Odometry Data Research Articles

Automated construction of cognitive maps with visual predictive coding

Humans construct internal cognitive maps of their environment directly from sensory inputs without access to a system of explicit coordinates or distance measurements. Although machine learning algorithms like simultaneous localization and mapping utilize specialized inference procedures to identify visual features and construct spatial maps from visual and odometry data, the general nature of cognitive maps in the brain suggests a unified mapping algorithmic strategy that can generalize to auditory, tactile and linguistic inputs. Here we demonstrate that predictive coding provides a natural and versatile neural network algorithm for constructing spatial maps using sensory data. We introduce a framework in which an agent navigates a virtual environment while engaging in visual predictive coding using a self-attention-equipped convolutional neural network. While learning a next-image prediction task, the agent automatically constructs an internal representation of the environment that quantitatively reflects spatial distances. The internal map enables the agent to pinpoint its location relative to landmarks using only visual information.The predictive coding network generates a vectorized encoding of the environment that supports vector navigation, where individual latent space units delineate localized, overlapping neighbourhoods in the environment. Broadly, our work introduces predictive coding as a unified algorithmic framework for constructing cognitive maps that can naturally extend to the mapping of auditory, sensorimotor and linguistic inputs.

Industrial Environment Multi-Sensor Dataset for Vehicle Indoor Tracking with Wi-Fi, Inertial and Odometry Data

This paper describes a dataset collected in an industrial setting using a mobile unit resembling an industrial vehicle equipped with several sensors. Wi-Fi interfaces collect signals from available Access Points (APs), while motion sensors collect data regarding the mobile unit’s movement (orientation and displacement). The distinctive features of this dataset include synchronous data collection from multiple sensors, such as Wi-Fi data acquired from multiple interfaces (including a radio map), orientation provided by two low-cost Inertial Measurement Unit (IMU) sensors, and displacement (travelled distance) measured by an absolute encoder attached to the mobile unit’s wheel. Accurate ground-truth information was determined using a computer vision approach that recorded timestamps as the mobile unit passed through reference locations. We assessed the quality of the proposed dataset by applying baseline methods for dead reckoning and Wi-Fi fingerprinting. The average positioning error for simple dead reckoning, without using any other absolute positioning technique, is 8.25 m and 11.66 m for IMU1 and IMU2, respectively. The average positioning error for simple Wi-Fi fingerprinting is 2.19 m when combining the RSSI information from five Wi-Fi interfaces. This dataset contributes to the fields of Industry 4.0 and mobile sensing, providing researchers with a resource to develop, test, and evaluate indoor tracking solutions for industrial vehicles.

MoMo: Mouse-Based Motion Planning for Optimized Grasping to Declutter Objects Using a Mobile Robotic Manipulator

The aim of this study is to develop a cost-effective and efficient mobile robotic manipulator designed for decluttering objects in both domestic and industrial settings. To accomplish this objective, we implemented a deep learning approach utilizing YOLO for accurate object detection. In addition, we incorporated inverse kinematics to facilitate the precise positioning, placing, and movement of the robotic arms toward the desired object location. To enhance the robot’s navigational capabilities within the environment, we devised an innovative algorithm named “MoMo”, which effectively utilizes odometry data. Through careful integration of these algorithms, our goal is to optimize grasp planning for object decluttering while simultaneously reducing the computational burden and associated costs of such systems. During the experimentation phase, the developed mobile robotic manipulator, following the MoMo path planning strategy, exhibited an impressive average path length coverage of 421.04 cm after completing 10 navigation trials. This performance surpassed that of other state-of-the-art path planning algorithms in reaching the target. Additionally, the MoMo strategy demonstrated superior efficiency, achieving an average coverage time of just 16.84 s, outperforming alternative methods.

