Can autonomous trucks find their path in mines and on construction sites even if they change their layout and shape or the visibility is difficult? At the voestalpine lime mine, the project partners voestalpine, Indurad, Bell and AIT Austrian Institute of Technology have successfully deployed an autonomous truck.
In the quarry
Lime is part of our everyday life. It is needed for construction, e.g. for concrete, cement and paint, for steel production, but also as a fertiliser and for wastewater and exhaust gas purification. We also find it in cosmetics, pharmaceuticals and other everyday objects.
At Steyrling the voestalpine Stahl GmbH extracts several 100,000 tonnes per year of this versatile raw material. In the quarry, the material is taken by a truck to primary crushing before it is transported down into the valley in order to be further processed.
What is special about this truck is that it operates without a driver. Despite poor sight caused by dust clouds after blasting or dense fog that often prevails on the mountain, it heads for its destination, manoeuvres, positions itself and dumps the material into the designated container.
itruck by Indurad; cooperation Indurad/ Bell/ Voestalpine/ AIT; videorelease by Voestalpine, 2021.
"Autonomous working machines are supposed to relieve us humans of assembly and dangerous work and counteract bottlenecks in the construction industry." Andreas Vrabl, Head of the AIT Center for Vision, Automation & Control
With this mining truck, the project partners voestalpine, Indurad, Bell and AIT Austrian Institute of Technology are dispelling the widespread opinion that autonomous work machines will hardly be able to find their way in environments that are constantly changing their shape - like on a construction site or mine. The project also shows that autonomous vehicles can avoid obstacles and operate - even under difficult visibility conditions.
Robust Radar Systems
"One of our research fields at the AIT Center for Vision, Automation & Control focuses on autonomous vehicles and working machines. Especially in construction areas or quarries, the work environment is often harsh. Dirt, dust and the weather all cause problems to the sensors that recognise the environment. Based on our previous research, we have further developed algorithms that we have mainly used for optical sensors (such as in cameras, LIDAR). Now we have adapted them to robust radar sensors," says Manfred Gruber, head of the Assistive & Autonomous Systems research group at the AIT center.
For the vehicle to find its path through the construction site and for it to react dynamically to the given situation, it needs a reliable 3D model of the environment. Based on the sensor data algorithms build and update the 3D map in real time. They also determine the current position of the vehicle. "The 3D modelling of the environment with sensor data from imaging radar systems is the basis for the autonomous path planning. We have fused the data from different sensor systems to precisely determine the vehicle's position in real time. This is the only way to ensure safety and productivity," explains Christian Zinner, researcher at the Assistive & Autonomous Systems group at the AIT Center for Vision, Automation & Control.
Led by Manfred Gruber, the Assistive & Autonomous Systems research group has already developed intelligent driver assistance systems for rail vehicles which are now installed in the trams made by Bombardier. The systems continuously scan the environment and compare the data with the tram's own position. It warns the driver or automatically initiates the braking process. The demand is rising worldwide as the system demonstrably increases road safety.
The research group is currently working on assistive systems for off-road vehicles as well as on autonomous machines that carry out loading and unloading processes in the transport and logistics sector.
Links
Contact
