System Architecture, Calibration, and Control for LiDAR Systems Onboard Unmanned Vehicles
The development of Unmanned Aerial Vehicles (UAVs) as a mobile platform for deploying portable systems has benefited several applications in the civilian and military fields in the past few decades. Also, the parallel advances in peripheral technology such as the enhancement of GNSS/INS modules have resulted in a remarkable development of both UAVs as well as UAVs applications. Such development leads to the establishment of a system architecture for UAV design that will be the basis of all discussion in this dissertation. Centrally, this dissertation introduces a generic framework for UAVs equipped with a GNSS/INS positioning and orientation module as well as low-cost LiDAR sensors for targeted mapping and monitoring applications. An essential aspect of this research proposes a LiDAR system calibration procedure for a mobile airborne platform. Such a calibration procedure can directly estimate the mounting parameters relating the laser scanners to the onboard GNSS/INS unit, i.e., the lever-arm and boresight angles for a LiDAR unit through an outdoor calibration procedure. This approach is based on the use of conjugate planar/linear features in overlapping point clouds derived from different flight lines. Furthermore, an optimal configuration of target primitives and flight lines is determined by analyzing the potential impact of bias in mounting parameters of a LiDAR unit. To add a degree of autonomy to this integrated framework, a Coverage Path Planning (CPP) approach is proposed. Such approach is performed to achieve complete coverage of the area of interest in a minimum time with the aid of real-time Simultaneous Localization and Mapping (SLAM) technique. The successful implementation of SLAM with this integrated framework furthermore offers insight into extending the system in conditions where one of the subsystems may not function properly. For example, in GNSS denied environments, the GNSS/INS modules fail to work correctly due to the absence of consistent GNSS signals. This dissertation introduces a Pseudo-GNSS/INS integrated framework that is implemented using probabilistic SLAM techniques. Such a framework allows for the extension of the operation of such systems for GNSS-denied environments and hence is a significant contribution towards increasing robustness and autonomy in terrestrial/aerial mapping systems.
Habib, Purdue University.
Off-Campus Purdue Users:
To access this dissertation, please log in to our