Bionic visual navigation model for enhanced template matching and loop closing in challenging lighting environments

In recent years, various computational models have been proposed in neuroscience, but only a few have been utilized in robot visual navigation through engineering approaches. Besides, these engineering methods lack environmental adaptability, especially in the aspect of visual place recognition in weak or uncontrollable light fluctuations. To address this issue and enhance the performance of visual template matching and map loop closure detection in challenging lighting environments, this paper proposed a bionic visual navigation model that combines two neural network architectures, the Pulse Coupled Neural Network (PCNN) and Continuous Attractor Neural Network (CANN). In our navigation model, the visual features of the environments are encoded as temporal information by the spiking model and input into the local view cells for visual template matching. Additionally, we utilize the pose cells network, which incorporates the similarity between current and previous templates, along with odometry data, to encode spatial information and store it as an experience map. To validate the effectiveness of the proposed model, we conducted evaluations on datasets collected within our library, campus, and an open-source office dataset. The experimental results reveal that our algorithm increases the F1-score of template matching by approximately 10.5%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$ \\% $$\\end{document}, 35.8%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$ \\% $$\\end{document}, 61.7%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$ \\% $$\\end{document}, and 1.9%\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$ \\% $$\\end{document} in each dataset, compared to the conventional RatSLAM method. Furthermore, our algorithm generates a more accurate map that closely correlates with the real-world trajectory without compromising on computation time. The results suggest that our bionic visual navigation model is reliable for both standard and extreme lighting conditions.

Trajectory planning approach for autonomous electric bus in dynamic environment

Path planning is a challenging task to achieve vehicle autonomy, and it becomes even more difficult in handling dynamic situations and big vehicle sizes. Its performance is affected by the accuracy of mapping and vehicle localization. Many path planning algorithms have been developed to address static environments, however, these do not accurately reflect real-world scenarios which are dynamic in nature. This paper introduces a new planning architecture consisting of a global planner that plans a path in an occupancy grid map generated using LIDAR and odometry data. The local planner uses this global plan and plans a local plan on a section of the map that takes into consideration dynamic obstacles. Both the local and global plans are used to calculate the optimum velocity profile, providing a feasible and comfortable trajectory. The trajectory is then fed to a predictive Stanley controller, and both the calculated steering angle and the optimum velocity are achieved by low-level controllers. The performance of the proposed architecture was tested on a golf bus in a constrained environment and compared to different manual driving attitudes using a key performance indicator. The results of the proposed architecture show that it has a better KPI, with an average of 37% better than manual driving.

Design and Implementation of an Integrated Control System for Omnidirectional Mobile Robots in Industrial Logistics

The integration of intelligent robots in industrial production processes has the potential to significantly enhance efficiency and reduce human adversity. However, for such robots to effectively operate within human environments, it is critical that they possess an adequate understanding of their surroundings and are able to navigate through narrow aisles while avoiding both stationary and moving obstacles. In this research study, an omnidirectional automotive mobile robot has been designed for the purpose of performing industrial logistics tasks within heavy traffic and dynamic environments. A control system has been developed, which incorporates both high-level and low-level algorithms, and a graphical interface has been introduced for each control system. A highly efficient micro-controller, namely myRIO, has been utilized as the low-level computer to control the motors with an appropriate level of accuracy and robustness. Additionally, a Raspberry Pi 4, in conjunction with a remote PC, has been utilized for high-level decision making, such as mapping the experimental environment, path planning, and localization, through the utilization of multiple Lidar sensors, IMU, and odometry data generated by wheel encoders. In terms of software programming, LabVIEW has been employed for the low-level computer, and the Robot Operating System (ROS) has been utilized for the design of the higher-level software architecture. The proposed techniques discussed in this paper provide a solution for the development of medium- and large-category omnidirectional mobile robots with autonomous navigation and mapping capabilities.

