Search for: All records

This paper presents a design approach for rigid wheels operating in highly variable, deformable terrain to improve the mobility, reliability, and efficiency of an autonomous vehicle driving on snow. The longstanding Bekker-Wong theory of terramechanics is used as the basis for the design changes with the wide range of terrain parameters for snow serving as inputs to the models and bounds for the problem. Modifications to the wheel width and diameter are evaluated based on their impacts to the rover as a system, with their effects on torque and drawbar pull being weighed against the resultant modifications in component sizing, rover weight, and energy use. Other factors, not included in the Bekker-Wong models but studied in single-wheel testbed experiments, such as bulldozing resistance and the observed dynamic effects of slip-sinkage, were also considered in the design decisions for the new wheel. Finally, to test these theories and assess the mobility improvements of the new design in situ, a four-wheeled rover, FrostyBoy, was developed for the new wheels and trialed in unmodified snow. While qualitatively showing an improvement in mobility on the Greenland ice sheet, the tests also uncovered dynamic modes of immobilization, in low cohesion, low stiffness snow that are not accounted for in terramechanics theory and require further investigation for trafficability to be maintained in all snow conditions. 

This paper presents mobility modes and control methods for the SnoWorm, a passively-articulated multi-segment autonomous wheeled vehicle concept for use in Earth’s polar regions. SnoWorm is based on FrostyBoy, a four-wheeled GPS guided rover built for autonomous surveys across ice sheets. Data collected from FrostyBoy were used to ground-truth a ROS/Gazebo model of vehicle-terrain interaction for simulations on snow surfaces. The first mobility mode, inchworm movement, uses active prismatic joints that link the SnoWorm’s segments, and allow them to push and pull one another. This pushing and pulling of individual segments can be coordinated to allow forward motion through terrain that would immobilize a single-segment vehicle. The second mobility mode utilizes fixed links between SnoWorm’s segments and uses the tension or compression measured in these links as a variable to control wheel speeds and achieve a targeted force distributions within the multi-segment vehicle. This ability to control force distribution can be used to distribute a towed load evenly across the entire SnoWorm. Alternatively, the proportion of the load carried by individual segments can be increased or decreased as needed based on each segment’s available drawbar pull or wheel slip. 

This paper presents a class of four-wheel drive autonomous robots designed to collaboratively traverse terrains with a deformable upper layer, where soil properties result in limited traction and have the potential to cause immobilization. The robots are designed to have front and rear axle yaw degrees of freedom, and front and rear axle roll degrees of freedom providing ground compliance and maneuverability on friable terrain. These degrees of freedom, along with four individually driven wheels and an actuated translational degree of freedom inside a mid-frame joint, enable poses and modes of mobility that differ significantly from a rigid vehicle. A primary goal of this work is to assess the capacity to use this vehicular form as a testbed that leverages these vehicle dynamics to assess mobility. Using a custom ROS-Gazebo simulation environment, a heterogenous driving surface is created and used to evaluate this capability. We show that the vehicle can sense imbalanced terrain resistances proprioceptively. Additionally, we show that rigidity of the vehicle can be controlled through a simple feedback control loop governing the robot’s unconstrained axles to maintain a proper heading angle and still can provide an avenue to monitor the dynamics related to full-vehicle immobilization.