A Novel Scheme For Low-Resolution Vision-based Recognition Of Autonomous Mobile Robots

The goal of our research is to develop algorithms that using low-resolution images to perceive and understand a typical indoor environment. The algorithm must solve three problems from autonomous navigation in a low-resolution image for indoor mobile robot: 1.Autonomous localization,2.Movement-direction determining,3.Simple geometric model representation of an indoor environment. We select the rectangular corridor for the experimental environment of the algorithm.First, we present a technique for mobile robot exploration in unknown indoor environments using only a forward facing camera. The robot got its own location through visual technology. Its location needs to detect the coordinates and orientation of the current robot position. The orientation is determined by combining the results of five complementary measures:Entropy, Symmetry by mutual information, Aggregate phase, Vanishing point using self-similarity, Median of bright pixels. While the estimated distance to the end combines the results of three complementary measures:Time-to-contact, Jeffrey divergence, Entropy.Second, we present an algorithm using image entropy to detect and classify corridor junctions from low resolution images. Because entropy can be used to perceive depth, it can be used to detect an open corridor in a set of images recorded by turning a robot at a junction by 360 degrees. Our algorithm involves detecting peaks from continuously measured entropy values and determining the angular distance between the detected peaks to determine the type of junction that was recorded (L-junction, T-junction, dead-end, or cross junction).Third, we propose a simple corridor representation consisting of the orientation line and the wall-floor boundaries. Our research discuss the impact of image resolution upon the accuracy of a geometry. The experimental results show that image resolution can be decreased by 99% while the additional error less than 2%.Application of simple robot from our own designs and the algorithm, we build a simulation system for the autonomous navigation base on Low-resolution visual recognition technology of the mobile robot. The system build map by the gray image of 32 X 24 resolution detected intermittently. Experimental results prove the processing speed of the system is fast, which lay a good foundation for the navigation and enables to complete the assigned task.