Browsing by Author "Kinnari, Jouko"

Now showing 1 - 8 of 8
  • No Thumbnail Available
    Item
    3D-rekonstruktiomenetelmät GNSS-riippumattomassa autonomisten miehittämättömien ilma-alusten paikannuksessa
    (2022-05-14) Karhunen, Aleksi; Kinnari, Jouko; Sähkötekniikan korkeakoulu; Forsman, Pekka
  • Thumbnail Image
    Item
    Data Generation for Visual Localization of Unmanned Aerial Vehicles
    (2022-06-13) Jussmäki, Sakke; Kinnari, Jouko; Verdoja, Francesco; Sähkötekniikan korkeakoulu; Kyrki, Ville
    Visual localization methods for unmanned aerial vehicles (UAV) aim to provide localization in situations where satellite-based methods are not available. Many visual localization methods rely on machine learning algorithms that require significant amounts of data. Collecting a sufficient amount of real-world data is expensive and time-consuming. An option to real data is to use a simulated environment for generating data. To study if the generated data can be used to increase localization performance, a data generation pipeline in Unreal Engine 4 is implemented in this thesis. To measure the changes in performance, a neural network based localization model is used. The effect of using different ratios of generated data to real data and the effect of different types of variance in the generated data is evaluated. The results show that introducing generated data to the model improves performance with some real test sets. However, seasonal variance remains an issue.
  • No Thumbnail Available
    Item
    Design of a measurement device and signal processing methods for automated wire rope condition monitoring
    (2010) Kinnari, Jouko; Sunio, Juha; Elektroniikan, tietoliikenteen ja automaation tiedekunta; Sähkötekniikan korkeakoulu; School of Electrical Engineering; Koivo, Heikki
  • Thumbnail Image
    Item
    Infrastructureless unmanned aerial vehicle localization
    (Aalto University, 2024) Kinnari, Jouko; Verdoja, Francesco, Academy Research Fellow, Aalto University, Department of Electrical Engineering and Automation, Finland; Sähkötekniikan ja automaation laitos; Department of Electrical Engineering and Automation; Intelligent Robotics; Sähkötekniikan korkeakoulu; School of Electrical Engineering; Kyrki, Ville, Prof., Aalto University, Department of Electrical Engineering and Automation, Finland
    The ability to localize, i.e., determine the position and orientation of a Unmanned Aerial Vehicle (UAV) with respect to a known frame of reference, is a basic requirement for autonomous flight. Common solutions for providing a UAV with localization ability have relied on the availability of an infrastructure built for this purpose, usually based on an arrangement of radio emitters, predominantly Global Navigation Satellite Systems (GNSSs). However, disruptions in the radio signal path, as well as actions taken by an adversary, such as spoofing and jamming, may hinder localization accuracy. This thesis focuses on UAV localization, in environments lacking infrastructure for that purpose, specifically using a low-size, weight and power (SWaP) sensor system consisting of a camera, an Inertial Measurement Unit (IMU), and a magnetometer. The challenges limiting this approach are associated with the difficulty of relating UAV environment measurements to a map, due to not only differences between the appearance of the map representation and the environment as observed using onboard sensors, but also natural ambiguities such as perceptual aliasing. This thesis addresses three specific problem areas and demonstrates a full localization solution running in real time on a small UAV. First, the thesis addresses the problem of how to perform localization with respect to an orthophoto map using a camera whose orientation is not strictly vertical. A method is presented for allowing variation in camera view direction by orthoprojecting camera images to a top-down view based on a planar assumption of the ground under the UAV. This would be an adequate assumption when flying over relatively flat terrain, as demonstrated through experimentation on real data. Second, this thesis addresses the problem of seasonal appearance change, where we learn a function for assessing the correspondence between an image acquired by an UAV and an orthophoto map by proposing a method that is tolerant to seasonal appearance change in the operating environment. The proposed method exceeds the state-of-the-art in the literature both in terms of the time to convergence and localization error. Third, this work addresses the wake-up robot problem. For this purpose, an approach is presented for learning a model to extract a compact descriptor vector representation from both a UAV image and from a map, thus enabling very fast confirmation or rejection of pose hypotheses, which allows localization to occur over large areas without knowledge of the initial pose. The presented method alleviates the computational challenges inherent in the problem of localization over a large area with an unknown prior starting position and orientation. The method also enables operation of a small UAV on a map covering an area of 100 square kilometers without requiring knowledge of the initial pose while tolerating seasonal appearance change and resolving ambiguities due to perceptual aliasing. Finally, the operation of the algorithm developed for the wake-up robot problem running on a small UAV is demonstrated in real time using real data. The thesis concludes by characterizing a number of open issues related to the problem domain.
