Outdoor mobile robot applications generally implement Global Positioning Systems (GPS) for localization tasks. However, GPS accuracy in outdoor localization has less accuracy in different environmental conditions. This paper presents two outdoor localization methods based on deep learning and landmark detection. The first localization method is based on the Faster Regional-Convolutional Neural Network (Faster R-CNN) landmark detection in the captured image. Then, a feedforward neural network (FFNN) is trained to determine robot location coordinates and compass orientation from detected landmarks. The second localization employs a single convolutional neural network (CNN) to determine location and compass orientation from the whole image. The dataset consists of images, geolocation data and labeled bounding boxes to train and test two proposed localization methods. Results are illustrated with absolute errors from the comparisons between localization results and reference geolocation data in the dataset. The experimental results pointed both presented localization methods to be promising alternatives to GPS for outdoor localization.

No abstract

Robot localization is an important task for mobile robot navigation. There are many methods focused on this issue. Some methods are implemented in indoor and outdoor environments. However, robot localization in textureless environments is still a challenging task. This is because in these environments, the scene appears the same in almost every position. In this work, we propose a method that can localize robots in textureless environments. We use Histogram of Oriented Gradients (HOG) and Speeded Up Robust Feature (SURF) descriptors together with Depth information to form a Depth-HOG-SURF multifeature descriptor, which is later used for image matching. K-means clustering is applied to partition the whole feature into groups that are collectively called visual vocabulary. All the images in the database are encoded using the vocabulary. The experimental results show a good performance of the proposed method.

No abstract

No abstract