Adaptive Multimodal Localisation Techniques for Mobile Robots in Unstructured Environments : A Review

A Comprehensive Survey of Indoor Localization Methods Based on Computer Vision

“…Surveys dedicated to positioning solutions for autonomous robots [4,5] present in-depth information on this topic.…”

“…Computing Time and Platform Accuracy [92] ICL-NUIMdataset [138] and TUM dataset [139] 296 ms to find the most similar frame and 277 ms to estimate the final pose on Intel Xeon E5-1650 v3 CPU 3.5 GHz, NVidia TITAN GPU more than 80% of the images are localized within 2. 5 If WiFi signals, inertial sensors, beacons, or other sensors can increase the accuracy of marker based localization solutions or can help reduce the number of synthesized images that should be placed on the ceiling/floor/walls of the building (as discussed in Section 3.2.6), a hybrid approach can be even more useful when dealing with natural features from the environment. Acquiring additional information from various sensors can help reduce the search space in the image matching stages.…”

“…In recent years, the field of indoor localization has increased in popularity due to both the increasing number of applications [1] in domains such as surveillance [2], navigation (both assistive and general purpose) [3], robotics [4][5][6], and Augmented Reality (AR) [7] and the many proposed solutions that differ in terms of the devices used for tracking, the type of sensor data, and the localization algorithms.…”

“…Even though a considerable number of surveys on indoor localization have been published [1,[4][5][6][9][10][11][12][13][14][15], as this research space continuously developed and the types of localization solutions diversified, we find that for the area of vision based localization, the majority of the previous surveys could have better focus as they are too general (encompassing all kinds of sensing devices) or too specific (addressing only a segment of the vision based localization problem, such as Simultaneous Localization and Mapping (SLAM) [16][17][18][19][20][21], Structure from Motion (SfM) [22], or image matching [23]). Other surveys discuss indoor positioning solutions particular to certain application domains.…”