Tag Archives: Loop Closure

Loop closure detection by optimization of finite sets of images that correspond to each place

Han, F., Wang, H., Huang, G. et al, Sequence-based sparse optimization methods for long-term loop closure detection in visual SLAM, Autonomous Robots, Volume 42, Issue 7, pp 1323–1335, DOI: 10.1007/s1051.

Loop closure detection is one of the most important module in Simultaneously Localization and Mapping (SLAM) because it enables to find the global topology among different places. A loop closure is detected when the current place is recognized to match the previous visited places. When the SLAM is executed throughout a long-term period, there will be additional challenges for the loop closure detection. The illumination, weather, and vegetation conditions can often change significantly during the life-long SLAM, resulting in the critical strong perceptual aliasing and appearance variation problems in loop closure detection. In order to address this problem, we propose a new Robust Multimodal Sequence-based (ROMS) method for robust loop closure detection in long-term visual SLAM. A sequence of images is used as the representation of places in our ROMS method, where each image in the sequence is encoded by multiple feature modalites so that different places can be recognized discriminatively. We formulate the robust place recognition problem as a convex optimization problem with structured sparsity regularization due to the fact that only a small set of template places can match the query place. In addition, we also develop a new algorithm to solve the formulated optimization problem efficiently, which guarantees to converge to the global optima theoretically. Our ROMS method is evaluated through extensive experiments on three large-scale benchmark datasets, which record scenes ranging from different times of the day, months, and seasons. Experimental results demonstrate that our ROMS method outperforms the existing loop closure detection methods in long-term SLAM, and achieves the state-of-the-art performance.

Using short- and long-term memories in SLAM

Labbé, M. & Michaud, F., Long-term online multi-session graph-based SPLAM with memory management, Auton Robot (2018) 42: 1133. DOI: 10.1007/s10514-017-9682-5.

For long-term simultaneous planning, localization and mapping (SPLAM), a robot should be able to continuously update its map according to the dynamic changes of the environment and the new areas explored. With limited onboard computation capabilities, a robot should also be able to limit the size of the map used for online localization and mapping. This paper addresses these challenges using a memory management mechanism, which identifies locations that should remain in a Working Memory (WM) for online processing from locations that should be transferred to a Long-Term Memory (LTM). When revisiting previously mapped areas that are in LTM, the mechanism can retrieve these locations and place them back in WM for online SPLAM. The approach is tested on a robot equipped with a short-range laser rangefinder and a RGB-D camera, patrolling autonomously 10.5 km in an indoor environment over 11 sessions while having encountered 139 people.

Faster long-term SLAM through direct use of Lie groups in filtering

Kruno Lenac, Josip Ćesić, Ivan Marković, and Ivan Petrović, Exactly sparse delayed state filter on Lie groups for long-term pose graph SLAM, The International Journal of Robotics Research Vol 37, Issue 6, pp. 585 – 610 DOI: 10.1177/0278364918767756.

In this paper we propose a simultaneous localization and mapping (SLAM) back-end solution called the exactly sparse delayed state filter on Lie groups (LG-ESDSF). We derive LG-ESDSF and demonstrate that it retains all the good characteristics of the classic Euclidean ESDSF, the main advantage being the exact sparsity of the information matrix. The key advantage of LG-ESDSF in comparison with the classic ESDSF lies in the ability to respect the state space geometry by negotiating uncertainties and employing filtering equations directly on Lie groups. We also exploit the special structure of the information matrix in order to allow long-term operation while the robot is moving repeatedly through the same environment. To prove the effectiveness of the proposed SLAM solution, we conducted extensive experiments on two different publicly available datasets, namely the KITTI and EuRoC datasets, using two front-ends: one based on the stereo camera and the other on the 3D LIDAR. We compare LG-ESDSF with the general graph optimization framework (g2o) when coupled with the same front-ends. Similarly to g2o the proposed LG-ESDSF is front-end agnostic and the comparison demonstrates that our solution can match the accuracy of g2o, while maintaining faster computation times. Furthermore, the proposed back-end coupled with the stereo camera front-end forms a complete visual SLAM solution dubbed LG-SLAM. Finally, we evaluated LG-SLAM using the online KITTI protocol and at the time of writing it achieved the second best result among the stereo odometry solutions and the best result among the tested SLAM algorithms.

