Mobile robots can significantly impact industrial applications by improving safety and productivity on dangerous or highly repetitive tasks. In areas such as Oil and Gas production, a key industry for Qatar, there are many tasks that are hazardous and/or require repetitive tedious inspection. For a mobile robot to perform these tasks it must be able to reliably and robustly localize itself while building maps of its operating environment. Vision sensors (cameras) equipped on a mobile robot can be used to simultaneously build a map of an environment, and use this map for localization. This process is referred to as SLAM, and can complement alternate localization methods. Examples include active beacon systems, which require additional and costly fixed infrastructure, and GPS which can be unreliable due to multi-path propagation and occlusions. This research would focus on the development of a visual SLAM system which uses an array of perspective stereo and omni-directional cameras. Each of these cameras has advantages when used as part of a visual SLAM system, and their optimized coupling has the potential to produce a reliable and robust system. The outcome of this research will be the development of such a system, suitable for use in industrial applications relevant to Qatar, which is validated using systematic experiments. My research goal is to advance the state of the art in visual SLAM, and this grant will enable me to make significant progress towards this goal.