Modeling of Wheel-Soil Interaction for Small Ground Vehicles Operating on Granular Soil.

Terramechanics;Discrete Element Method;Small Unmanned Ground Vehicles;Mechanical Engineering;Engineering;Mechanical Engineering

Smith, William ClarkePerkins, Noel C. ;

University of Michigan

关键词: Terramechanics;    Discrete Element Method;    Small Unmanned Ground Vehicles;    Mechanical Engineering;    Engineering;    Mechanical Engineering;   

Others  :  https://deepblue.lib.umich.edu/bitstream/handle/2027.42/108811/wsmithw_1.pdf?sequence=1&isAllowed=y

瑞士|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

Unmanned ground vehicles continue to increase in importance for many industries, from planetary exploration to military defense.These vehicles require significantly fewer resources compared to manned vehicles while reducing risks to human life.Terramechanics can aid in the design and operation of small vehicles to help ensure they do not become immobilized due to limited traction or energy depletion.In this dissertation methods to improve terramechanics modeling for vehicle design and control of small unmanned ground vehicles (SUGVs) on granular soil are studied.Various techniques are developed to improve the computational speed and modeling capability for two terramechanics methods.In addition, a new terramechanics method is developed that incorporates both computational efficiency and modeling capability.First, two techniques for improving the computation performance of the semi-empirical Bekker terramechanics method are developed.The first technique stores Bekker calculations offline in lookup tables.The second technique approximates the stress distributions along the wheel-soil interface.These techniques drastically improve computation speed but do not address its empirical nature or assumption of steady-state operation.Next, the discrete element method (DEM) is modified and tuned to match soil test data, evaluated against the Bekker method, and used to determine the influence of rough terrain on SUGV performance.A velocity-dependent rolling resistance term is developed that reduced DEM simulation error for soil tests.DEM simulation shows that surface roughness can potentially have a significant impact on SUGV performance.DEM has many advantages compared to the Bekker method, including better locomotion prediction, however large computation costs limit its applicability for design and control.Finally, a surrogate DEM model (S-DEM) is developed to maintain the simulation accuracy and capabilities of DEM with reduced computation costs.This marks one of the first surrogate models developed for DEM, and the first known model developed for terramechanics.S-DEM stores wheel-soil interaction forces and soil velocities extracted from DEM simulations.S-DEM reproduces drawbar pull and driving torque for wheel locomotion on flat and rough terrain, though wheel sinkage error can be significant.Computational costs are reduced by three orders of magnitude, bringing the benefits of DEM modeling to vehicle design and control.