Partial-Map-Based Monte Carlo Localization in Architectural Floor Plans

Modern automation demands mobile robots to be robustly localized in complex scenarios. Current localization systems typically use maps that require to be built and interpreted by experienced operators, increasing deployment costs as well as reducing the adaptability of robots to rearrangements in the environment. In contrast, architectural floor plans can be easily understood by non-expert users and typically represent only the non-rearrangeable parts of buildings. In this paper we propose a system for robot localization in architectural CAD drawings. Our method employs a simultaneous localization and mapping approach to online augment the floor plan with a map represented as a pose-graph with LiDAR measurements. Whenever the environment is accurately mapped in the vicinity of the robot, we use the graph to perform relative localization. We thoroughly evaluate our system in challenging real-world scenarios. Experiments demonstrate that our method is able to robustly track the robot pose even when the floor plan shows major discrepancies from the real-world. We show that our system achieves sub-centimeter accuracy and is suitable for real-time application.

Reliability is a key factor for realizing safety guarantee of fully autonomous robot systems. In this paper, we focus on reliability in mobile robot localization. Monte Carlo localization (MCL) is widely used for mobile robot localization. However, it is still difficult to guarantee its safety because there are no methods determining reliability for MCL estimate. This paper presents a novel localization framework that enables robust localization, reliability estimation, and quick relocalization, simultaneously. The presented method can be implemented using a similar estimation manner to that of MCL. The method can increase localization robustness to environment changes by estimating known and unknown obstacles while performing localization; however, localization failure of course occurs by unanticipated errors. The method also includes a reliability estimation function that enables a robot to know whether localization has failed. Additionally, the method can seamlessly integrate a global localization method via importance sampling. Consequently, quick relocalization from a failure state can be realized while mitigating noisy influence of global localization. We conduct three types of experiments using wheeled mobile robots equipped with a two‐dimensional LiDAR. Results show that reliable MCL that performs robust localization, self‐failure detection, and quick failure recovery can be realized.

Robust robot localization with visually adaptive consensus filters in dynamic corridor environments

As a probabilistic localization algorithm, Monte Carlo localization (MCL) method has been widely used for mobile robot localization over the past decade. In this paper, an extended MCL method (EMCL) is developed by incorporating two different resampling processes, namely importance resampling and sensor-based resampling, to conventional MCL for improvement of localization performance. Different resampling processes are utilized based on a matching of sample distribution and observations. Two additional processes for validating over-convergence and uniformity are introduced for examination of such matching. A visual based EMCL is further implemented using a triangulation-based resampling from visual features recognized by Bayesian networks. Experiments are conducted to demonstrate the validity of the proposed approach

We propose the robust 2D localization applies an Auxiliary Particle Filter (APF) to Monte Carlo Localization (MCL). Urban environments have fewer landmarks than two-dimensional (2D) indoor maps for efficiently finding a unique location. Localization using MCL have the problem that few landmarks pose divergence of the particles of MCL. We use APF for MCL, because APF continues resampling until convergence particle occurs in one localization step. Another problem with 2D urban mapping is that of data association posed by three-dimensional (3D) surfaces. Pitching and rolling may, for example, adversely affect 2D scan-data metrics due to 3D surfaces, causing mismatching data association in 2D maps. We therefore use a Laplacian filter for 2D grid maps. Experimental results show that our localization method is more highly stable in urban environments than MCL.

Recently, indoor space construction information has been actively carried out primarily in large buildings and in underground facilities. However, the building of this data was done by only a handful of people, and it was a time- and money-intensive task. Therefore, the technology of automatically extracting a wall and constructing a 3D model from architectural floor plans was developed. Complete automation is still limited by accuracy issues, and only a few sets of floor plan data to which the technology can be applied exist. In addition, it is difficult to extract complicated walls and their thickness to build the wall-junction structure of indoor spatial information, which requires significant topological information in the automation process. In this paper, we propose an automatic method of extracting the wall from an architectural floor plan suitable for the restoration of the indoor spatial information according to the indoor spatial information standard.

Inertial navigation systems (INS) are composed of inertial sensors, such as accelerometers and gyroscopes. An INS updates its orientation and position automatically; it has an acceptable stability over the short term, however this stability deteriorates over time. Odometry, used to estimate the position of a mobile robot, employs encoders attached to the robot’s wheels. However, errors occur caused by the integrative nature of the rotating speed and the slippage between the wheel and the ground. In this paper, we discuss mobile robot position estimation without using external signals in indoor environments. In order to achieve optimal solutions, a Kalman filter that estimates the orientation and velocity of mobile robots has been designed. The proposed system combines INS and odometry and delivers more accurate position information than standalone odometry.

This paper presents an automatic system for analyzing and labeling architectural floor plans. In order to detect the locations of the rooms, the proposed systems extracts both, structural and semantic information from given floor plans. Furthermore, OCR is applied on the text layer to retrieve the meaningful room labeling. Finally, a novel post-processing is proposed to split rooms into several sub-regions if several semantic rooms share the same physical room. Our fully automatic system is evaluated on a publicly available dataset of architectural floor plans. In our experiments, we could clearly outperform other state-of-the-art approaches for room detection.

