Object-spatial layout-route based hybrid map and global localization for mobile robots

This paper presents a new object entity based global localization approach with stereo camera. A local invariant feature and stereo depth information are used as visual features. The map we use here is a hybrid of global topological map and local object location map. The topological map includes some semantic information about the representing space and the object entities in the space. The object location map has the pose information of each object entity and visual features for object recognition. The localization process consists of two stages: coarse pose estimation and refined pose estimation. The coarse pose is computed by using the object recognition and point cloud fitting method. And the refined pose is estimated with particle filtering algorithm. An experiment shows that our approach can be an effective vision-based global localization method.

Ultrasonic guided-wave testing can greatly benefit from (1) an ability to provide quantitative information on the damage that is being detected, and (2) an ability to select the best mode-frequency combination for maximum sensitivity to a given type of damage. Achieving these capabilities in complex structures (e.g. nonprismatic structures such as a stiffened panel in aerospace fuselages) is a nontrivial task. This paper will discuss an improved Global-Local (GL) method where the geometrical “local” discontinuity (e.g. the stiffener) is modelled by traditional FE discretization and the rest of the structure (“global” part) is modelled by Semi-Analytical Finite Element (SAFE) cross-sectional discretization. The boundaries of the “local” domain and the “global” domain are then matched in terms of wave displacement and stresses. GL models have been proposed in the past using theoretical (Lamb) wave solutions that only apply to isotropic plates. The authors have also previously studied GL methods using the SAFE approach for application to multi-layered anisotropic plates for which theoretical solutions are either not existent or hard to obtain. This work will extend recent research on these methods by optimizing the Matlab routine that is used to run the GL code, correcting some formulation errors that were present in the previous edition, and studying the specific case of a composite panel stiffened with cocured stringers that is representative of modern commercial aircraft construction (e.g. Boeing 787). The newlyformulated GL method will be shown to provide excellent results that can help designing a guided-wave test on these aircraft components for optimum detection of relevant damage that can be induced by impacts (including skin delaminations, stringer heel cracks, and stringer to skin disbonds). Other applications of the GL methods beyond stiffened aircraft panels will be discussed.

Robotic technologies will continue to enter new applications in addition to automated manufacturing and logistics. Once mobile robots can also operate outside of today's special facilities, they have the potential to relieve us of dirty and dangerous labor in various areas. However, for this purpose, these machines will need to be able to navigate autonomously in complex natural, urban, and industrial settings. This thesis addresses the development of locomotion skills for legged robots in challenging environments. Our work focuses on perceptive locomotion where exteroceptive sensing of the surrounding is exploited to plan and control the robot’s motion. This enables quadrupedal robots to negotiate rough terrain through carefully selected contacts. In this work, we evaluate different sensing technologies and analyze their performance for local dense terrain mapping on a mobile robot. We include special conditions such as close range objects and the influence of ambient light as we find them in real-world applications. By modeling the error characteristics of the sensors, the robot can judge the quality of the resulting terrain reconstruction. As the robot moves, the surrounding is continuously mapped to capture new areas and update regions which have changed. We contribute with a mapping framework that models the terrain from a robot-centric perspective. To this end, we present a novel approach for the error propagation from the robot's state estimation to the representation of the map. This formulation allows for robust and high-rate local mapping that is independent of a global localization method. We introduce our approach to locomotion planning, which finds safe footholds along with collision-free swing-leg motions, leveraging the generated terrain map. A nonlinear optimization finds postures that respect kinematic and stability constraints. We experimentally verify this work with torque-controllable quadrupedal robots that autonomously traverse obstacles, such as rubble, steps, gaps, and stairs without prior knowledge of the scene or any additional equipment. The locomotion planner re-plans its motion at every step in real-time, to cope with disturbances and dynamic environments. For the control of the legged robot, we contribute architecturally to the versatile and task-oriented motion execution. This method enables the robust tracking of motion plans, even with significant mismatches between the models and reality. In addition to rough terrain locomotion, we demonstrate the integration of our method for applications, such as whole-body stair climbing, manipulation, jumping, docking, inspection, payload delivery, dancing, and more. Our approach is thoroughly validated with the quadrupedal robot ANYmal in realistic long-term missions for autonomous industrial inspection and search and rescue. Finally, we extend our work with the design and implementation of a collaborative navigation framework for ground and flying robots. The ground vehicle utilizes the data captured by the flying robot to navigate uncharted…

