Predicting the mobility of tracked forestry machines operating on Nordic forest soil

R. Palaniappan, Machine Design, KTH Royal Institute of Technology, Stockholm, Sweden, revathi@kth.se

A. Pirnazarov, Machine Design, KTH Royal Institute of Technology, Stockholm, Sweden, abpirnaz@kth.se

U. Sellgren, Machine Design, KTH Royal Institute of Technology, Stockholm, Sweden, ulfse@kth.se

B. Löfgren, Swedish Forestry Research Institute Skogforsk, Uppsala, Sweden, bjorn.lofgren@skogforsk.se

The predominant process for forest harvesting in Northern Europe is the fully mechanized cut-to-length method that is based on two articulated and 6-8 wheeled machines; a harvester that fells the trees and cuts them in predefined lengths, and a forwarder that transports the logs to a loading area for further transport to a processing facility.

North European forest soil is a very complicated and sensitive bed. It consists of large areas of sand and clay with embedded stones and rocks and with several root layers of significant importance, and marshland with low bearing capacity. Furthermore, in thinning operations, the root layers not only significantly contribute to the bearing capacity but one requirement is also that the roots are protected, i.e. they must not break. Wheeled vehicles are typically more agile and maneuverable than tracked vehicles, but the tracked vehicles give lower ground pressures and thus smaller rutting and less damage to the roots that are embedded in the soil.

Track-soil interaction covers issues on track-soil friction, rutting, soil compaction, shearing, traction and drawbar pull. The track rut depth is one of the key factors to determine vehicle performance and energy consumption and the damages caused to ground and vegetation.  Measurement and characterization of the mechanical properties of a multilayer soil that efficiently may be used for mobility predictions is a complex task. Usually, there are two techniques used to determine soil mechanical properties, the cone penetrometer technique and the Bevameter technique. Since forestry soil is not homogeneous, the Bevameter technique is comparatively less efficient for characterizing Nordic forest soil. The widely used WES-method is a set of semi-empirical methods that are based on cone-penetrometer data to describe soil mechanical properties and to model wheel-soil and track-soil interaction.

This paper presents a comparison of the capabilities of different empirically-based terrain interaction models to predict the rutting measured in full scale field tests. Soil penetration and track rut depth (both first and multi-pass) were measured in the field tests for a medium sized forwarder (loaded and unloaded) equipped with three different types of tracks mounted on bogie wheels. In the tests, the ground pressure at different depths below the wheel-soil interface and the soil moisture, were also measured. Comparisons between tracked and wheeled forwarders are also presented. Matlab was the primary modeling and analysis tool.

The applicable conference tracks where the paper will fit are indicated in the table below:

Track	Description	Highly applicable	Applicable	Not applicable
1	Soil and terrain modeling and characterisation	X
2	Planetary rovers and mobile robotics			X
3	Tires, wheels and tracks modeling	X
4	Agricultural and earth moving equipment		X
5	Operation on snow and ice			X
6	Vehicle dynamics, mobility and safety		X
7	Terramechanics		X
8	Other - Forestry machines	X