• Visual SLAM algorithms and their application for AR, mapping, localization and wayfinding

      Theodorou, Charalambos; Velisavljevic, Vladan; Dyo, Vladimir; Nonyelu, Fredi; ; University of Bedfordshire; Briteyellow (Elsevier, 2022-08-03)
      Visual simultaneous localization and mapping (vSLAM) algorithms use device camera to estimate agent’s position and reconstruct structures in an unknown environment. As an essential part of augmented reality (AR) experience, vSLAM enhances the real-world environment through the addition of virtual objects, based on localization (location) and environment structure (mapping). From both technical and historical perspectives, this paper categorizes and summarizes some of the most recent visual SLAM algorithms proposed in research communities, while also discussing their applications in augmented reality, mapping, navigation, and localization.
    • Visual SLAM for dynamic environments based on object detection and optical flow for dynamic object removal

      Theodorou, Charalambos; Velisavljevic, Vladan; Dyo, Vladimir; University of Bedfordshire; Briteyellow Ltd (MDPI, 2022-10-03)
      In dynamic indoor environments and for a Visual Simultaneous Localization and Mapping (vSLAM) system to operate, moving objects should be considered because they could affect the system’s visual odometer stability and it is position estimation accuracy. vSLAM can use feature points or a sequence of images as it is only source of input in order to perform localization while simultaneously creating a map of the environment. A vSLAM system based on ORB-SLAM3 and on YOLOR was proposed in this paper. The newly proposed system in combination with an object detection model (YOLOX) applied on extracted feature points is capable of achieving 2-4% better accuracy as compared to VPS-SLAM and DS-SLAM. Static feature points such as signs and benches were used to calculate the camera position and dynamic moving objects were eliminated by using the tracking thread. A specific custom personal dataset that includes indoor and outdoor RGB-D pictures of train stations including dynamic objects and high density of people, ground truth data, sequence data, video recording with the train stations and X, Y, Z data was used to validate and evaluate the proposed method. The results show that ORB-SLAM3 with YOLOR as object detection achieves 89.54% of accuracy in dynamic indoor environments compared to previous systems such as VPS-SLAM.
    • Wireless magnetic sensor network for road traffic monitoring and vehicle classification

      Velisavljević, Vladan; Cano, Eduardo; Dyo, Vladimir; Allen, Ben; University of Bedfordshire; European Commission, Joint Research Centre; University of Oxford (De Gruyter Open, 2016-11-23)
      Efficiency of transportation of people and goods is playing a vital role in economic growth. A key component for enabling effective planning of transportation networks is the deployment and operation of autonomous monitoring and traffic analysis tools. For that reason, such systems have been developed to register and classify road traffic usage. In this paper, we propose a novel system for road traffic monitoring and classification based on highly energy efficient wireless magnetic sensor networks. We develop novel algorithms for vehicle speed and length estimation and vehicle classification that use multiple magnetic sensors. We also demonstrate that, using such a low-cost system with simplified installation and maintenance compared to current solutions, it is possible to achieve highly accurate estimation and a high rate of positive vehicle classification.
    • Wrinkle measurement in glass-carbon hybrid laminates comparing ultrasonic techniques: a case study

      Larrañaga-Valsero, Beatriz; Smith, Robert A.; Tayong-Boumda, Rostand; Fernández-López, Antonio; Güemes, Alfredo; Universidad Politécnica de Madrid; University of Bristol (Elsevier Ltd, 2018-08-15)
      Wrinkles, (also known as out-of-plane waviness) are, unfortunately, a common phenomenon that has caused some wind-turbine blades to unexpectedly fail in service. Being able to detect the wrinkles while in the factory will reduce the risk of catastrophic failure and characterising the wrinkles would minimise the repaired area, thus increasing the efficiency of the repair and the design. This work compares the effectiveness of three different ultrasound techniques for detecting and characterising out-of-plane wrinkles in the typical glass-carbon hybrid laminates that are used for wind-turbine blades. The tests samples were manufactured so that the laminates and the defects are representative of those used in the wind-turbine industry. Basic mechanical tests were performed to check the drop in mechanical properties due to wrinkling. The ideal probe frequency was determined as the resonance frequency of the plies using an analytical ultrasonic-propagation model. The three different ultrasound techniques used are: full-matrix capture (FMC) with the total focusing method (TFM), a commercial phased-array instrument and an immersion test with a raster-scanned single-element focused probe. When possible, severity parameters of the wrinkle were measured on the ultrasonic images and compared with the measurements of the actual sample in order to determine which method best characterises such wrinkles and which would be more appropriate to implement in an industrial environment. Not all of the techniques allowed full characterisation of out-of-plane waviness on the specimens. The FMC/TFM method gave better results whilst phased-array technology and single-element immersion testing presented more challenges. An additional enhancement to the TFM imaging was achieved using an Adapted-TFM method with an angle-dependent velocity correction.