Autonomous Robot Localization Research Articles

Particle Filter Enhanced by CNN-LSTM for Localization of Autonomous Robot in Indoor Environment

Particle filters play a crucial role in indoor localization tasks where GPS or similar sensors are unavailable, owing to their adaptability to dynamic environments and contribution to precise navigation. However, despite their effectiveness in non-Gaussian and non-linear settings, particle filters face challenges such as issues with particle initialization and recovery from scenarios like robot kidnapping. These limitations impede the complete autonomy of robots, a critical aspect of their functionality. This research presents a novel solution that integrates particle filters with a CNN-LSTM network to address these challenges. Leveraging the time-series image processing capabilities of CNN-LSTMs, this architecture aids in particle filter initialization and facilitates recovery from challenging situations. The integration enhances the overall performance and autonomy of robots, making them more efficient in indoor navigation tasks.

Integrating artificial intelligence with SLAM technology for robotic navigation and localization in unknown environments

In the era of advancing technology, unmanned inspection robots have become indispensable for their efficiency, precision, and safety. Key to their autonomous operation is Simultaneous Localization and Mapping (SLAM) technology, which allows robots to navigate and create maps of unknown environments in real-time. This article explores the integration of SLAM with artificial intelligence, highlighting its role in robotic navigation, localization, and obstacle avoidance. Specifically, we delve into SLAM's principles, its implementation with LiDAR technology, and its application in autonomous robot localization. Furthermore, we introduce a collaborative mapping algorithm based on ORB-SLAM3, enhancing map construction efficiency and real-time performance. Through our exploration, we illustrate the transformative potential of SLAM technology, paving the way for safer and more efficient robotic inspection systems across various industries.

SPINS: A structure priors aided inertial navigation system

AbstractWe propose a navigation system combining sensor‐aided inertial navigation and prior‐map‐based localization to improve the stability and accuracy of robot localization in structure‐rich environments. Specifically, we adopt point, line, and plane features in the navigation system to enhance the feature richness in low‐texture environments and improve the localization reliability. We additionally integrate structure prior information of the environments to constrain the localization drifts and improve the accuracy. The prior information is called structure priors and parameterized as low‐dimensional relative distances/angles between different geometric primitives. The localization is formulated as a graph‐based optimization problem that contains sliding‐window‐based variables and factors, including Inertial Measurement Unit, heterogeneous features, and structure priors. A limited number of structure priors are selected based on the information gain to alleviate the computation burden. Finally, the proposed framework is extensively tested on synthetic data, public data sets, and, more importantly, on the real Unmanned Aerial Vehicle flight data obtained from both indoor and outdoor inspection tasks. The results show that the proposed scheme can effectively improve the accuracy and robustness of localization for autonomous robots in civilian applications.

Consistent Localization for Autonomous Robots With Inter-Vehicle GNSS Information Fusion

In this letter, we reveal a new precise localization framework based on inter-vehicle GNSS information fusion (IGIF) for autonomous robot systems. In this framework, we derive a new hybrid GNSS filter (HGF) using information from both onboard sensors and the inter-vehicle network, and a consistency checking (CC) method to improve the system reliability. The main novelty of the proposed method is the fusion of ionosphere-free (IF) and double differenced (DD) combinations to improve the global positioning accuracy of robots and the two-level probabilistic detection of erroneous inter-vehicle measurements. A real-world experiment was conducted and it turned out that our method provided more consistent, continuous, and accurate position estimation than the state-of-the-art methods while only requiring low communication bandwidth.

Bayesian In-Situ Calibration of Multiport Antennas for DoA Estimation: Theory and Measurements

The DoA of radio waves is used for many applications, e.g. the localization of autonomous robots and smart vehicles. Estimating the DoA is possible with a multiport antenna, e.g. an antenna array or a multi-mode antenna (MMA). In practice, DoA estimation performance decisively depends on accurate knowledge of the antenna response, which makes antenna calibration vital. As the antenna surroundings influence its response, it is necessary to measure the entire device with installed antenna to obtain the installed antenna response. Antenna calibration is often done in a dedicated measurement chamber, which can be inconvenient and costly, especially for larger devices. Thus, auto- and in-situ calibration methods aim at making antenna calibration in a measurement chamber redundant. However, existing auto- and in-situ calibration methods are restricted to certain antenna types and certain calibrations. In this paper, we propose a Bayesian in-situ calibration algorithm based on a maximum a posteriori (MAP) estimator, which is suitable for arbitrary multiport antennas. The algorithm uses received signals from a transmitter, noisy external DoA observations, takes multipath propagation into account and does not require synchronization. Furthermore, we take an estimation theoretic perspective and provide an in-depth theoretical discussion of in-situ antenna calibration in unknown propagation conditions based on Bayesian information and the Bayesian Cramér-Rao bound (BCRB). Extensive simulations show that the proposed algorithm operates close to the BCRB and the achieved DoA estimation performance asymptotically approaches the case of a perfectly known antenna response. Finally, we provide an experimental validation, where we calibrate the antenna on a robotic rover and evaluate the DoA estimation performance using measurement data. With the proposed in-situ antenna calibration algorithm, DoA estimation performance is considerably improved compared to using an antenna response obtained by simulation or in a measurement chamber.

