Estimating optimal regions for improvement in range acquisition from a single point of view

It is well established that acquiring large amount of data can quickly lead to important data management challenges where processing capabilities become saturated and preempt full usage of the information available for autonomous systems to make educated decisions. While sub-sampling offers a naïve solution for reducing dataset dimension, it does not capitalize on the knowledge available in already acquired data to selectively drive further acquisition over the most significant regions. This paper discusses the development of a probabilistic occupancy grid based algorithm to automatically establish which regions within a single point of view from a range sensor would provide the most improvement if further acquisitions were made. The algorithm, which is independent of the sensor used, is validated with range data acquired from the popular Kinect multi-modal imaging sensor.