  • Thumbnail Image
    Item
    LSVL: Large-scale season-invariant visual localization for UAVs
    (Elsevier Science, 2023-10) Kinnari, Jouko; Renzulli, Riccardo; Verdoja, Francesco; Kyrki, Ville; Department of Electrical Engineering and Automation; Intelligent Robotics
    Localization of autonomous unmanned aerial vehicles (UAVs) relies heavily on Global Navigation Satellite Systems (GNSS), which are susceptible to interference. Especially in security applications, robust localization algorithms independent of GNSS are needed to provide dependable operations of autonomous UAVs also in interfered conditions. Typical non-GNSS visual localization approaches rely on known starting pose, work only on a small-sized map, or require known flight paths before a mission starts. We consider the problem of localization with no information on initial pose or planned flight path. We propose a solution for global visual localization on large maps, based on matching orthoprojected UAV images to satellite imagery using learned season-invariant descriptors, and test with environment sizes up to 100 km2. We show that the method is able to determine heading, latitude and longitude of the UAV at 12.6–18.7 m lateral translation error in as few as 23.2–44.4 updates from an uninformed initialization, also in situations of significant seasonal appearance difference (winter–summer) between the UAV image and the map. We evaluate the characteristics of multiple neural network architectures for generating the descriptors, and likelihood estimation methods that are able to provide fast convergence and low localization error. We also evaluate the operation of the algorithm using real UAV data and evaluate running time on a real-time embedded platform. We believe this is the first work that is able to recover the pose of an UAV at this scale and rate of convergence, while allowing significant seasonal difference between camera observations and map.
  • No Thumbnail Available
    Item
    Millimeter Wave Multi-radar Testbed
    (2020-01-20) Hussaini, Syeda; Kinnari, Jouko; Sähkötekniikan korkeakoulu; Ruttik, Kalle
    The research in the 5G network and millimeter-wave technology deployment for automotive and industrial applications has rapidly increased due to the large scale availability of compact high-resolution single-chip radar solutions. In this thesis, multiple radar sensor modules are utilized for generating the raw ADC data. This thesis work develops an application to acquire the captured data utilizing the automotive and industrial millimeter-wave sensing evaluation modules. The radar sensors operating at millimeter-wave frequency spectrum comprises of the Digital Signal Processor (DSP) and hardware accelerator (HWA) to perform real-time signal processing algorithm and stream the processed output of detected object. However, the thesis aims to stream the raw ADC data output and perform the Fast Fourier Transform (FFT) computations without the utilization of DSP or HWA for signal processing. The assessment of the captured raw ADC data will be beneficial for developers to modify the design of the digital signal processing chain as per the application requirements. Initially, the study commences with a detailed description of the FMCW radar theory and signal-processing methods, crucial to understand the functioning of the digital front end of the device and techniques to detect objects. Furthermore, the implementation of raw ADC data transfer from the internal memory of both the sensing modules is executed using the Texas Instruments equipment, software tools and development kit. Hence, the post-processing of the accumulated data is performed to estimate the range and visualize the complex ADC signal and detected objects peaks in the stationary scenario.
  • No Thumbnail Available
    Item
    Radar Based Simultaneous Localization and Mapping for UAV use
    (2023-05-25) Lindell, Mikael; Kinnari, Jouko; Sähkötekniikan korkeakoulu; Forsman, Pekka
  • Thumbnail Image
    Item
    Season-Invariant GNSS-Denied Visual Localization for UAVs
    (IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC, 2022-10-01) Kinnari, Jouko; Verdoja, Francesco; Kyrki, Ville; Department of Electrical Engineering and Automation; Intelligent Robotics
    Localization without Global Navigation Satellite Systems (GNSS) is a critical functionality in autonomous operations of unmanned aerial vehicles (UAVs). Vision-based localization on a known map can be an effective solution, but it is burdened by two main problems: places have different appearance depending on weather and season, and the perspective discrepancy between the UAV camera image and the map make matching hard. In this letter, we propose a localization solution relying on matching of UAV camera images to georeferenced orthophotos with a trained convolutional neural network model that is invariant to significant seasonal appearance difference (winter-summer) between the camera image and map. We compare the convergence speed and localization accuracy of our solution to six reference methods. The results show major improvements with respect to reference methods, especially under high seasonal variation. We finally demonstrate the ability of the method to successfully localize a real UAV, showing that the proposed method is robust to perspective changes.