Omnidirectional Vision Based Mobile Robot Hierarchical SLAM

This paper first demonstrates an interesting property of bundle adjustment (BA), drift correction. Here drift correction means that BA can converge to the correct solution (up to a scale) even if the initial values of the camera pose translations and point feature positions are calculated using very different scale factors. This property together with other properties of BA makes it the best approach for monocular Simultaneous Localization and Mapping (SLAM), without considering the computational complexity. This naturally leads to the idea of using local BA and map joining to solve large-scale monocular SLAM problem, which is proposed in this paper. The local maps are built through Scale-Invariant Feature Transform (SIFT) for feature detection and matching, random sample consensus (RANSAC) paradigm at different levels for robust outlier removal, and BA for optimization. To reduce the computational cost of the large-scale map building, the features in each local map are judiciously selected and then the local maps are combined using a recently developed 3D map joining algorithm. The proposed large-scale monocular SLAM algorithm is evaluated using a publicly available dataset with centimeter-level ground truth.

Simultaneous localization and mapping (SLAM) becomes popular in the autonomous robot research community. SLAM requests a mobile robot, placed at an unknown location in an unknown environment, to incrementally build a consistent map of this environment while simultaneously determining its location within this map. The odometer data is helpful but the noisy error accumulates. Assistant with measurements from the other sensors, SLAM improves the accuracy by filtering. The extended Kalman filter is the most used method to feedback the error. In this thesis, a monocular vision based SLAM for a wheeled robot is proposed. A single web camera is the only measurement input to correct the odometer data. Extract features and match them based on the scale invariant feature transform (SIFT). Absence of the feature depth information is the key issue for the monocular vision in SLAM. The inverse depth parametrization method solves the problem by present the position in inverse coordinate with some reasonable assumptions. Convergence of the feature position is more easily achieved by the EKF due to the linearity. Combining the odometer data with the inverse depth method, this system provides an accurate map by tests of the indoor image sequence.

Map learning and self-localization based only on a perception of an environment’s structure are fundamental cognitive capacities required for intelligent robot’s to realize true autonomy. Simultaneous localization and mapping (SLAM) is an effective technique for such robots, as it addresses the problem of incrementally building an environment map from noisy sensory data and tracking the robot’s pose with the built map. While the Rao-Blackwellized particle filter (RBPF) is a popular SLAM technique, it tends to accumulate errors introduced by inaccurate linearization of the SLAM nonlinear function. Accordingly, RBPF-SLAM will usually fail to close large loops when applied to large-scale environments. To overcome this drawback, a new Jacobian-free RBPF-SLAM algorithm is derived in this paper. The main contribution of the algorithm lies in the utilization of the 5th-order conjugate unscented transform, which calculates the SLAM transition density up to the 5th order, to give a better distribution of the particle filter and discover local features and landmarks. The performance of the proposed SLAM is investigated and compared with that of FastSLAM2.0 and UFastSLAM in both indoor and outdoor experiments. The results verify that the proposed algorithm improves the SLAM performance in large-scale environments.

An artificial intelligent grey wolf optimizer (GWO)-assisted resampling scheme is applied to the Rao-Blackwellized particle filter (RBPF) in the simultaneous localization and mapping (SLAM). By doing this, we can make the diversity of the particles resampling and then obtain a better localization accuracy and fast convergence to realize indoor mobile robot SLAM. In addition, we propose an adaptive local data association (Range-SLAM) scheme to improve the computational efficiency for the algorithm of the nearest neighbor (NN) data association in the iteration of the RBPF prediction. Through the experiment and simulations, the proposed SLAM schemes have fast convergence, accuracy, and heuristics. Therefore, the improved RBPF and new data association schemes presented in this paper can provide a feasible method for the indoor mobile robot SLAM.

The simultaneous localization and mapping (SLAM) is considered as a crucial prerequisite for purely autonomous mobile robots. In this paper, we demonstrate the mobile robot SLAM using Rao-Blackwellized particle filters (RBPF) through computer simulations under MATLAB platform, while an analytical investigation into the involved algorithms is presented. Then we make further comparisons, not only in parallel between the FastSLAM 1.0 and FastSLAM 2.0, also in vertical between FastSLAM performance and EKF-SLAM performance which used to be the dominant approach to the SLAM problem. Vivid simulations and numerical analysis make the paper illustrated with clarity and perception.

