Terrain Modeling and Simulation of a Tumbleweed Rover Traversing Martian Rock Fields

A tumbleweed rover is a spherical wind-driven vehicle designed to negotiate harsh Martian terrain while searching for areas of scientific interest. This paper models Martian rock fields and analyzes the rover's dynamics on these terrains. Computational procedures are established for creating randomized Martian rock fields based on statistical models. Optimization techniques allow for terrain generation to coincide with the rover's motion, reducing the computational cost of exploring broad regions. Efficient collision detection routines reduce the number of tests of potential collisions that must be performed between the rover and the terrain while establishing new contact constraints. Simulations demonstrate that bouncing is the rover's dominant mode of travel through the rock fields, and Monte Carlo simulations illustrate how the rover's downrange position depends on the rover design and atmospheric conditions. The rover's capacity for long-distance travel over Martian rock fields is verified.