Efficient Selection of Disambiguating Actions for Stereo Vision
This work addresses the challenge of active sensing for mobile robots in low-texture environments, offering an incremental improvement in computational efficiency.
The paper tackles the problem of efficiently selecting disambiguating actions for stereo vision in robotics, where scenes lack texture, by proposing a method that reduces the time complexity for selecting the optimal laser aim point from O(nd^2) to O(nd), matching the stereo algorithm's cost.
In many domains that involve the use of sensors, such as robotics or sensor networks, there are opportunities to use some form of active sensing to disambiguate data from noisy or unreliable sensors. These disambiguating actions typically take time and expend energy. One way to choose the next disambiguating action is to select the action with the greatest expected entropy reduction, or information gain. In this work, we consider active sensing in aid of stereo vision for robotics. Stereo vision is a powerful sensing technique for mobile robots, but it can fail in scenes that lack strong texture. In such cases, a structured light source, such as vertical laser line can be used for disambiguation. By treating the stereo matching problem as a specially structured HMM-like graphical model, we demonstrate that for a scan line with n columns and maximum stereo disparity d, the entropy minimizing aim point for the laser can be selected in O(nd) time - cost no greater than the stereo algorithm itself. In contrast, a typical HMM formulation would suggest at least O(nd^2) time for the entropy calculation alone.