University of Tartu launches initiative to improve safety of autonomous heavy machinery

An international team of researchers and entrepreneurs, led by the University of Tartu, is starting to develop artificial intelligence-based solutions that will enable self-driving heavy vehicles in earth-moving, logistics and forestry to operate safely and reliably in difficult conditions. The European Commission supports the four-year project with 8 million euros.

Heavy machinery is often used in harsh environments, such as excavations, forests or urban traffic in northern weather conditions. Self-driving vehicles need to operate on icy roads in winter, and excavators need to be able to cope on terrain with loose soil. All kinds of unpredictable environmental conditions – such as ice, snow, mud and unmovable large rocks – make the use of self-driving technology difficult and dangerous, as an autonomous vehicle may not be able to cope with rapid and unpredictable changes in the weather.

An international team of researchers, led by Arun Kumar Singh, Associate Professor of Collaborative Robotics at the Institute of Technology of the University of Tartu, will look for AI-based ways to ensure the safety and reliability of self-driving heavy vehicles in such conditions. According to Singh, the aim is to improve autonomous operations in land management, forestry and logistics to reduce the fuel consumption of vehicles and environmental impacts such as soil damage. The results of the project will be put to practical use thanks to partners, including Novatron, a Finnish heavy machinery technology company, Komatsu, a Swedish engineering company, and Clevon, a developer of autonomous vehicles in Estonia.

According to Martin Appo, Head of Autonomous Driving at Clevon, the project represents a significant opportunity for the development of both the vehicles developed by Clevon and the industry as a whole. "Our involvement in the project means that the final solutions will be tested and deployed on the Clevon platform, allowing us to further refine and prove the capabilities of our technology. Together with cutting-edge research and industry partners, we are developing systems that ensure the safety and efficiency of self-driving vehicles in the most demanding conditions," said Appo.

Research partners in the European Commission-funded project “Explainable, Safe, Contact-Aware Planning and Control for Heavy Machinery Manipulation and Navigation” (XSCAVE) include Tampere University and Aalto University from Finland, technology university CTU from the Czech Republic and Umeå University, research institute FZI from Germany, Toshiba Europe from the United Kingdom as well as Algoryx Simulation AB from Sweden.

Further information: Arun Kumar Singh, Associate Professor of Collaborative Robotics at the Institute of Technology, arun.singh@ut.ee