LogoLogo
ISTVSConferences
  • 11th Asia-Pacific Regional Conference of the ISTVS
  • Details
  • Program
  • Submissions
  • Registration
  • Sponsors
  • Papers of the 11th Asia-Pacific Regional Conference of the ISTVS
    • 0303 / Composite Beam Tests with Closed Cell Polyurethane and Aluminum Foam
    • 0356 / Design and Simulation Analysis of Intelligent Suspension for Manned Lunar Rover
    • 0861 / Review of the Reconfigurable Wheel-Tracked System
    • 0963 / A Wheel and Vehicle Mobility Index Based on Traction and Velocity...
    • 1128 / Semi-Active Reinforcement Learning Suspension Control for the Off-Road Vehicles
    • 1491 / Design and Verification of a Creeping Mars Rover
    • 1534 / Foothold Selection Considering Constraint and Slippage Evaluation for Legged Robots
    • 1561 / Prominent Problems and Thoughts of “Paddy Soil-Terrain Machine System”...
    • 1655 / Modeling of Lunar Rover Vehicle Wheel-Soil Interaction Using Fem-Dem Method
    • 2034 / A Comprehensive Lumped Parameter Approach for the Dynamic Simulation...
    • 2149 / Investigation of the Shear Stress Dynamics on Silty Loam Soil and Measurement...
    • 2190 / Tyre Parameterization Tests: Dynamic vs. Static
    • 2539 / Model Predictive Control of a Robot Driven Vehicle for Testing of Advanced Driver...
    • 2632 / Energy Consumption Analysis of Door Opening with a Mobile Manipulator...
    • 2643 / An Improved Simultaneous Localization and Mapping Method Base on LeGO-LOAM and Motion Compens
    • 3351 / Benchmarking of Compression Testing Devices in Snow
    • 4054 / Field Validation of Egress Process for Planetary Rover
    • 4243 / Soil Compaction Monitoring Technique Using Deep Learning
    • 4260 / The Running Gear Construction Impact on Obstacles Overcoming by Light High-Mobility UGV
    • 4409 / Design of Self-Driving Bulldozer System
    • 4744 / Terrain Classification Using Mars Raw Images Based on Deep Learning Algorithms...
    • 4774 / Steadily Learn to Drive with Virtual Memory
    • 4782 / Experimental Study of Track-Soil Interactions of the Steering Performance of Tracked...
    • 4812 / Multi-Fidelity Machine Learning Modeling for Wheeled Locomotion on Soft Soil
    • 4827 / Introducing Polibot: A High Mobility Tracked Robot with Innovative Passive Suspensions
    • 5060 / Bionic Quadruped Robot for Mars Surface Exploration
    • 5408 / Ride Comfort Comparison Between Suspension Modes: Input Towards Designing Difference...
    • 5800 / Interaction Modeling and Dynamic Control Strategy for C-Shaped Leg with Sandy Terrain...
    • 5979 / Research on Drag Reduction Performance of Sliding Plate of Rice Direct Seeding Machine...
    • 6174 / Factors Affecting Bevameter Soil Characterization
    • 6316 / Perceptive Locomotion of Legged Robot Coupling Model Predictive Control and Terrain Mapping
    • 6718 / Research on Vehicle Running Performance on Paved Roads Covered with Falling Volcanic Ash
    • 6796 / Nonparametric Terrain Estimation Based on the Interaction Simulation Between Planetary...
    • 7018 / A Review of Modeling and Validation Techniques for Tire-Deformable Soil Interactions
    • 7092 / A Time Domain Passivity Controller for Teleoperation of Four Wheeled Differential...
    • 7199 / Vehicle Dynamic Factor Characterized by Actual Velocity and Combined Influence...
    • 7233 / Study of Passive Steering Mechanism for Mars Surface Exploration Rovers
    • 7399 / Tire-Soil Tangential Force Reinforcement Learning Modeling
    • 7878 / A Method for Fast Obtaining of Soil Shear Strength Index Based on Dem Free-Fall Cone...
    • 8131 / Parameters Calibration of Red Clay Soil in Hilly Area of Southwest China for Discrete...
    • 8349 / The Effect of Integrating a Bio-Inspired Convex Structure with a Low-Surface Energy...
    • 8654 / Construction of a Soil Clods Recognition Bench-Scale Experiment for Discrete Element...
    • 8658 / Investigation of the Relationship between the Cone Index and the Physical and...
    • 9352 / 3D-DEM Simulation and Post-Process Method of Wheel-Terrain Interaction for Planetary Rovers
    • 9768 / Design and Traction Performance Test of Bionic Paddy Wheel Based on Cattle Hoof
    • 9913 / Acquisition of Flipper Motion in Step-Climbing of Tracked Robot Using Reinforcement Learning
  • Abstract-only submissions
  • Statement on Publication Ethics and Malpractice
Powered by GitBook

© International Society for Terrain-Vehicle Systems :: www.istvs.org

On this page
  1. Papers of the 11th Asia-Pacific Regional Conference of the ISTVS

4812 / Multi-Fidelity Machine Learning Modeling for Wheeled Locomotion on Soft Soil

Paper presented at the 11th Asia-Pacific Regional Conference of the ISTVS

Previous4782 / Experimental Study of Track-Soil Interactions of the Steering Performance of Tracked...Next4827 / Introducing Polibot: A High Mobility Tracked Robot with Innovative Passive Suspensions

Last updated 2 years ago

Title: Multi-Fidelity Machine Learning Modeling for Wheeled Locomotion on Soft Soil

Authors: Vladyslav Fediukov* /a,b, Felix Dietrich /b, and Fabian Buse /a

a/ Institute of System Dynamics and Control, German Aerospace Center (DLR), Weßling, Germany b/ Institut für Informatik, TU München, Garching b. München, Germany

*Corresponding author: vladyslav.fediukov@dlr.de

Abstract: Wheeled vehicles are the most convenient and widespread locomotion machines for the majority of research, industrial or private tasks. A perceptible share of wheeled vehicles is used on soft soil. Modeling wheel locomotion in these situations is challenging, because of the non-proportional relation between applied shear stress and the soil’s deformation. Currently, various conventional simulation approaches are used to describe wheel–soil interaction, ranging from detailed numerical methods with particle-level simulations to simpler empirical models, where a big part of physical formulas are set up a priori, empirically. The ultimate wheel locomotion modelling tool should have high-quality onboard predictions but within a reasonable time. The trade-off is unachievable with the current simulation tools. In this project, we argue that using Machine Learning (ML) we can build a tool with the quality of high-fidelity and speed of lower-fidelity simulations. To fit this requirement, we are combining data from several models with different fidelities, in order to build a multi-fidelity ML model. In the model, forces and torques acting on the wheel are predicted using input data like the wheel’s trajectory, surface and soil characteristics. The quality of this model will be validated by Terramechanics Robotics Locomotion Laboratory (TROLL) at Deutsche Zentrum für Luft- und Raumfahrt (DLR), a robotic single-wheel test bed designed to perform wheel–soil interaction experiments automatically. Early results show that, in simplified scenarios, our proposed method can be used to create efficient, multi-fidelity numerical models for locomotion prediction, including uncertainty estimation for the predictions.

Order the full paper:

ISTVS members: receive three papers per year as part of your membership via the ISTVS Member Portal:

https://doi.org/10.56884/WGPV6693
https://www.istvs.org/proceedings-orders/paper
https://istvs.knack.com/member-portal/