
President Lee said, “We will do our best to apply our newly developed technology not only to the autonomous driving of industrial machinery such as excavators, dump trucks, and tractors, but also to the autonomous driving of unmanned vehicles.” Military vehicles such as tanks and search vehicles. ”
