Path planning algorithms in rough terrain environments

Abstract

Path planning is concerned with finding an optimal path from a starting point to a destination. This process involves using different types of algorithms. Path planning in rough terrain environments poses a unique challenge in implementing these algorithms. In recent times, autonomous vehicles have become more common, yet they have been mostly constrained to on-road and indoor environments. In this thesis the process of path planning, path planning algorithms and recent research about implementing these algorithms into rough terrain environments are reviewed. The difference between conventional and rough terrain environments is that the latter is very heterogeneous. More conventional graph search and sampling-based algorithms such as A* and RRT can be well adapted to the situation, yet the novel machine learning approach offers versatility.