GNSS Jamming Detection and Exclusion for Trustworthy Virtual Balise Capture in Satellite-Based Train Control

Jamming to satellite navigation signals has become a major threat to the safety critical train control systems using Global Navigation Satellite System (GNSS), where the Virtual Balise (VB) concept enables a specification-compatible solution to utilize GNSS and reduce the physical Balises. This paper presents a novel GNSS jamming detection and exclusion solution to achieve trustworthy capture of VBs. An advanced architecture of Virtual Balise Reader (VBR) is proposed by integrating an Interference Qualifier Module (IQM) into the conventional GNSS-enabled VB framework. To enhance the trustworthiness of the VB capture, the IQM detects and identifies the existence of interference by examining the residuals between the real and predicted GNSS pseudo-ranges, which are generated by the odometry data and the trackmap. A discrimination test approach is utilized to evaluate the availability of raw satellite measurements according to the statistical analysis. We embed the performance indicator for each pseudo-range into the VB capture logic to isolate the degraded measurements and, meantime, adopt an adaptive Capture Interval Limit (CIL) to avoid the unexpected missed capture. Data sets from a GNSS jamming injection-based test platform are used for comparative studies, demonstrating the necessity and superiority of the proposed solution in improving the interference protection capability and trustworthiness of VB capture under the GNSS jamming environment.

A 3D LiDAR odometry for UGVs using coarse-to-fine deep scene flow estimation

Light detection and ranging (LiDAR) odometry plays a crucial role in autonomous mobile robots and unmanned ground vehicles (UGVs). This paper presents a deep learning–based odometry system using two successive three-dimensional (3D) point clouds to estimate their scene flow and then predict their relative pose. The network consumes continuous 3D point clouds directly and outputs their scene flow and uncertain mask in a coarse-to-fine fashion. A pose estimation layer without trainable parameters is designed to compute the pose with the scene flow. We also introduce a scan-to-map optimization algorithm to enhance the robustness and accuracy of the system. Our experiments on the KITTI odometry data set and our campus data set demonstrate the effectiveness of the proposed deep learning–based point cloud odometry.

Construction and benchmark of an autonomous tracked mobile robot system

Robots require a certain set of skills to perceive and analyse the environment and act accordingly. For tracked mobile robots getting good odometry data from sensory information is a challenging key prerequisite to perform in an unstructured dynamic environment, thus an essential issue in the tracked mobile robotics domain. In this article, we construct a ROS-based tracked mobile robot system taking the Jaguar V4 mobile robot as the base platform. On which several visual odometry solutions based on different cameras and methods (Intel RealSense T265, Zed camera, RTAB-Map RGBD) are integrated and benchmark comparison is performed. Analysis of new challenges faced by different methods while applied on a tracked vehicle as well as recommendations and conclusions are presented. Intel RealSense T265 solution proved to perform well in uncertain conditions which involves bounded vibrations and low lighting conditions with low latency, which result in good map generation. Further evaluations with a path planning algorithm and Intel RealSense T265 were conducted to test the effect of the robot’s motion profiles on odometry data accuracy.

Real-Time Map Matching with a Backtracking Particle Filter Using Geospatial Analysis.

Inertial odometry is a typical localization method that is widely and easily accessible in many devices. Pedestrian positioning can benefit from this approach based on inertial measurement unit (IMU) values embedded in smartphones. Fitting the inertial odometry outputs, namely step length and step heading of a human for instance, with spatial information is an ubiquitous way to correct for the cumulative noises. This so-called map-matching process can be achieved in several ways. In this paper, a novel real-time map-matching approach was developed, using a backtracking particle filter that benefits from the implemented geospatial analysis, which reduces the complexity of spatial queries and provides flexibility in the use of different kinds of spatial constraints. The goal was to generalize the algorithm to permit the use of any kind of odometry data calculated by different sensors and approaches as the input. Further research, development, and comparisons have been done by the easy implementation of different spatial constraints and use cases due to the modular structure. Additionally, a simple map-based optimization using transition areas between floors has been developed. The developed algorithm could achieve accuracies of up to 3 m at approximately the 90th percentile for two different experiments in a complex building structure.

