Στόχος της παρούσας διπλωματικής εργασίας είναι η ανάπτυξη ολοκληρωμένου συστήματος οδομετρίας που θα υπολογίζει την απόσταση μετακίνησης ενός κινητού μέσου με χρήση τεχνικών όρων της όρασης των υπολογιστών. Στα κλασικά μετρητικά συστήματα εμφανίζονται σημαντικές αποκλίσεις μεταξύ της πραγματικής και της υπολογισθείσας θέσης του ρομπότ. Σκοπός της ολοκληρωμένης διάταξης οδομετρίας που θα κατασκευαστεί είναι ο περιορισμός των σφαλμάτων αυτών. / Aim of this diploma thesis is the development of a dead-reackoning (odometry) system through a hollistic approach, in order to calculate the distance travelled by a mobile system (robot) by means of computer vision. In traditional systems there exist important deviations between the real and the calculated positions. Goal of the current work is to limit (minimize) the aforementioned deviations.

Recent studies have demonstrated that images constructed from lidar reflectance information exhibit superior robustness to lighting changes. However, due to the scanning nature of the lidar and assumptions made in previous implementations, data acquired during continuous vehicle motion suffer from geometric motion distortion and can subsequently result in poor metric visual odometry (VO) estimates, even over short distances (e.g., 5-10 m). The first part of this thesis revisits the measurement timing assumption made in previous systems, and proposes a frame-to-frame VO estimation framework based on a pose-interpolation scheme that explicitly accounts for the exact acquisition time of each intrinsic, geometric feature measurement. The second part of this thesis investigates a novel method of lidar calibration that can be applied without consideration of the internal structure of the sensor. Both methods are validated using experimental data collected from a planetary analogue environment with a real scanning laser rangefinder.

Recent studies have demonstrated that images constructed from lidar reflectance information exhibit superior robustness to lighting changes. However, due to the scanning nature of the lidar and assumptions made in previous implementations, data acquired during continuous vehicle motion suffer from geometric motion distortion and can subsequently result in poor metric visual odometry (VO) estimates, even over short distances (e.g., 5-10 m). The first part of this thesis revisits the measurement timing assumption made in previous systems, and proposes a frame-to-frame VO estimation framework based on a pose-interpolation scheme that explicitly accounts for the exact acquisition time of each intrinsic, geometric feature measurement. The second part of this thesis investigates a novel method of lidar calibration that can be applied without consideration of the internal structure of the sensor. Both methods are validated using experimental data collected from a planetary analogue environment with a real scanning laser rangefinder.

A classic Computer Vision task is the estimation of a 3D map from a collection of images. This thesis explores the online simultaneous estimation of camera poses and map points, often called Visual Simultaneous Localisation and Mapping [VSLAM]. In the near future the use of visual information by autonomous cars is likely, since driving is a vision dominated process. For example, VSLAM could be used to estimate the position of the car in relation to objects of interest, such as the road, other cars and pedestrians. Aimed at the creation of a real-time, robust, loop closing, single camera SLAM system, the properties of several state-of-the-art VSLAM systems and related techniques are studied. The system goals cover several important, if difficult, problems, which makes a solution widely applicable. This thesis makes two contributions: A rigorous qualitative analysis of VSLAM methods and a system designed accordingly. A novel tracking by matching scheme is proposed, which, unlike the trackers used by many similar systems, is able to deal better with forward camera motion. The system estimates general motion with loop closure in real time. The system is compared to a state-of-the-art monocular VSLAM algorithm and found to be similar in speed and performance.

This work presents a novel technique for lifelong localization of robots. It performs a tight fusion of GPS and Multi-State Constraint Kalman Filter, a visual-inertial odometry method for robot localization. It is shown in exper- iments that the proposed algorithm achieves better position accuracy than either GPS and Multi-State Constraint Kalman Filter alone. Additionally, the experiments demonstrate that the algorithm is able to reliably operate when the GPS signal is highly corrupted by noise or even in presence of substantial GPS outages. 1

This thesis deals with algorithms of visual odometry and its application on the experimental vehicle Car4. The first part contains different researches in this area on which the solution process is based. Next chapters introduce theoretical design and ideas of monocular and stereo visual odometry algorithms. The third part deals with the implementation in the software MATLAB with the use of Image processing toolbox. After tests done and based on real data, the chosen algorithm is applied to the vehicle Car4 used in practical conditions of interior and exterior. The last part summarizes the results of the work and address the problems which are asociated with the application of visual obmetry algorithms.