Position estimation based on scan matching is a type of popular approaches for mobile robot localization in indoor office-like environments. Global scan matching and local scan matching are two basic position estimation problems. Global scan matching is subject to global localization for mobile robot and local scan matching is used to deal with tracking mobile robot position over time. Global localization based on scan matching is represented in this paper. A global map is built to represent whole environment feature and a local map is built to represent current range scan. Global map and local map consist of a lot of line segments. Complete line segment that represents a complete environment object is singled out and used to match between local map and global map. Line segments are sorted counterclockwise in two maps. This type of ordinal map improves the searching efficiency of matching and localization algorithm is timesaving. In addition, matching between current range scan and global map is based on relative position relationships of line segments in local and global map. Global localization based on relative position relationship avoids frequent coordinates transformation and lowers computational cost. All these techniques have been implemented on our mobile robot ATRVII equipped with 2D laser range scanner SICK.

We present a method for solving the correspondence problem in Simultaneous Localisation and Mapping (SLAM) in a topological map. The nodes in the topological map are a representation for each local space the robot visits. The approach is feature based – a neural network algorithm is used to learn a signature from a set of features extracted from each local space representation. Newly encountered local spaces are classified by the neural network as to how well they match the signatures of the nodes in the topological network. Of equal importance as the correspondence problem is its dual, that of perceptual aliasing which occurs when parts of the environment which appear the same are in fact different. It manifests itself as false positive matches from the neural network classification. Our approach to solving this aspect of the problem is to use the context provide by nodes in the neighbourhood of the (mis)matched node. When neural network classification indicates a correspondence then subsequent local spaces the robot visits should also match nodes in the topological map where appropriate.KeywordsMobile RobotLocal SpaceCorrespondence ProblemMatched NodeNeural Network ClassificationThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

This paper presents a global localization method for mobile robots based on the geometric information of structured indoor environments. With a global/local point cloud and projection map, lines are extracted from the projection maps using Hough transform. According to the directions of the obtained lines, the orientations of projection maps and point clouds are normalized. Next, the template matching algorithm is applied to the normalized global and local projection maps. Once coarse localization is completed, final accurate localization is achieved using the Iterative Closest Points (ICP) algorithm. Experimental results on several point clouds show that the proposed method can achieve high localization accuracy in real-time. The proposed method can be used for other global localization applications in structured environments.

A novel topological map is proposed in this paper. By virtue of the topological map whose nodes are represented with the range finderpsilas free beams together with the visual scale-invariant features, the mobile robot can autonomously navigates in unknown, large-scale and indoor environments. Different from the traditional navigation methods that rely on accurate global localization, the robot locates itself imprecisely by location recognition rather than calculating its global pose in the world reference frame. By combining the reactive navigational method, BCM, with the topological map, a smooth, real-time navigation without global localization can be realized.

Global localization in a prior map is an important field in virtual and augmented reality systems, but it is always a challenge to conduct point cloud based localization in the large-scale scene prior map. A large-scale prior map usually means huge amount of calculation for point cloud processing, which leads to the long time required for global localization. To deal with this problem, we propose a fast point cloud global localization method based on point clouds segmentation and learning-based descriptor. On the one hand, cylindrical filtering, ground-point removal and point cloud segmentation are adopted to eliminate a large number of useless points and retain points with rich structures, which improves the efficiency of point cloud registration. On the other hand, reliable 3D point cloud descriptor, two-phase search strategy for place recognition and geometric consistency verification are used to ensure the localization accuracy. Experiments prove that the proposed method achieves good localization effect on both KITTI and MVSEC datasets. Under the condition of ensuring the high localization accuracy, the time for point clouds to complete the global localization is greatly reduced.

A hybrid mapping framework is presented in this work. The goal is to obtain better computational efficiency than pure metrical mapping techniques and better accuracy as well as usability for robot guidance and navigation compared to the topological mapping. Image sequences acquired in an environment by manually driving a robot are used to build a hierarchical map representation by using an image sequence partitioning (ISP) technique that uses local image features. The hierarchical map built can be understood as a topological map with nodes corresponding to certain regions in the environment. Each node in turn is made up of a set of images acquired in that region. These maps are further augmented with metrical information at those nodes which correspond to image subsequences acquired while the robot is turning as a part of its trajectory. Metrical information becomes invaluable during autonomous robot navigation through these places. Hence, we call the resulting maps hybrid since they primarily contain topological information and metrical information at places that are important for navigation. Experimental results obtained on a sequence acquired in an outdoor environment are provided to demonstrate our approach.

