Accurate perception is a principal challenge of autonomous off-road driving. Perceptive technologies generally focus on obstacle avoidance. However, at high speed, terrain roughness is also important to control shock the vehicle experiences. The accuracy required to detect rough terrain is significantly greater than that necessary for obstacle avoidance. We present a self-supervised machine lea...

This report investigates the application of two Computer Vision techniques for characterizing critical regions in digital terrain maps: Skeletonization and Topographical Region Detection. The Navigation Path Planning problem involves nding the optimal path through a digital terrain Map and must minimize a cost function while satisfying the constraints of navigation. This process is computationa...

A map of extant slope failures is the most basic element of any landslide assessment. Without an accurate inventory of slope instability, it is not possible to analyze the controls on the spatial and temporal patterns of mass movement or the environmental, human, or geomorphic consequences of slides. Landslide inventory maps are tedious to compile, difficult to make in vegetated terrain using c...

The U.S. Anny currently uses the root mean square of elevation and power spectral density to characterize road/terrain (off-road) roughness for durability. This paper describes research aimed toward improving these metrics. One method currently under consideration involves running a relatively simple, yet vehicle class specific, model over a given terrain and using predicted vehicle response(s)...

The generation of a topographical surface or digital elevation model for given set points in space is known problem civil engineering and topography. In this article, we propose simple efficient way to obtain the terrain by using structural shell finite element model, giving advice on how implement it. proposed methodology does not need large number define terrain, so it especially suitable be ...

A stereo-based terrain classification for traversability estimation of all terrains in offroad mobile robots is presented. The proposed method defines the roughness of the surrounding terrain for every single pixel in the image or point in a point cloud using surface normals and also explains how this is applied to all terrain by knowing the kinematics capability, mechanical constraints and siz...

The ability to measure terrain roughness is essential for safe and e cient navigation of outdoor mobile robots. This paper describes a prototype system consisting of a colour camera and a laser stripe generator for the measurement of ground surface roughness. With outdoor mobile robotic applications in mind, the system provides a cheap, fast and robust alternative to other methods of roughness ...