Object goal navigation (ObjectNav) is the task of finding a target object in an unseen environment. It is one of the fundamental challenges in visual navigation as it requires both structural and semantic understanding. In this letter, we present OVG-Nav, a novel ObjectNav framework that leverages a topological graph structure called object value graph (OVG), which contains visual observations and commonsense prior knowledge. The high-level planning of OVG-Nav prioritizes subgoal nodes for exploration based on a metric called object value , which reflects the closeness to the target object. Here, we propose OVGNet, a model designed to predict the object values of each node of an OVG using observed features along with commonsense knowledge. The structure of high-level planning using OVG and low-level action decisions reduces sensitivity to accumulating sensor noises, leading to robust navigation performance. Experimental results show that OVG-Nav outperforms the baseline in success rate (SR) and success rate weighted by path length (SPL) in the MP3D dataset both in accurate sensing and noisy sensing. In addition, we show that the OVG-Nav can be transferred to the real-world robot successfully.

Compound eyes, also known as insect eyes, have a unique structure. They have a hemispheric surface, and a lot of single eyes are deployed regularly on the surface. Thanks to this unique form, using the compound images has several advantages, such as a large field of view (FOV) with low aberrations. We can exploit these benefits in high-level vision applications, such as object recognition, or semantic segmentation for a moving robot, by emulating the compound images that describe the captured scenes from compound eye cameras. In this paper, to the best of our knowledge, we propose the first convolutional neural network (CNN)-based ego-motion classification algorithm designed for the compound eye structure. To achieve this, we introduce a voting-based approach that fully utilizes one of the unique features of compound images, specifically, the compound images consist of a lot of single eye images. The proposed method classifies a number of local motions by CNN, and these local classifications which represent the motions of each single eye image, are aggregated to the final classification by a voting procedure. For the experiments, we collected a new dataset for compound eye camera ego-motion classification which contains scenes of the inside and outside of a certain building. The samples of the proposed dataset consist of two consequent emulated compound images and the corresponding ego-motion class. The experimental results show that the proposed method has achieved the classification accuracy of 85.0%, which is superior compared to the baselines on the proposed dataset. Also, the proposed model is light-weight compared to the conventional CNN-based image recognition algorithms such as AlexNet, ResNet50, and MobileNetV2.