In robotics, the problem of concurrently addressing the localization and mapping is well defined as simultaneous localization and mapping (SLAM) problem. Since the SLAM procedure is usually recursive, maintaining a certain error bound on the current position estimate is a critical issue. However, when the robot is kidnapped (i.e., the robot is moved by an intentional or unintentional user) or suffers from locomotion failure (due to large slip and falling), the robot will inevitably lose its current position. In this case, immediate recovery of the robot position is essential for seamless operation. In this paper, we present a method of solving both SLAM and relocation problems by employing ambient magnetic and radio measurements. The proposed SLAM is realized in the Rao-Blackwellized particle filter- and grid-based SLAM frameworks, where we exploit the local heading corrections from the magnetic measurements. For the relocation, we design the location signatures using the magnetic and radio measurements, and examine each of the Monte Carlo localization-based and multilayer perceptron-based relocation methods with real-world data. We implement the proposed SLAM and relocation algorithms in an embedded system and verify the feasibility of the proposed methods as an online robot navigation system.

This paper describes a new system, dubbed Continuous Appearance-based Trajectory Simultaneous Localisation and Mapping (CAT-SLAM), which augments sequential appearance-based place recognition with local metric pose filtering to improve the frequency and reliability of appearance-based loop closure. As in other approaches to appearance-based mapping, loop closure is performed without calculating global feature geometry or performing 3D map construction. Loop-closure filtering uses a probabilistic distribution of possible loop closures along the robot’s previous trajectory, which is represented by a linked list of previously visited locations linked by odometric information. Sequential appearance-based place recognition and local metric pose filtering are evaluated simultaneously using a Rao–Blackwellised particle filter, which weights particles based on appearance matching over sequential frames and the similarity of robot motion along the trajectory. The particle filter explicitly models both the likelihood of revisiting previous locations and exploring new locations. A modified resampling scheme counters particle deprivation and allows loop-closure updates to be performed in constant time for a given environment. We compare the performance of CAT-SLAM with FAB-MAP (a state-of-the-art appearance-only SLAM algorithm) using multiple real-world datasets, demonstrating an increase in the number of correct loop closures detected by CAT-SLAM.

We propose a robust simultaneous localization and mapping (SLAM) with a Rao-Blackwellized particle filter (RBPF) algorithm for mobile robots using sonar sensors in non-static environments. The algorithm consists of three parts: sampling from multiple ancestor sets, estimating intermediate paths for map updates and eliminating spurious landmarks using negative information from sonar sensors. The proposed sampling method, in which particles are sampled from multiple ancestor sets, increases the robustness of the estimation of the robot's pose, even if environmental changes corrupt observations. This step increases the probability of some particles being sampled from correct ancestor sets that are updated by observations reflected from stationary objects. When particles are sampled from several time steps earlier, however, observations at intermediate time steps cannot be used to update the map because of the lack of information about the intermediate path. To update the map with all sensor information, the intermediate path is estimated after particles are sampled from ancestor sets. Finally, spurious landmarks still exist on the map representing objects that were eliminated or that were extracted by error in cluttered areas. These are eliminated in the final step using negative information from the sonar sensors. The performance of the proposed SLAM algorithm was verified through simulations and experiments in various non-static environments.

Feature point extraction and matching are the key modules in the front-end of visual simultaneous localization and mapping (SLAM), which determines the accuracy of locating and map Nbuilding robots. To obtain a good performance, the image extraction algorithm being applied should have high efficiency and robustness to the changes of the surrounding environment. In this paper, three widely used feature extraction algorithms, namely Scale-invariant feature transform (SIFT), speed up robust features (SURF) and oriented FAST, and rotated BRIEF (ORB), are studied. The basic concepts and details of the three algorithms are introduced. Then the execution time of the three algorithms is compared to evaluate their efficiency. Meanwhile, the key points matching rate of the three algorithms are compared under various transformations and distortions. The experiment results show that ORB has higher efficiency than SIFT and SURF. However, SIFT has better robustness to image rotation, illumination change, scaling transformation, Gaussian noise and salt-and-pepper noise. In addition, the performance of the ORB-SLAM system is tested on different datasets. It has great ability in camera trajectory estimation and map building and therefore is suitable for deploying in industry.

