A service robot that operates in a previously-unseen home environment should be able to recognize the functionality of the rooms it visits, such as a living room, a bathroom, etc. We present a novel part-based model and an approach for room categorization using data obtained from a visual sensor. The proposed approach uses a convolutional neural network (CNN) and is robust to image degradation, such as occlusions, modifications of image scaling, and aspect changes.

The next generation service robots are expected to co-exist with humans in their homes. Such a mobile robot requires an efficient representation of space, which should be compact and expressive, for effective operation in real-world environments. We present the category-independent lower layers of a novel compositional hierarchical representation of space that is based on 2D ground-plan-like laser-range-data. The effectiveness of the model is demonstrated in the context of room categorization problem.

A model of the category-specific higher layer of a range-data-based hierarchical representation of space is presented. A method for discriminative exemplar learning based on pair-wise part similarities is introduced and applied for part dictionary selection. The method is general and can easily be applied to other modalities. Furthermore, an approach for range-data-based room categorization using the category-specific parts is proposed..

We developed algorithms and representations for autonomous mapping and visual localization of mobile robots from omni-directional images. We also developed a purely emergent hierarchical mapping algorithm based on “recover and select” that creates maps of wide and unstructured environments from local appearance and odometry.