Flexible and Resource-Efficient Multi-Robot Collaborative Visual-Inertial-Range Localization

In multi-robot systems, two important research problems are relative localization between the robots and global localization of all robots in a common frame. Traditional methods rely on detecting inter and intra-robot loop closures, which can be restrictive operation-wise since the robot must form loops. Ultra-wideband sensors, which provide direct distance measurements and robot ID, can replace loop closures in many applications. However, existing research on UWB-aided multi-robot state estimation often ignores the odometry drift which leads to inaccurate global position in the long run. In this work, we present a UWB-aided multi-robot localization system that does not rely on loop closure (flexible) and only requires odometry data from neighbors (resource-efficient). We propose a two-stage approach: 1) with a long sliding window, the relative transformation is refined based on range and odometry data, 2) onboard visual-inertial-range data are tightly fused in a short-term sliding window to provide more accurate local and global estimates. Simulation and real-life experiments with two quadrotors show that the system as a whole outperforms previous approaches as well as its individual parts.

Human localization in robotized warehouses based on stereo odometry and ground-marker fusion

Modern logistic solutions for large warehouses consist of a fleet of robots that transfer goods, move racks, and perform other physically difficult and repetitive tasks. The shopfloor is usually enclosed with a safety fence and if a human needs to enter the warehouse all the robots are stopped, as opposed to only the ones in the most immediate vicinity of the human, thus significantly limiting the warehouse efficiency. To tackle this challenge, an integrated safety system is needed with human localization as one of its essential components. In this paper, we propose a novel human localization method for robotized warehouses that is based on a suite of wearable visual sensors installed on a vest worn by humans. The proposed method does not require any modifications of the warehouse environment and relies on the already existing infrastructure. Specifically, we estimate the human location by fusing stereo visual-inertial odometry data and distances to the known absolute poses of the detected ground-markers which robots use for their localization. Fusion is performed by building a pose graph, where we treat estimated human poses relative to markers as graph nodes and odometry estimates as graph edges. We conducted extensive laboratory and warehouse facility experiments, where we tested the reliability and accuracy of the proposed method and compared its performance to a state-of-the-art visual SLAM solution, namely ORB-SLAM2. The results indicate that our method can track absolute position in real-time and has competitive accuracy with respect to ORB-SLAM2, while ensuring higher localization reliability when faced with structural changes in the environment. Furthermore, we provide publicly the experimental datasets to the research community.

A UHF-RFID Multi-Antenna Sensor Fusion Enables Item and Robot Localization

This paper describes a UHF-RFID robotic system for tagged-item inventory and localization. A mobile robot is equipped with wheeled rotary encoders and a UHF-RFID reader connected to multiple antennas. At first, the robot reconstructs its trajectory by exploiting a sensor-fusion method combining odometry data with phase data gathered by on-board antennas from an infrastructure of passive reference tags. Then, it leverages its reconstructed trajectory to localize target tags placed at unknown locations through a multi-antenna synthetic-aperture-radar (SAR) approach. A Particle Swarm Optimization is applied to speed up the position estimation. An experimental campaign conducted in an office environment is presented to verify the system features and feasibility. The performance of the proposed method for both tag localization and robot self-localization is compared with respect to the case of trajectories reconstructed only by odometry data or through a commercial Laser Range Finder mounted on the robot. Particularly, the effect of the cumulated drift of the estimated trajectory on the tag localization performance is investigated.

A Cost-Efficient Hybrid UHF RFID and Odometry-Based Mobile Robot Self-Localization Technique With Centimeter Precision

This paper presents an indoor localization approach that determines the absolute position of a mobile robot platform with centimeter precision by fusing RFID localization results based on cost-effective, standard passive UHF RFID technology with the robot’s odometry data. The mobile robot platform is equipped with a multistatic UHF RFID interrogation system, and several RFID tags are arbitrarily placed within the localization environment, serving as landmarks. The RFID localization concept is based on phase evaluations. To overcome the problem of ambiguous position estimations due to the 2π-phase ambiguity of the RFID signal phase and mitigate the linearization problem of the nonlinear system, a novel algorithm based on an iterative multihypothesis Kalman filter is introduced. A realistic simulation setup is developed to validate the proposed filter algorithm. By tracking a UHF RFID-equipped mobile robot platform in a real-world office environment, the proposed approach is also practically tested in terms of real-time capability, everyday suitability, and multipath resistance. Given that centimeter precision is only achieved in environments with weak multipath propagation, the RFID localization results are fused with odometry data provided by the robot. This effectively compensates for offset and drift in the odometry sensor, achieving a root-mean-square localization error of 2.7 cm.