Visual(-Inertial) Odometry (VIO) and Simultaneous Localization and Mapping (SLAM) are hot topics in Computer Vision today. These technologies have various applications, including robotics, autonomous driving, and virtual reality. They may also be valuable in studying human behavior and navigation through head-mounted visual systems. A complication to SLAM and VIO systems could potentially be visual degeneration such as motion blur. This thesis attempts to evaluate the robustness to motion blur of two open-source state-of-the-art VIO and SLAM systems, namely Delayed Marginalization Visual-Inertial Odometry (DM-VIO) and ORB-SLAM3. There are no real-world benchmark datasets with varying amounts of motion blur today. Instead, a semi-synthetic dataset was created with a dynamic trajectory-based motion blurring technique on an existing dataset, TUM VI. The systems were evaluated in two sensor configurations, Monocular and Monocular-Inertial. The systems are evaluated using the Root Mean Square (RMS) of the Absolute Trajectory Error (ATE).  Based on the findings, the visual input highly influences DM-VIO, and performance decreases substantially as motion blur increases, regardless of the sensor configuration. In the Monocular setup, the performance decline significantly going from centimeter precision to decimeter. The performance is slightly improved using the Monocular-Inertial configuration. ORB-SLAM3 is unaffected by motion blur performing on centimeter precision, and there is no significant difference between the sensor configurations. Nevertheless, a stochastic behavior can be noted in ORB-SLAM3 that can cause some sequences to deviate from this. In total, ORB-SLAM3 outperforms DM-VIO on the all sequences in the semi-synthetic datasets created for this thesis. The code used in this thesis is available at GitHub https://github.com/cmannila along with forked repositories of DM-VIO and ORB-SLAM3 / Visual(-Inertial) Odometry (VIO) och Simultaneous Localization and Mapping (SLAM) är av stort intresse inom datorseende (Computer Vision). Dessa system har en variation av tillämpningar såsom robotik, själv-körande bilar och VR (Virtual Reality). En ytterligare potentiell tillämpning är att integrera SLAM/VIO i huvudmonterade system, såsom glasögon, för att kunna studera beteenden och navigering hos bäraren. En komplikation till SLAM och VIO skulle kunna vara en visuell degration i det visuella systemet såsom rörelseoskärpa. Detta examensarbete försöker utvärdera robustheten mot rörelseoskärpa i två tillgängliga state-of-the-art system, DM-VIO (Delayed Marginalization Visual-Inertial Odometry) och ORB-SLAM3. Idag finns det inga tillgängliga dataset som innehåller specifikt varierande mängder rörelseoskärpa. Således, skapades ett semisyntetiskt dataset baserat på ett redan existerande, vid namn TUM VI. Detta gjordes med en dynamisk rendering av rörelseoskärpa enligt en känd rörelsebana erhållen från datasetet. Med denna teknik kunde olika mängder exponeringstid simuleras.  DM-VIO och ORB-SLAM3 utvärderades med två sensor konfigurationer, Monocular (en kamera) och Monokulär-Inertial (en kamera med Inertial Measurement Unit). Det objektiva mått som användes för att jämföra systemen var Root Mean Square av Absolute Trajectory Error i meter. Resultaten i detta arbete visar på att DM-VIO är i hög-grad beroende av den visuella signalen som används, och prestandan minskar avsevärt när rörelseoskärpan ökar, oavsett sensorkonfiguration. När enbart en kamera (Monocular) används minskar prestandan från centimeterprecision till diameter. ORB-SLAM3 påverkas inte av rörelseoskärpa och presterar med centimeterprecision för alla sekvenser. Det kan heller inte påvisas någon signifikant skillnad mellan sensorkonfigurationerna. Trots detta kan ett stokastiskt beteende i ORB-SLAM3 noteras, detta kan ha orsakat vissa sekvenser att bete sig avvikande. I helhet, ORB-SLAM3 överträffar DM-VIO på alla sekvenser i det semisyntetiska datasetet som skapats för detta arbete. Koden som använts i detta arbete finns tillgängligt på GitHub https://github.com/cmannila tillsammans med forkade repository för DM-VIO och ORB-SLAM3.

