Rheological Interaction Physics of Wheeled Locomotion in Soft Substrates for Improved Mobility: MIT Component
Abstract
When roads are unavailable, wheeled and tracked vehicles must traverse complex terrain. The terrain can have variable slope, cohesion, heterogeneity, and ability to support load, which can impede progress or cause the wheels to become stuck. The disciplines of off-road engineering and terramechanics have led to successful empirical models of wheeled locomotion for large, heavy vehicles. However, recent evidence suggests these models may not be applicable to locomotors that are small and lightweight, such as rovers and unmanned ground vehicles. In this proposal, we will undertake an investigation of the rheological physics of wheel-substrate interaction. Our theoretical/computational studies, in close collaboration with Dan Goldman s experimental laboratory at Georgia Tech will lead to, terra dynamic principles and models that can aid development of wheeled-locomotion control in a variety of settings --- spanning a range of substrate media, wheel geometry, and loading --- and will lead potentially to model-driven smart-wheel algorithms for mutable/deformable wheels. Achieving reliable motion requires new insight into the physics of wheeled locomotion in soft substrates across a wide range of rotation speeds. Many current terramechanics models are empirical and apply to large, slowly moving vehicular systems. Further, materials like sand and mud are not easily described constitutively. Our group at MIT has recently made progress in determining the key constitutive ingredients to accurately computing intrusion force in dry granular media. Using these continuum models and novel numerical solvers, we have shown that the experimentally observed Resistive Force Theory of granular media, a rapid intrusion force model developed in Goldman s group, can be completely reconciled from tension-free frictional plasticity theory. Pushing further, from this connection we have also shown that despite the known complexities of granular rheology, there actually exist a number of exploitable and accurate scaling relations that can be used in designing wheeled granular locomotors. We have also written fast solvers for RFT s, which can be run fast enough to be used in feedback control to explore a new model-driven paradigm for in-situ locomotive optimization. Motivated by our success developing intrusion and locomotive models and numerical tools, we will develop a joint three-year program with Goldman s group to investigate and discover principles of wheeled locomotion and general intrusion on complex terrain. We will examine the physics arising in wheeled vehicular locomotion scenarios involving sloped terrain, wet granular terrain, high inertia, and heterogeneous obstacles. Specifically, by combining continuum models with our state-of-the-art meshless solvers, together with Goldman s experimental tests and validations, we aim to produce new RFT s applicable in cohesive muds exhibiting yield stress behavior, high speed situations where various inertial and fluidization effects come in, and inclined substrates. Once the continuum-level underpinnings are validated in experiments for these various circumstances, we will seek out corresponding scaling laws to permit wheel design using downscaled lab tests. We will also produce our own wheeled control algorithm based on optimization theory applied to RFT, and test its ability to maintain optimal traction in our own laboratory using a custom morphable wheel design. We will also use our collaboration with Goldman s group to develop computational tools, such as Discrete Element and Material Point Methods, for accurate and/or rapid rheological modeling. These computational tools will give further insight into how macroscopic response results from microscopic interactions.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- May 07, 2018
- Source ID
- W911NF1810118
Entities
People
- Ken Kamrin
Organizations
- Army Contracting Command
- Massachusetts Institute of Technology
- United States Army