NEWS Get rid of the cabin and the driver. Then, let the timber truck move by itself through the forest, over stumps, shrubs and stones, to a transport route, unload the felled trees, turn around and drive back to pick up a new load.
This may sound like science fiction but it is a challenge which researchers at the Intelligent Off-Road Vehicles project (IFOR) at Umeå University, in cooperation with forest machine maker Komatsu, believe they can overcome. The fact that the forwarder, as the vehicle is called, also weighs ten tonnes, is nine metres long, with an articulated joint and supposed to perform its tasks during day and night and in all types of weather does not make the task any easier for the researchers, who are now in the process of developing a prototype. – The vehicle must be able to answer several questions such as – where am I, where do I want to go, how will I get there and where have I been, explains Thomas Hellström from the Department of Computing Science and Principal Investigator of the Autonomous Navigation Project. If the vehicle cannot answer these questions, it will not know where it is and it will not be able to find its way back if, for some reason, it strays off the course that it is familiar with.
– However, the vehicle must also be capable of seeing obstacles. It needs to understand what is ahead of it so that it does not drive into something or injure a person who gets in the way, Thomas Hellström continues.
The vehicle has several sensors which can detect position, speed, movements, obstacles and many other things. One of the researcher’s tasks is to make these sensors function optimally. Lasers, cameras, radar and satellite navigation together can provide data the vehicle can react to. However, only solving the vehicle’s ability to know where it is can be a tough nut to crack. GPS guidance via satellite is one way of doing it but this needs to be supplemented, as navigation via GPS does not always work in dense forest with poor visibility.
The need for driverless vehicles is considerable and the researchers know that there are several industries interested in autonomous navigation. For instance, there are examples of driverless machines in the mining industry. Vehicles which are capable of doing heavy demolition work or mine clearance in war-affected areas are also in demand.
Photo: Samuel Bengtson
Editor: Karin Wikman