Project

Underground Machine Vision

Camera-based positioning and environmental sensing for underground vehicles where satellite navigation is unavailable and conditions are difficult.

Machine vision can provide vehicle motion and environmental information without relying on expensive, data-intensive sensors. Adam Neethling’s visual-odometry work targets robust camera-based positioning with uncertainty-aware drift correction. Franco Prins’s road-condition topic extends the sensing question to the surface on which machines operate.

Position estimates can become productivity indicators—cycle time, harsh braking and slow corners—as well as inputs to future autonomous systems.