Global localization and kidnapping are two challenging problems in robot localization. The popular method, Monte Carlo Localization (MCL) addresses the problem by iteratively updating a set of particles with a “sampling-weighting” loop. Sampling is decisive to the performance of MCL [1]. However, traditional MCL can only sample from a uniform distribution over the state space. Although variants of MCL propose different sampling models, they fail to provide an accurate distribution or generalize across scenes. To better deal with these problems, we present a distribution proposal model named Deep Samplable Observation Model (DSOM). DSOM takes a map and a 2D laser scan as inputs and outputs a conditional multimodal probability distribution of the pose, making the samples more focusing on the regions with higher likelihood. With such samples, the convergence is expected to be more effective and efficient. Considering that the learning-based sampling model may fail to capture the accurate pose sometimes, we furthermore propose the Adaptive Mixture MCL (AdaM MCL), which deploys a trusty mechanism to adaptively select updating mode for each particle to tolerate this situation. Equipped with DSOM, AdaM MCL can achieve more accurate estimation, faster convergence and better scalability than previous methods in both synthetic and real scenes. Even in real environments with long-term changes, AdaM MCL is able to localize the robot using DSOM trained only by simulation observations from a SLAM map or a blueprint map. Source code for this paper is available here: https://github.com/Runjian-Chen/AdaM_MCL.

The paper describes an intelligent computer aided architectural design system (ICAAD) called ICADS. ICADS encapsulates different types of design knowledge into independent "critic" modules. Each "critic" module possesses expertise in evaluating an architect's work in different areas of architectural design and can offer expert advice when needed. This research focuses on the representation of spatial information encoded in architectural floor plans and the representation of expert design knowledge. Described in the paper is our research in designing and developing two particular "critic" modules. The first module, FPDX, checks a residential apartment floor plan, verifies that the plan meets a set of government regulations, and offers suggestions for floor plan changes if regulations are not met. The second module, IDX, analyzes room and furniture layout according to a set of interior design guidelines and offers ideas on how furniture should be moved if the placement does not follow good design principles.

Most studies on map segmentation and recognition are focused on architectural floor plans, while there are very few analyses of shopping mall plans. The objective of the work is to accurately segment and recognize the shopping mall plan, obtaining location and semantic information for each room via segmentation and recognition. This work can be used in other applications such as indoor robot navigation, building area and location analysis, and three-dimensional reconstruction. First, we identify and match the catalog of a mall floor plan to obtain matching text, and then we use the two-stage region growth method we proposed to segment the preprocessed floor plan. The room number is then obtained by sending each segmented room section to an OCR (optical character recognition) system for identification. Finally, the system retrieves the matching text to match the room number in order to obtain the room name, and outputs the needed room location and semantic information. It is considered a successful detection when a room region can be successfully segmented and identified. The proposed method is evaluated on a dataset including 1340 rooms. Experimental results show that the accuracy of room segmentation is 92.54%, and the accuracy of room recognition is 90.56%. The total detection accuracy is 83.81%.

Mobile robot localization is a very important problem in robotics as most robot's tasks need the positional information. Monte Carlo Localization(MCL) is one of the most popular and efficient localization algorithms for mobile robot localization. MCL algorithm represents a robot's pose by a set of weighted particles. In order to further improve the performance of MCL, many extensions have been proposed. In this paper, we proposed an algorithm called dynamic size MCL, an extension of MCL. We incorporate the clustering approach into traditional MCL. With the help of clustering information, our algorithm could reduce the number of particles during the process of localization, which lower the computational cost. Experimental results demonstrate the effectiveness of the proposed method.

An improved 2D weighted centroid localization algorithm for Wireless Sensor Networks (WSNs) is proposed. Compared with most previous approaches, the improved algorithm uses a optimized threshold to restrict the number of anchors involved. Simulation results show that our improved weighted centroid localization algorithm increased the positioning accuracy by at least 15% than traditional Monte Carlo Localization (MCL) algorithm. Furthermore, the algorithm is extended from 2D to 3D and comparison with MCL method both in accuracy and in efficiency are carried out as well, showing that 3D-WCL algorithm has a high accuracy about 13% localization average error significantly with little hardware cost, which is suitable for practical application, such as medical surveillance in hospital etc.

In this paper, we propose a LiDAR‐based robot localization method in a complex oil and gas environment. Localization is achieved in six degrees of freedom (DoF) thanks to a particle filter framework. A new time‐efficient likelihood function, based on a precalculated three‐dimensional likelihood field, is introduced. Experiments are carried out in real environments and their digitized point clouds. Six DoF real‐time localization is achieved with spatial and angular errors of less than 2.5 cm and 1°, respectively, in a real environment of . The proposed approach focuses on real‐time performance on embedded platforms. It enabled the Vikings team to win the first two ARGOS Challenge contests.

Mobile robot localization is the task of determining a robot's pose in a known environment, which is one of the most important problems in mobile robotics. The state-of-the-art Monte Carlo Localization (MCL) algorithm requires a large amount of particles and thus converges slowly. Also, MCL performs poorly in low noise sensor input.