Published 2020-09-30This post is also available in Swedish
Digital forestry shapes up
Planning robots and autonomous forest machines. Digital maps and decision support in real time. In Mistra Digital Forest, digital forestry is taking shape. The aim is to facilitate application of the research in the relatively near future.
In Mistra Digital Forest, research is being developed in close collaboration with Swedish forest companies. Two current examples are algorithms that help in planning forest thinning and forest machines that can navigate automatically.
‘We’ve tackled digitisation in forestry in a way that allows the knowledge to be applied and have an impact in the business world. Back when we were working on the programme application, we already had a workshop with the forest companies to pick up on their needs and see how the research results can be put to practical use.’
So says Erik Willén, Process Manager of digitisation at Skogforsk (the Forestry Research Institute of Sweden), one of Mistra Digital Forest’s programme partners. He heads Efficient Forestry, one of the programme’s four work packages, which focuses on digitisation and automation of forestry operations such as felling, planting and thinning.
‘There are many sustainability benefits of using more digital opportunities and autonomous forest machines — lighter ones that run more efficiently, reducing both carbon emissions and site damage. Other benefits also arise. The work environment for operators of these machines is a tough one, and improves when they can be controlled remotely instead. Then more people may be attracted to this kind of work.’
Optimised, emission-cutting routes
As an example of environmental and climate benefits, Willén mentions the ‘planning robot’ that is now being developed and tested. With this robot, forestry operations can be planned in detail, using algorithms. All types of important data are collected, such as the nearest paved roads, off-road terrain and sensitive natural areas. The data are combined to produce the most optimised route. The forester receives a completed proposal instead of spending time on manual planning. The research is conducted in collaboration with Laval University in Quebec, Canada.
‘Tests show that using the planning robot shortens the driving distance per felling operation, compared with manual planning, by eight kilometres. The hope is that the method might already be used in forestry within the programme period,’ Willén says.
Precision forest management using digital maps is another important component of the work package. With laser scanning of forest surfaces, data on volume, density, soil conditions and more are collected. Digital maps existed already, but in Mistra Digital Forest even more detailed information is gathered. The purpose is to create more precise documentation for assessing the effects of afforestation, regrowth and thinning in terms, for example, of ecosystem services.
Digital twin for automated forest machines
The biggest challenge, and one focus area of the work package, is development of autonomous forest machines. Going from manual tasks to greater automation and driverless vehicles requires a great deal of research and testing along the road. In Mistra Digital Forest, the focus is on developing forest machines that, based on data, can scan and respond to the terrain, choosing not only the best routes but which trees to fell.
‘Unlike self-driving cars that can relate to black roads with white lines, forest machines encounter all types of terrain. Now we’re training the machine how to manoeuvre in forests using data on the physical characteristics of the ground by means of a “digital twin” at Umeå University,’ Willén relates.
Beyond Mistra’s research programme, in Vinnova’s Auto2 (‘Automation for autonomous terrain mobility’) project, the first steps towards remote control have been taken. In a lab for remote control, drivers sit in front of monitors, wearing virtual-reality (VR) glasses, and test various solutions. In Mistra Digital Forest, further research is in progress on design of the stations — the nature of the driver environment, and the driver support and types of tool needed for efficient operation that spares the land and terrain from damage, as well as promoting comfort for the individual.
Forest variation spells challenges
In early October there will be a presentation for forest companies that are showing keen interest in the work. Some are providing operators to test the autonomous solutions that, as Willén points out, are a key part of the work. Forest-machine manufacturers, too, are displaying great commitment to development. These include some of the industry’s giants, and it is important for the forest companies’ views to reach them. But replacing the entire machine park at a fast pace is not the right way to go about it, according to Willén.
‘Before completely new machine concepts are developed, there are still a lot of research questions to answer. Although we can make use of research in other industries, like mining, forests are more varied. Now we’re working with an empty vehicle cab and sensors to test the solutions conceptually and, as you should appreciate, it takes time to reach the finish line.’
Another area in Willén’s work package is developing decision support in real time, to streamline every step of the production process and enhance environmental awareness.
To disseminate the research results as the work proceeds, stakeholders from industry are invited to quarterly meetings on various themes. These provide opportunities to field questions and comments about ongoing research, and report on developments in the various subject areas. One meeting has been about the planning robot, and another about how felling methods facilitate automation. In November, the theme is autonomous off-road driving.
‘The meetings are reflective and get people on side. Instead of just presenting results, we jointly identify challenges and needs in the ongoing research. We believe this gives us great potential for change,’ Willén says.
Mistra Digital Forest
Mistra Digital Forest focuses on creating functional digitisation that integrates the entire forest value chain, and where industrial processes start at source: in the forest. The vision is to create digital solutions that contribute to a circular bioeconomy.
The programme consists of four work packages (WPs):
- Programme leadership, WP coordination and communication
- Forest facts
- Efficient forestry
- Assets and benefits of forests and forest products.