Sammanfattning

We explore the potential to control terrain vehicles using deep reinforcement in scenarios where human operators and traditional control methods are inadequate. This letter presents a controller that perceives, plans, and successfully controls a 16-tonne forestry vehicle with two frame articulation joints, six wheels, and their actively articulated suspensions to traverse rough terrain. The carefully shaped reward signal promotes safe, environmental, and efficient driving, which leads to the emergence of unprecedented driving skills. We test learned skills in a virtual environment, including terrains reconstructed from high-density laser scans of forest sites. The controller displays the ability to handle obstructing obstacles, slopes up to 27 degrees, and a variety of natural terrains, all with limited wheel slip, smooth, and upright traversal with intelligent use of the active suspensions. The results confirm that deep reinforcement learning has the potential to enhance control of vehicles with complex dynamics and high-dimensional observation data compared to human operators or traditional control methods, especially in rough terrain.

Nyckelord

Deep learning methods; reinforcement learning; autonomous vehicle navigation; model learning for control; robotics and automation in agriculture and forestry

Publicerad i

IEEE Robotics and Automation Letters
2022, Volym: 7, nummer: 1, sidor: 390-397
Utgivare: IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC

SLU författare

• Nordfjell, Tomas

  • Institutionen för skogens biomaterial och teknologi, Sveriges lantbruksuniversitet
    

Globala målen

SDG9 Bygga motståndskraftig infrastruktur, verka för en inkluderande och hållbar industrialisering samt främja innovation


UKÄ forskningsämne

Robotteknik och automation


Publikationens identifierare

DOI: https://doi.org/10.1109/LRA.2021.3126904

Permanent länk till denna sida (URI)

https://res.slu.se/id/publ/114522