Indoor Simultaneous Localization And Mapping Research Articles

FPP-SLAM: indoor simultaneous localization and mapping based on fringe projection profilometry.

Simultaneous localization and mapping (SLAM) plays an important role in autonomous driving, indoor robotics and AR/VR. Outdoor SLAM has been widely used with the assistance of LiDAR and Global Navigation Satellite System (GNSS). However, for indoor applications, the commonly used LiDAR sensor does not satisfy the accuracy requirement and the GNSS signals are blocked. Thus, an accurate and reliable 3D sensor and suited SLAM algorithms are required for indoor SLAM. One of the most promising 3D perceiving techniques, fringe projection profilometry (FPP), shows great potential but does not prevail in indoor SLAM. In this paper, we first introduce FPP to indoor SLAM, and accordingly propose suited SLAM algorithms, thus enabling a new FPP-SLAM. The proposed FPP-SLAM can achieve millimeter-level and real-time mapping and localization without any expensive equipment assistance. The performance is evaluated in both simulated controlled and real room-sized scenes. The experimental results demonstrate that our method outperforms other state-of-the-art methods in terms of efficiency and accuracy. We believe this method paves the way for FPP in indoor SLAM applications.

High-Precision SLAM Based on the Tight Coupling of Dual Lidar Inertial Odometry for Multi-Scene Applications

Simultaneous Localization and Mapping (SLAM) is an essential feature in many applications of mobile vehicles. To solve the problem of poor positioning accuracy, single use of mapping scene, and unclear structural characteristics in indoor and outdoor SLAM, a new framework of tight coupling of dual lidar inertial odometry is proposed in this paper. Firstly, through external calibration and an adaptive timestamp synchronization algorithm, the horizontal and vertical lidar data are fused, which compensates for the narrow vertical field of view (FOV) of the lidar and makes the characteristics of vertical direction more complete in the mapping process. Secondly, the dual lidar data is tightly coupled with an Inertial Measurement Unit (IMU) to eliminate the motion distortion of the dual lidar odometry. Then, the value of the lidar odometry after correcting distortion and the pre-integrated value of IMU are used as constraints to establish a non-linear least-squares objective function. Joint optimization is then performed to obtain the best value of the IMU state values, which will be used to predict the state of IMU at the next time step. Finally, experimental results are presented to verify the effectiveness of the proposed method.

Research on SLAM based on RBPF algorithm in indoor environment

In order to realize the simultaneous localization and mapping (SLAM) of robots in indoor environments, a SLAM method for four-wheel mobile robots based on RBPF algorithm and lidar is proposed. The mobile robot realizes its own positioning during the movement. The lidar scans the location of indoor obstacles, updates the map in real time, and gradually realizes the construction of a local map to a global map through data association. Aiming at the particle barrenness that may occur when the RBPF algorithm realizes SLAM during resampling, an adaptive resampling method is adopted to ensure that there are enough particles to realize SLAM every time. The experimental results show that when the linear velocity and angular velocity of the four-wheel mobile robot are small, indoor SLAM can be better realized.

LFM: A Lightweight LCD Algorithm Based on Feature Matching between Similar Key Frames.

Loop Closure Detection (LCD) is an important technique to improve the accuracy of Simultaneous Localization and Mapping (SLAM). In this paper, we propose an LCD algorithm based on binary classification for feature matching between similar images with deep learning, which greatly improves the accuracy of LCD algorithm. Meanwhile, a novel lightweight convolutional neural network (CNN) is proposed and applied to the target detection task of key frames. On this basis, the key frames are binary classified according to their labels. Finally, similar frames are input into the improved lightweight feature matching network based on Transformer to judge whether the current position is loop closure. The experimental results show that, compared with the traditional method, LFM-LCD has higher accuracy and recall rate in the LCD task of indoor SLAM while ensuring the number of parameters and calculation amount. The research in this paper provides a new direction for LCD of robotic SLAM, which will be further improved with the development of deep learning.