This thesis investigates techniques and designs an autonomous visual stereo based navigation sensor to improve stereo visual odometry for purpose of navigation in unknown environments. In particular, autonomous navigation in a space mission context which imposes challenging constraints on algorithm development and hardware requirements. For instance, Global Positioning System (GPS) is not available in this context. Thus, a solution for navigation cannot rely on similar external sources of information. Support to handle this problem is required with the conception of an intelligent perception-sensing device that provides precise outputs related to absolute and relative 6 degrees of freedom (DOF) positioning. This is achieved using only images from stereo calibrated cameras possibly coupled with an inertial measurement unit (IMU) while fulfilling real time processing requirements. Moreover, no prior knowledge about the environment is assumed. Robotic navigation has been the motivating research to investigate different and complementary areas such as stereovision, visual motion estimation, optimisation and data fusion. Several contributions have been made in these areas. Firstly, an efficient feature detection, stereo matching and feature tracking strategy based on Kanade-Lucas-Tomasi (KLT) feature tracker is proposed to form the base of the visual motion estimation. Secondly, in order to cope with extreme illumination changes, High dynamic range (HDR) imaging solution is investigated and a comparative assessment of feature tracking performance is conducted. Thirdly, a two views local bundle adjustment scheme based on trust region minimisation is proposed for precise visual motion estimation. Fourthly, a novel KLT feature tracker using IMU information is integrated into the visual odometry pipeline. Finally, a smart standalone stereo visual/IMU navigation sensor has been designed integrating an innovative combination of hardware as well as the novel software solutions proposed above. As a result of a balanced combination of hardware and software implementation, we achieved 5fps frame rate processing up to 750 initials features at a resolution of 1280x960. This is the highest reached resolution in real time for visual odometry applications to our knowledge. In addition visual odometry accuracy of our algorithm achieves the state of the art with less than 1% relative error in the estimated trajectories.

In this research, image segmentation and visual odometry estimations in real time are addressed, and two main contributions were made to this field. First, a new image segmentation and classification algorithm named DilatedU-NET is introduced. This deep learning based algorithm is able to process seven frames per-second and achieves over 84% accuracy using the Cityscapes dataset. Secondly, a new method to estimate visual odometry is introduced. Using the KITTI benchmark dataset as a baseline, the visual odometry error was more significant than could be accurately measured. However, the robust framerate speed made up for this, able to process 15 frames per second. / Includes bibliography. / Thesis (M.S.)--Florida Atlantic University, 2018. / FAU Electronic Theses and Dissertations Collection

Accurate registration of 3D data is one of the most challenging problems in a number of Computer Vision applications. Visual Odometry is one such application, which determines the motion, or change in position of a moving rover by registering 3D data captured by an on-board range sensor, in a pairwise manner. The performance of Visual Odometry depends upon two main factors, the first being the quality of 3D data, which itself depends upon the type of sensor being used. The second factor is the robustness of the registration algorithm. Where sensors like stereo cameras and LIDAR scanners have been used in the past to improve the performance of Visual Odometry, the introduction of the Velodyne LIDAR scanner is fairly new and has been less investigated, particularly for odometry applications. This thesis presents and examines a new method for registering 3D point clouds generated by a Velodyne scanner mounted on a moving rover. The method is based on one of the the most widely used registration algorithms called Iterative Closest Point (ICP). The proposed method is divided into two steps. The first step, which is also the main contribution of this work, is the introduction of a new point sampling method, which prudently select points that belong to the regions of greatest geometric variance in the scan. Interest Point (Region) Sampling plays an important role in the performance of ICP by effectively discounting the regions with non-uniform resolution and selecting regions with a high geometric variance and uniform resolution. Second step is to use sampled scan pairs as the input to a new plane-to-plane variant of ICP, known as Generalized ICP. Several experiments have been executed to test the compatibility and robustness of Interest Point Sampling (IPS) for a variety of terrain landscapes. Through these experiments, which include comparisons of variants of ICP and past sampling methods, this work demonstrates that the combination of IPS and GICP results in the least localization error as compared to all other tested method. / Thesis (Master, Electrical & Computer Engineering) -- Queen's University, 2011-11-03 11:12:43.596