An important milestone for building affordable robots that can become widely popular is to address robustly the simultaneous localization and mapping (SLAM) problem with inexpensive, off-the-shelf sensors, such as monocular cameras. These sensors, however, impose significant challenges on SLAM procedures because they provide only bearing data related to environmental landmarks. This paper starts by providing an extensive comparison of different techniques for bearing-only SLAM in terms of robustness under different noise models, landmark densities and robot paths. We have experimented in a simulated environment with a variety of existing online algorithms including Rao-Blackwellized particle filters (RB-PFs). Our experiments suggest that RB-PFs are more robust compared to other existing methods and run considerably faster. Nevertheless, their performance suffers in the presence of outliers. In order to overcome this limitation we proceed to propose an augmentation of RB-PFs with: (a) Gaussian sum filters for landmark initialization and (b) an online, unsupervised outlier rejection policy. This framework exhibits impressive robustness and efficiency even in the presence of outliers

Simultaneous localization and mapping (SLAM) for mobile robots has attracted research interests in the past two decades. Recent years, Rao-Blackwellized particle filter (RBPF) approach proved to be an effective means to estimate the full SLAM posterior. However, most of the SLAM practices are implemented in static environment. To navigate in dynamic environment, model based approaches have been implemented. The approaches require modeling, classification, and data association. Modeling and classification often require prior knowledge of the environment and are less robust in highly complex environment. To obviate modeling, classification and data association in dynamic environment, a solution framework based on Rao-Blackwellised genetic algorithmic filter (RBGAF) is extended for recovering the full SLAM posterior, using raw sensor measurements. The resultant extended RBGAF-SLAM recovers the negative sensor information during SLAM so that the moving objects can be efficiently identified in the particle maps. This approach permits the implementation of SLAM without prior knowledge of the environment and the static environment assumption. Simulations and experimental results obtained in an outdoor environment using a laser measurement system are presented to demonstrate the method's effectiveness.

Autonomous and safe robot navigation requires the capability to simultaneously building a map of the environment and a selflocalization of the robot itself. This is known as the SLAM (Simultaneous Localization and Mapping) problem. In such a context, omnidirectional camera looks like a very interesting sensor since it allows a full 360 degrees field of vision. Complexity of the SLAM methods dramatically increases when the size of the environment grows up. Conversely, accuracy and integrity of the estimation process cannot be guaranteed any more. In this paper, we present an efficient way to reduce the complexity of the algorithms thanks to the definition of local submaps. In contrast with other papers dealing with local maps, our main purpose is not only to reduce the computational cost but also to keep the global map correlated. A method to consistently share information between local maps is also provided. These methods are used in the context of the omnidirectional bearing-only SLAM. Due to the specificity of the model of projection, images provided by the omnidirectional cameras cannot be used in a SLAM method without taking some precaution. In this paper, we provide an original method for computing the covariance matrix taking into account this specificity. Finally, we provide a validation with real data in an indoor environment subject to large illumination changes.

This contribution aims at unifying two recent trends in applied particle filtering (PF). The first trend is the major impact in simultaneous localization and mapping (SLAM) applications, utilizing the FastSLAM algorithm. The second one is the implications of the marginalized particle filter (MPF) or the Rao-Blackwellized particle filter (RBPF) in positioning and tracking applications. Using the standard FastSLAM algorithm, only low-dimensional vehicle models are computationally feasible. In this work, an algorithm is introduced which merges FastSLAM and MPF, and the result is an algorithm for SLAM applications, where state vectors of higher dimensions can be used. Results using experimental data from a UAV (helicopter) are presented. The algorithm fuses measurements from on-board inertial sensors (accelerometer and gyro) and vision in order to solve the SLAM problem, i.e., enable navigation over a long period of time.

A new approach to vision-based simultaneous localization and mapping (SLAM) is proposed. the scale invariant feature transform (SIFT) features is landmarks, The minimal connected dominating set(CDS) approach is used in data association which solve the problem that the scale of data association increase with the map grows in process of SLAM . SLAM is completed by fusing the information of binocular vision and robot pose with Extended Kalman Filter (EKF).the system has been implemented and tested on data gathered with a mobile robot in a typical office environment. Experiments presented demonstrate that proposed method improves the data association and in this way leads to more accurate maps.

SIFT (scale invariant feature transform) is one of the most robust and the widely used image matching algorithms based on local features. It has been used in the aspect of the cameras target matching. For its high complexity, this paper proposes a target matching method for multiple cameras by combining SIFT and the rotation-invariant LBP (local binary patterns). The object detection is obtained by improved Back-ground subtraction technique, and particle filter is used for region-based object tracking. The descriptor combining SIFT and the rotation-invariant LBP is calculated for every object region. The object matching is obtained by comparing the distance of feature matrix. Experimental results show that the proposed algorithm is more rapid than the standard SIFT algorithm while the performance is favorably compared to the standard SIFT algorithm, and it is robust in various scenes.