A Vertex-to-Edge Weighted Closed-Form Method for Dense RGB-D Indoor SLAM

RGB-D dense mapping systems are widely used for indoor navigation and augmented reality. However, the drift in egomotion estimation systems accumulates as the systems expand over space and time. In this paper, we develop a vertex-to-edge weighted closed-form algorithm to reduce camera drift for dense RGB-D indoor simultaneous localization and mapping (SLAM). First, due to the high-frame rate of common RGB-D sensors, in our approach, we employ a robust key-frame selection strategy. We then consider the typical depth noise properties of the sensor and exploit both visual and geometric information to determine the camera motion of the key frames. To enforce global consistency, we employ a combination of techniques to enhance the efficiency and accuracy of loop closure detection and then factor graph optimization is used to mitigate trajectory drift; specifically, the graph edges are weighted by residual errors. To demonstrate its accuracy and robustness, our SLAM system is tested on seven sets of sequences from publicly available benchmark datasets collected with Kinect v1 and on three sets of sequences taken with a handheld structure sensor. In a direct comparison, our approach yields significantly less trajectory error than several other state-of-the-art methods.

Research on simultaneous localization and mapping of indoor mobile robot

Simultaneous localization and mapping (SLAM) is the key technology to fulfill mobile robot obstacle avoidance and autonomous navigation. As cameras such as binocular cameras or RGB-D cameras exist wide-angle limitations, the use of such sensors prone to map mismatches and localization errors. In order to solve the aforementioned problems, 2D laser radar and odometer are used as the core sensor to design an experimental system for indoor SLAM, which is configured on a mobile robot to examine the effect of indoor SLAM. Through the Rao-Blackwellized [1] particle filter algorithm, the data of laser radar and odometer are fused with real-time processing, which realizes the construction of map as well as localization and navigation of the mobile robot in the unknown environment. Experiments on simulation and physical verification were carried out with open source robot operating system (ROS). The simulation and experimental results confirm the feasibility and practicability of the platform.

Improved Map Generation by Addition of Gaussian Noise for Indoor SLAM using ROS

Rao-Blackwellized Particle Filter (RBPF) is used in this paper to solve the Simultaneous Localization and Mapping (SLAM) problem. RBPF algorithm uses particle filter where each particle carries an individual map of the environment. With the usage of Robot Operating System (ROS), GMapping package was used as a basis for map generation and SLAM. To improve the map generation, Gaussian noise was introduced to the data from laser range finder and also the odometry from the robot Pioneer P3AT’s pose. The introduced algorithm was successful in decreasing the uncertainty as well as increased the knowledge of each particle in the estimation of the robot’s pose, proven through practical experiment. Exploration experiments were also carried out to test the performance of P3AT based on our proposed method.

Indoor laser-based SLAM for micro aerial vehicles

This article presents a laser-based 2D simultaneous localization and mapping (SLAM) algorithm for indoor environments. An adaption and optimization of a ground vehicle SLAM solution (TinySLAM) for the use with Micro Aerial Vehicles is proposed. Optimizations of the map update strategy and a motion model improves the accuracy strongly. An extension to 3D mapping is introduced. The presented algorithm is tested with simulated and real world data. The optimized SLAM solution maps a whole floor of an office building very accurately and achieves embedded real-time capability.

A New Method For Indoor SLAM Based On Artificial Landmark

A mobile robot must be capable of localizing itself in unknown environments and constructing a map of the environments at the same time. Simultaneous localization and mapping (SLAM) is a challenging problem in indoor envi- ronments because GPS information isn't available. An algorithm is proposed in this paper for SLAM with vision sensors, which is designed by utilizing the artificial landmark MR code and integrating the FastSLAM algorithm, the method can decrease the time cost compared with particle filter and improve the accuracy of localization and mapping compared with natural landmarks. Experiments show the effectiveness and robustness of this algorithm.