Map building and localization are essential to an autonomous mobile robot system. A global topological map is commonly constructed from a global grid map using a Voronoi diagram. In this hierarchical method, however, the advantage of a topological map such as compactness may not be fully utilized. In this paper sensor data are used to build a local grid map from which a local topological map is created using a thinning method. The thinning method is an alternative of the Voronoi diagram. The local topological map is then updated to the global one. Localization based on the topological map is usually difficult, but additional nodes created by the thinning method can improve localization performance. Various experiments with a laser scanner demonstrate validity and feasibility of the proposed algorithms for map building and localization.

In conventional navigation systems, inertial sensors consist of accelerometers and gyroscopes. These sensors suffer from in‐built errors, accumulated drift and high‐level noise sensitivity. The accurate gyroscopes are expensive and not suitable in cost‐effective applications. To minimise such disadvantages, one solution is the combination of inertial sensors with different aiding sensors. To lower the cost, utilisation of redundant accelerometer structure as gyro‐free inertial measurement unit (GFIMU) has been proposed. In this study, Gyro‐Free navigation errors using four tri‐axial accelerometers are illustrated. Compensation of errors in terms of angular velocity and position estimation is verified based on adding a simple gyroscope, inexpensive stereo cameras as well as creating an easy to use topological map. The topological map is easily created by means of scale‐invariant feature transform method. The estimation of angular velocity is corrected on the basis of fusing the measurements from GFIMU and a simple gyroscope using unscented Kalman filter. The correction of position is performed by comparing the estimated position from GFIMU and observation of stereo cameras together with topological map. The results of the research show that the collaboration of GFIMU, stereo cameras and simple gyroscope will improve the robustness and accuracy of navigation, significantly.

This article presents an adaptive dynamic clustered particle filtering method for mobile robot global localization. The posterior distribution of robot pose in global localization is usually multimodal due to the symmetry of the environment and ambiguous detected features. Moreover, the multimodal distribution of the posterior varies as the robot moves and observations be obtained. Considering these characteristics, we use a set of clusters of particles to represent the posterior. These clusters are dynamically evolved corresponding to the varying posterior by merging the overlapping clusters and splitting the diffuse clusters or those whose particles gather to some sub-clusters inside. Further, in order to improve computational efficiency without sacrificing estimation accuracy, a mechanism for adapting the sample size of clusters is proposed. The theoretical lower bound of the number of particles needed to limit the estimation error is derived, based on the central limit theorem in multi-dimensional space and the statistic theory of Importance Sampling. Then, a method for tuning the sample size for each cluster according to the derived lower bound is presented. Experiment results show the effectiveness of the proposed method, which is sufficient to achieve robust tracking of robot's real pose and meanwhile significantly enhance the computational efficiency.

H-SLAM: Hybrid direct–indirect visual SLAM

In the task of hyperspectral image classification, how to learn features of hyperspectral image is the important and difficulty issue which may directly affect the classification results. Inspired by the idea of natural language processing, in this paper, we propose a local space long short-term memory network based hyperspectral image classification, which constructs sequential features in the local area of hyperspectral images. This method is based on the integration features of two traditional low-level features, and from these integration features to extract sequential features of the center sample in the local space, then use the long short-term memory network to learn high-level semantic features, finally use them to classify image. This method can not only obtain more representative and discriminative high-level semantic features, and through constructing the local space sequence to enhance positive impact of the useful pixels, inhibit negative effects of useless pixels, it improves the classification accuracy.

The technique of simultaneous localization and mapping has received much attention recently in mobile robotics. When map is being built, robot memorizes environmental information on the plane of grid or topology. Several approaches about this research have been presented so far, but most of them use mapping technique as either grid-based map or topology-based map. This paper proposes some algorithms integrating grid and topology map and its performance is faster and more stable. The proposed scheme uses an occupancy grid map in representing the environment and then formulate topological information in path finding by A* algorithm. The localization process and the dynamic obstacle avoidance can be accomplished by topological information on grid map. On simulator mapping process is shown and the shortest path is decided on grid based map. Then topological information such as direction, distance is calculated on simulator program then transmitted to robot hardware devices. Three ultrasonic sensors and electronic compass are implemented on our mobile robot and two DC geared motor wheels are individually controlled by PD control method. Simulations and experimental results show the performance and efficiency of the proposed scheme.