Monocular Visual-Inertial Odometry for Agricultural Environments

The accuracy of autonomous robot localization using a monocular visual-inertial odometry system (VIO) is significantly reduced in an agricultural environment compared to an urban and indoor environment due to the unstructured scenes with unstable features, variation of light conditions, and rugged terrain. To address those challenges, we propose a monocular VIO system with modifications to the existing state-of-the-art monocular VIO system VINS-mono. Three modules of VINS-mono have improved in this work: vision processing front end, pose optimization, failure detection and recovery. In the vision processing front-end module, we proposed a keyframe selection algorithm base on vertical movement smoothness verification to prevent the rapid loss of feature tracks. In the pose optimization module, some feature points are removed using a depth-limited approach to improve the efficiency and accuracy of trajectory tracking. In the failure detection and recovery module, we propose a failure recovery method using the old trajectory poses, which can recover the trajectory tracking process interrupted by the failure and alleviate the orientation drift. Experiments on the Rosario dataset have shown that our system outperforms VINS-mono, and the absolute trajectory error in the agricultural environment can be at least reduced by 69% compared with the VINS-mono, which can effectively improve the localization accuracy of agriculture robots in the agricultural environment.

Localization of Autonomous Robot in an Urban Area Based on SURF Feature Extraction of Images

An autonomous robot is now an internationally discussed topic to ease the life of humans. Localization and movement are two rudimentary necessities of the autonomous robots before accomplishing any job. So, many researchers have proposed methods of localization using external tools like network connectivity, global positioning system (GPS), etc. However, if these tools are lost, either the movement will be paused, or the robot will be derailed from the actual mission. In these circumstances, the authors propose an approach to localize an autonomous robot in a specific area using the given set of images without external help. The image database has been prepared and kept in the internal memory of robot so that image matching can be done quickly. The localization method has been accomplished using three algorithms: (1) SURF, (2) ICP-BP, and (3) EMD. In the evaluation, SURF has been found better than ICP-BP and EMD in terms of accuracy and elapsed time. The authors believe that the proposed method will add value to other methods using some external tools even when those tools are unavailable.

Cost-effective Indoor Localization for Autonomous Robots using Kinect and WiFi Sensors

Indoor localization has been considered to be the most fundamental problem when it comes to providing a robot with autonomous capabilities. Although many algorithms and sensors have been proposed, none have proven to work perfectly under all situations. Also, in order to improve the localization quality, some approaches use expensive devices either mounted on the robots or attached to the environment that don't naturally belong to human environments. This paper presents a novel approach that combines the benefits of two localization techniques, WiFi and Kinect, into a single algorithm using low-cost sensors. It uses separate Particle Filters (PFs). The WiFi PF gives the global location of the robot using signals of Access Point devices from different parts of the environment while it bounds particles of the Kinect PF, which determines the robot's pose locally. Our algorithm also tackles the Initialization/Kidnapped Robot Problem by detecting divergence on WiFi signals, which starts a localization recovering process. Furthermore, new methods for WiFi mapping and localization are introduced.

Secondary Radar Beacons for Local Ad-Hoc Autonomous Robot Localization Systems.

In this paper, we present a detailed analysis and implementation of secondary radar beacons designed for a local ad-hoc localization and landing system (LAOLa) to support the navigation of autonomous ground and aerial vehicles. We discuss a switched linear feedback network as a virtually coherent oscillator and show how to use it as a secondary radar transponder. Further, we present a signal model for the beat signal of the transponder response in an FMCW radar system, which is more detailed than in previously published papers. An actual transponder realization in the 24 ISM band is presented. Its RF performance was evaluated both in the laboratory and in the field. Finally, we put forward some ideas on how to overcome the range measurement inaccuracy inherent in this transponder concept.