Indoor SLAM Using Laser and Camera with Closed-Loop Controller for NAO Humanoid Robot

We present a SLAM with closed-loop controller method for navigation of NAO humanoid robot from Aldebaran. The method is based on the integration of laser and vision system. The camera is used to recognize the landmarks whereas the laser provides the information for simultaneous localization and mapping (SLAM ). K-means clustering method is implemented to extract data from different objects. In addition, the robot avoids the obstacles by the avoidance function. The closed-loop controller reduces the error between the real position and estimated position. Finally, simulation and experiments show that the proposed method is efficient and reliable for navigation in indoor environments.

Improvement of speeded-up robust features for robot visual simultaneous localization and mapping

SUMMARYA robot mapping procedure using a modified speeded-up robust feature (SURF) is proposed for building persistent maps with visual landmarks in robot simultaneous localization and mapping (SLAM). SURFs are scale-invariant features that automatically recover the scale and orientation of image features in different scenes. However, the SURF method is not originally designed for applications in dynamic environments. The repeatability of the detected SURFs will be reduced owing to the dynamic effect. This study investigated and modified SURF algorithms to improve robustness in representing visual landmarks in robot SLAM systems. Many modifications of the SURF algorithms are proposed in this study including the orientation representation of features, the vector dimension of feature description, and the number of detected features in an image. The concept of sparse representation is also used to describe the environmental map and to reduce the computational complexity when using extended Kalman filter (EKF) for state estimation. Effective procedures of data association and map management for SURFs in SLAM are also designed to improve accuracy in robot state estimation. Experimental works were performed on an actual system with binocular vision sensors to validate the feasibility and effectiveness of the proposed algorithms. The experimental examples include the evaluation of state estimation using EKF SLAM and the implementation of indoor SLAM. In the experiments, the performance of the modified SURF algorithms was compared with the original SURF algorithms. The experimental results confirm that the modified SURF provides better repeatability and better robustness for representing the landmarks in visual SLAM systems.

A Fast and Robust Feature-Based Scan-Matching Method in 3D SLAM and the Effect of Sampling Strategies

Simultaneous localization and mapping (SLAM) plays an important role in fully autonomous systems when a GNSS (global navigation satellite system) is not available. Studies in both 2D indoor and 3D outdoor SLAM are based on the appearance of environments and utilize scan-matching methods to find rigid body transformation parameters between two consecutive scans. In this study, a fast and robust scan-matching method based on feature extraction is introduced. Since the method is based on the matching of certain geometric structures, like plane segments, the outliers and noise in the point cloud are considerably eliminated. Therefore, the proposed scan-matching algorithm is more robust than conventional methods. Besides, the registration time and the number of iterations are significantly reduced, since the number of matching points is efficiently decreased. As a scan-matching framework, an improved version of the normal distribution transform (NDT) is used. The probability density functions (PDFs) of the reference scan are generated as in the traditional NDT, and the feature extraction - based on stochastic plane detection - is applied to the only input scan. By using experimental dataset belongs to an outdoor environment like a university campus, we obtained satisfactory performance results. Moreover, the feature extraction part of the algorithm is considered as a special sampling strategy for scan-matching and compared to other sampling strategies, such as random sampling and grid-based sampling, the latter of which is first used in the NDT. Thus, this study also shows the effect of the subsampling on the performance of the NDT.

Optimal Filtering for Non-parametric Observation Models: Applications to Localization and SLAM

In this work we address the problem of optimal Bayesian filtering for dynamic systems with observation models that cannot be approximated properly as any parameterized distribution. In the context of mobile robots this problem arises in localization and simultaneous localization and mapping (SLAM) with occupancy grid maps. The lack of a parameterized observation model for these maps forces a sample-based representation, commonly through Monte Carlo methods for sequential filtering, also called particle filters. Our work is grounded on the demonstrated existence of an optimal proposal distribution for particle filters. However, this optimal distribution is not directly applicable to systems with non-parametric models. By integrating ideas from previous works on adaptive sample size, auxiliary particle filters, and rejection sampling, we derive a new particle filter algorithm that enables the usage of the optimal proposal to estimate the true posterior density of a non-parametric dynamic system. This new filter is better suited, both theoretically and in practice, than previous approximate methods for indoor and outdoor localization and SLAM, as confirmed by experiments with real robots.