Control of Rough Terrain Vehicles Using Deep Reinforcement Learning

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...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	IEEE robotics and automation letters Ročník 7; číslo 1; s. 390 - 397
Hlavní autoři:	Wiberg, Viktor, Wallin, Erik, Nordfjell, Tomas, Servin, Martin
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	Piscataway IEEE 01.01.2022 The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Témata:	Artificial Intelligence Automatic Control autonomous vehicle navigation Biomedical Engineering Blocking Computational modeling Computer Science Computer Science Applications Computer Vision and Pattern Recognition Control and Optimization Control and Systems Engineering Control methods Controllers datalogi Deep learning Deep learning methods Forestry fysik Human-Computer Interaction Joints Mechanical Engineering model learning for control Operators Physics reglerteknik Reinforcement learning Robotics robotics and automation in agriculture and forestry Robotteknik och automation Rough terrain Skills Solid modeling Suspensions (mechanical systems) Vehicles Virtual environments Wheels
ISSN:	2377-3766, 2377-3766
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	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<inline-formula><tex-math notation="LaTeX">^\circ</tex-math></inline-formula>, 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.
Bibliografie:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ISSN:	2377-3766 2377-3766
DOI:	10.1109/LRA.2021.3126904