Analysing Willerding’s formula for solving the planar three point resection problem

Abstract The main mathematical problem in the planar three point resection problem of surveying is to find a necessary and sufficient condition that a point on a plane is uniquely determined if the directions from that point to three given points are known only up to an unknown offset. It will be shown in this paper that such a condition is the applicability of Willerding’s formula for computing the point in question, and that this condition is equivalent to the well-known necessary condition that the four considered points are not located on a circle or a line. Because of this result, the easy evaluation and the short derivation of Willerding’s formula, it can be expected, that this formula will be widely used in surveying, and also for autonomous robot localization.

Global Localization of Autonomous Robots in Forest Environments

A method for the autonomous geolocation of ground vehicles in forest environments is presented. The method provides an estimate of the global horizontal position of a vehicle strictly based on finding a geometric match between a map of observed tree stems, scanned in 3D by Light Detection and Ranging (lidar) sensors onboard the vehicle, to another stem map generated from the structure of tree crowns analyzed from high resolution aerial orthoimagery of the forest canopy. The method has been tested with real-world data and has been able to estimate vehicle geoposition with an average error of less than 2m. The method has two key properties that are significant: (a) The algorithm does not require a priori knowledge of the area surrounding the robot, and (b) Based on estimated vehicle state, it uses the geometry of detected tree stems as the only input to determine horizontal geoposition.

Autonomous Robot Localization Using WiFi Fingerprinting

Autonomous Robot Localization Using WiFi Fingerprinting

Simultaneous sensing of location and mapping for autonomous robots

PurposeTo describe an innovative new technology, simultaneous localization and mapping (SLAM), for creating maps and localization for autonomous robot operations.Design/methodology/approachA series of stereo cameras gather images of a 3D area. A software package then analyzes the images to create a 3D map of the area and to monitor the real‐time location of the robot in the area.FindingsThe SLAM approach will open up the range of possible autonomous mobile robot applications, especially service and non‐industrial applications.Practical implicationsThe scalable SLAM system and algorithms will enable improved performance in current as well as new mobile autonomous robot applications.Originality/valueUsers now have a totally new means for creating “maps” for autonomous mobile robotic activities.

The Reverse Monte Carlo localization algorithm

Global localization is a very fundamental and challenging problem in Robotic Soccer. Here, the main aim is to find the best method which is very robust and fast and requires less computational resources and memory compared to similar approaches and is precise enough for robot soccer games and technical challenges. In this work, the Reverse Monte Carlo localization (R-MCL) method is introduced. The algorithm is designed for fast, precise and robust global localization of autonomous robots in the robotic soccer domain, to overcome the uncertainties in the sensors, environment and the motion model. R-MCL is a hybrid method based on Markov localization (ML) and Monte Carlo localization (MCL), where the ML based module finds the region where the robot should be and the MCL based part predicts the geometrical location with high precision by selecting samples in this region. It is called Reverse since the MCL routine is applied in a reverse manner in this algorithm. In this work, this method is tested on a challenging data set that is used by many other researchers and compared in terms of error rate against different levels of noise, and sparsity. Additionally, the time required to recover from kidnapping and the processing time of the methods are tested and compared. According to the test results R-MCL is a considerable method against high sparsity and noise. It is preferable when its recovery from kidnapping and processing times are considered. It gives robust and fast but relatively coarse position estimations against imprecise and inadequate perceptions, and coarse action data, including regular misplacements, and false perceptions.

Interpretation of Ultrasonic Readings for Autonomous Robot Localization

The work described in this paper is a contribution to providing mobility aid for people with motor disability. It constitutes a part of the VAHM project which aims to design a smart powered wheelchair able to control its displacements in a known environment. Original methods established for the static localisation of the wheelchair using readings provided by a belt of 14 ultrasonic sensors is presented. This approach is based on a classical matching of occupancy grids. Yet because of the presence of the person on the wheelchair any complementary movement intended to obtain additional measures is impossible. That is why our study is centred on the search for the best way to represent ultrasound measures, to model environment and to define the matching criterion in order to mitigate the imperfections of ultrasonic sensors. The method thus developed is implemented on our prototype. Examples are given of the tests carried out in real-life conditions in a typical environment consisting of a flat recreated in our laboratory. The results obtained using real and simulated readings show that the approach is reliable and fitted to our project.