Masters Thesis

Optimization of LIDAR Simulations for Multi-Robot Coordination of Movement

To understand and improve the efficiency of LIDAR (Light Imaging, Detection, And Ranging) simulations used for robotic movement, an analysis of multiple variants of scanning algorithms was performed. Many challenges arose during the course of experimentation: creating a meaningful user interface, restructuring the problem into smaller segments, validating uncommon conditions, and enhancing overall performance. Time trials were observed for verified algorithms when appending multiple types of data used in a given simulation. An objective analysis was completed about the overall mathematical structure of the valid algorithms. This thesis confirms that performance for LIDAR simulations can be vastly accelerated with no loss in fidelity or precision. These findings demonstrate an exponential speed increase in LIDAR simulator execution for the average use case. This capability has the potential to improve 3D simulations as well as enable more comprehensive testing of fully autonomous mobile robots in the future.

Items