Using sequences of images for loop closure instead of only one

Loukas Bampis, Angelos Amanatiadis, and Antonios Gasteratos, Fast loop-closure detection using visual-word-vectors from image sequences, The International Journal of Robotics Research Vol 37, Issue 1, pp. 62 – 82, DOI: 10.1177/0278364917740639.

In this paper, a novel pipeline for loop-closure detection is proposed. We base our work on a bag of binary feature words and we produce a description vector capable of characterizing a physical scene as a whole. Instead of relying on single camera measurements, the robot’s trajectory is dynamically segmented into image sequences according to its content. The visual word occurrences from each sequence are then combined to create sequence-visual-word-vectors and provide additional information to the matching functionality. In this way, scenes with considerable visual differences are firstly discarded, while the respective image-to-image associations are provided subsequently. With the purpose of further enhancing the system’s performance, a novel temporal consistency filter (trained offline) is also introduced to advance matches that persist over time. Evaluation results prove that the presented method compares favorably with other state-of-the-art techniques, while our algorithm is tested on a tablet device, verifying the computational efficiency of the approach.

Interesting review of approaches to visually detect loop closings in robotics, and a novel, very efficient method that is independent on the image representation and based on not using the typical l2 norm (least squares), which leads to dense optimization problems

Yasir Latif, Guoquan Huang, John Leonard, José Neira, Sparse optimization for robust and efficient loop closing, Robotics and Autonomous Systems, Volume 93, July 2017, Pages 13-26, ISSN 0921-8890,DOI: 10.1016/j.robot.2017.03.016.

It is essential for a robot to be able to detect revisits or loop closures for long-term visual navigation. A key insight explored in this work is that the loop-closing event inherently occurs sparsely, i.e., the image currently being taken matches with only a small subset (if any) of previous images. Based on this observation, we formulate the problem of loop-closure detection as a sparse, convex
ℓ 1 -minimization problem. By leveraging fast convex optimization techniques, we are able to efficiently find loop closures, thus enabling real-time robot navigation. This novel formulation requires no offline dictionary learning, as required by most existing approaches, and thus allows online incremental operation. Our approach ensures a unique hypothesis by choosing only a single globally optimal match when making a loop-closure decision. Furthermore, the proposed formulation enjoys a flexible representation with no restriction imposed on how images should be represented, while requiring only that the representations are “close” to each other when the corresponding images are visually similar. The proposed algorithm is validated extensively using real-world datasets.

Incorporating spatial info into the symbolic (bag-of-words) info used for loop closure detection

Nishant Kejriwal, Swagat Kumar, Tomohiro Shibata, High performance loop closure detection using bag of word pairs, Robotics and Autonomous Systems, Volume 77, March 2016, Pages 55-65, ISSN 0921-8890, DOI: 10.1016/j.robot.2015.12.003.

In this paper, we look into the problem of loop closure detection in topological mapping. The bag of words (BoW) is a popular approach which is fast and easy to implement, but suffers from perceptual aliasing, primarily due to vector quantization. We propose to overcome this limitation by incorporating the spatial co-occurrence information directly into the dictionary itself. This is done by creating an additional dictionary comprising of word pairs, which are formed by using a spatial neighborhood defined based on the scale size of each point feature. Since the word pairs are defined relative to the spatial location of each point feature, they exhibit a directional attribute which is a new finding made in this paper. The proposed approach, called bag of word pairs (BoWP), uses relative spatial co-occurrence of words to overcome the limitations of the conventional BoW methods. Unlike previous methods that use spatial arrangement only as a verification step, the proposed method incorporates spatial information directly into the detection level and thus, influences all stages of decision making. The proposed BoWP method is implemented in an on-line fashion by incorporating some of the popular concepts such as, K-D tree for storing and searching features, Bayesian probabilistic framework for making decisions on loop closures, incremental creation of dictionary and using RANSAC for confirming loop closure for the top candidate. Unlike previous methods, an incremental version of K-D tree implementation is used which prevents rebuilding of tree for every incoming image, thereby reducing the per image computation time considerably. Through experiments on standard datasets it is shown that the proposed methods provide better recall performance than most of the existing methods. This improvement is achieved without making use any geometric information obtained from range sensors or robot odometry. The computational requirements for the algorithm is comparable to that of BoW methods and is shown to be less than the latest state-of-the-art method in this category.