Design and calibration of a multi-modal range sensor using passive stereo, structured lighting, and active triangulation laser range finder

Collecting dense range measurements in uncontrolled environments is a challenging problem as lighting and surface textures significantly influence the quality of the measurements. This dependency affects the registration and data fusion processes and consequently degrades the accuracy of surface or occupancy models that are computed from the range measurements. Typical approaches to address this issue have concentrated on improving a specific type of range sensor. On the other hand, the overall quality of the sensing can also be enhanced through the development of a mechanism that combines various range sensing technologies to form a multi-modal range sensor. The resulting problem of merging datasets collected with the respective modalities can then be solved in two ways: system calibration of the multi-modal sensor or data fitting of all datasets into a single model, the latter being more widely used. The lack of multi-modal system calibration approaches is due to their complicated and lengthy nature, where individual calibration approaches must be applied to each subsystem and then applied between subsystems of the multi-modal range sensor. This thesis proposes a technique to alleviate the problems encountered in multi-modal systems calibration. Straightforward and generic guidelines for calibration are defined and applied to an in-house integrated multi-modal system built from a laser range finder, two structured lighting systems, and a stereovision system. The system's intra- and inter-calibration processes are detailed. Reconstructed renderings of datasets collected with the calibrated multi-modal range sensor, without the use of data fitting, are also presented. From these results, the potential benefits of multi-modal calibration that reduces the need for data fitting and the advantages of merging subsystem's strengths to complement other subsystem's weaknesses are put into evidence