Probabilistic Appearance-Invariant Topometric Localization With New Place Awareness

Probabilistic state-estimation approaches offer a principled foundation for designing localization systems, because they naturally integrate sequences of imperfect motion and exteroceptive sensor data. Recently, probabilistic localization systems utilizing appearance-invariant visual place recognition (VPR) methods as the primary exteroceptive sensor have demonstrated state-of-the-art performance in the presence of substantial appearance change. However, existing systems 1) do not fully utilize odometry data within the motion models, and 2) are unable to handle route deviations, due to the assumption that query traverses exactly repeat the mapping traverse. To address these shortcomings, we present a new probabilistic topometric localization system which incorporates full 3-dof odometry into the motion model and furthermore, adds an "off-map" state within the state-estimation framework, allowing query traverses which feature significant route detours from the reference map to be successfully localized. We perform extensive evaluation on multiple query traverses from the Oxford RobotCar dataset exhibiting both significant appearance change and deviations from routes previously traversed. In particular, we evaluate performance on two practically relevant localization tasks: loop closure detection and global localization. Our approach achieves major performance improvements over both existing and improved state-of-the-art systems.

Robust Approach to Supervised Deep Neural Network Training for Real-Time Object Classification in Cluttered Indoor Environment

Unmanned aerial vehicles (UAVs) are being widely utilized for various missions: in both civilian and military sectors. Many of these missions demand UAVs to acquire artificial intelligence about the environments they are navigating in. This perception can be realized by training a computing machine to classify objects in the environment. One of the well known machine training approaches is supervised deep learning, which enables a machine to classify objects. However, supervised deep learning comes with huge sacrifice in terms of time and computational resources. Collecting big input data, pre-training processes, such as labeling training data, and the need for a high performance computer for training are some of the challenges that supervised deep learning poses. To address these setbacks, this study proposes mission specific input data augmentation techniques and the design of light-weight deep neural network architecture that is capable of real-time object classification. Semi-direct visual odometry (SVO) data of augmented images are used to train the network for object classification. Ten classes of 10,000 different images in each class were used as input data where 80% were for training the network and the remaining 20% were used for network validation. For the optimization of the designed deep neural network, a sequential gradient descent algorithm was implemented. This algorithm has the advantage of handling redundancy in the data more efficiently than other algorithms.

Augmentation of Mapping and Autonomous Navigation for Hexapod Robots by using a Visual Inertial System

In our previous work, we focused on the development of Razbot, a hexapod platform, and achieving its autonomy using the Robot Operating System (ROS). Therein, we also explained the difficulties faced with Razbot while trying to execute complex operations autonomously. Hence, in this work we address the problem and make an attempt at improving the odometry measurements of Razbot by introducing a Visual Inertial System (VIS) that includes a pair of synchronized monocular camera and inertial measurement unit which makes it capable of mapping the environment and autonomously navigating in it with increased effectiveness. To introduce the VIS, we made minor modifications in the components used and the control scheme of Razbot. In this work, we take help of the Extended Kalman Filter (EKF) for the fusion of odometry sources. Further, we also analyse the accuracy of the fused odometry by comparing the same and evaluating the errors or drifts in odometry before and after introducing the VIS in Razbot. Based on these results, we make a conclusion that the odometry source from VIS can significantly lower errors and drifts in fused odometry data, thus improving autonomous navigation and mapping of the hexapod robot. The implementation of this system will allow Razbot to be used in even more varied applications than those specified earlier.

Ultrawideband (UWB)-based precise short-range localization for wireless power transfer to electric vehicles in parking environments.

As the necessity of wireless charging to support the popularization of electric vehicles (EVs) emerges, the development of a wireless power transfer (WPT) system for EV wireless charging is rapidly progressing. The WPT system requires alignment between the transmitter coils installed on the parking lot floor and the receiver coils in the vehicle. To automatically align the two sets of coils, the WPT system needs a localization technology that can precisely estimate the vehicle’s pose in real time. This paper proposes a novel short-range precise localization method based on ultrawideband (UWB) modules for application to WPT systems. The UWB module is widely used as a localization sensor because it has a high accuracy while using low power. In this paper, the minimum number of UWB modules consisting of two UWB anchors and two UWB tags that can determine the vehicle’s pose is derived through mathematical analysis. The proposed localization algorithm determines the vehicle’s initial pose by globally optimizing the collected UWB distance measurements and estimates the vehicle’s pose by fusing the vehicle’s wheel odometry data and the UWB distance measurements. To verify the performance of the proposed UWB-based localization method, we perform various simulations and real vehicle-based experiments.

Scan Matching for Navigation of a Mobile Robot in Semi-Structured Terrain Conditions

To ensure unmanned autonomous movement of ground robotic means, it is required to accurately determine the position and orientationof the robot. The present study is related to the estimation of coordinates by comparing the scans of a laser scanning rangefinder in conditionsof semi-structed infrastructure and the absence of a global satellite communications signal. The existing methods of comparing scans havesignificant drawbacks in the conditions of movement over a semi-structured terrain, associated both with the processing time of data fromthe laser scanning rangefinder, and with the quality of the results obtained. The scan is preliminarily placed in a map consisting of cells.Each cell of around point scan is described by forces represented by the laws of physics or probability theory. In the cells of the map, wetake into account the mutual influence of all forces from each point of the scan and thus we obtain the resulting artificial potential field ofthe scan. The position of the robot is estimated by the change in the number of acting forces of one scan per points of the next scan takinginto account their direction. We estimate the orientation of the robot based on the sum of the vector products of the forces and distancesto the given forces acting on the points of the next scan. This method allows you to calculate the displacement of the robot between scansregardless of road conditions and terrain. This article presents the results of an experimental verification of the method on a mock-up of amobile robot equipped with a Velodyne HDL-32 LIDAR. We indicate the operating conditions of the method for a given LIDAR, as wellas the time spent on calculating the bias estimate. Given the peculiarities of the LIDAR, we present a method for eliminating the DopplerEffect (distortion) for the original point cloud. A comparative analysis of the developed method for integrating wheel odometry data, inertialand satellite navigation using the Extended Kalman Filter shows the applicability of this method to assess the position and orientation of therobot in conditions of its movement over rough terrain.

Design and Implementation of an Autonomous Electric Vehicle for Self-Driving Control under GNSS-Denied Environments

In this study, the hardware and software design and implementation of an autonomous electric vehicle are addressed. We aimed to develop an autonomous electric vehicle for path tracking. Control and navigation algorithms are developed and implemented. The vehicle is able to perform path-tracking maneuvers under environments in which the positioning signals from the Global Navigation Satellite System (GNSS) are not accessible. The proposed control approach uses a modified constrained input-output nonlinear model predictive controller (NMPC) for path-tracking control. The proposed localization algorithm used in this study guarantees almost accurate position estimation under GNSS-denied environments. We discuss the procedure for designing the vehicle hardware, electronic drivers, communication architecture, localization algorithm, and controller architecture. The system’s full state is estimated by fusing visual inertial odometry (VIO) measurements with wheel odometry data using an extended Kalman filter (EKF). Simulation and real-time experiments are performed. The obtained results demonstrate that our designed autonomous vehicle is capable of performing path-tracking maneuvers without using Global Navigation Satellite System positioning data. The designed vehicle can perform challenging path-tracking maneuvers with a speed of up to 1 m per second.