Navigation Model Research Articles

A Temporal Learning Approach to Inpainting Endoscopic Specularities and Its Effect on Image Correspondence

Video streams are utilised to guide minimally-invasive surgery and diagnosis in a wide range of procedures, and many computer-assisted techniques have been developed to automatically analyse them. These approaches can provide additional information to the surgeon such as lesion detection, instrument navigation, or anatomy 3D shape modelling. However, the necessary image features to recognise these patterns are not always reliably detected due to the presence of irregular light patterns such as specular highlight reflections. In this paper, we aim at removing specular highlights from endoscopic videos using machine learning. We propose using a temporal generative adversarial network (GAN) to inpaint the hidden anatomy under specularities, inferring its appearance spatially and from neighbouring frames, where they are not present in the same location. This is achieved using in-vivo data from gastric endoscopy (Hyper Kvasir) in a fully unsupervised manner that relies on the automatic detection of specular highlights. System evaluations show significant improvements to other methods through direct comparison and ablation studies that depict the importance of the network’s temporal and transfer learning components. The generalisability of our system to different surgical setups and procedures was also evaluated qualitatively on in-vivo data of gastric endoscopy and ex-vivo porcine data (SERV-CT, SCARED). We also assess the effect of our method in comparison to other methods on computer vision tasks that underpin 3D reconstruction and camera motion estimation, namely stereo disparity, optical flow, and sparse point feature matching. These are evaluated quantitatively and qualitatively and results show a positive effect of our specular inpainting method on these tasks in a novel comprehensive analysis. Our code and dataset are made available at https://github.com/endomapper/Endo-STTN.

Automation of escape route analysis for BIM-based building code checking

The use of BIM is becoming increasingly important in the AEC industry because the included dataset enables the automation of many processes. One emerging application of BIM is the automation of building code checking by building authorities. Building code checking is a time-consuming process, particularly for escape routes. Therefore, its automation promises considerable optimisation potential. This study presents a semi-automated escape route analysis for building code checking. Our developed concept consists of two automated steps, with a legally required manual step in between: (1) escape route generation, which provides possible routes from which designers can select a valid combination, and (2) escape route validation, which checks the selected combination of escape routes for validity. The basis of both automated steps is a novel indoor navigation model. We implemented the concept in Solibri Office and validated it with real-world models. The developed escape route analysis method shows potential to improve designers’ escape route planning and building authorities’ escape route checking.

Designing E-Course Navigation

In the design of the e-course learning process, it is proposed to use the concept of a navigation model. This model provides convenient and didactically based navigation through the educational material of the course. The work is based on the recent research of the authors on the structuring of educational content in the form of an e-course content model, ordered in the form of a hierarchical structure of the ontologies of learning elements that make up the course. The navigation model concept includes a set of matrices of relations of order and logical coherence of learning elements and corresponding to these matrices of orgraphs of the sequence of study and logical coherence of fragments of educational content. The navigation model answers two important questions of designing an electronic course: 1) what should be the didactically rational sequence of studying the learning elements in the course that is being created; 2) what logical connections should be established between the individual learning elements of the course in order to ensure convenient and didactically based navigation of the course. The mathematical substantiation of the navigation model is given, its properties are investigated, and integral characteristics are introduced. The use of the e-course navigation model helps to: determine and visually represent a rational sequence of studying the educational material, as well as the logical reference links between its various fragments; provide effective assistance to students in navigating the course; analyze and compare different educational materials, assess the level of didactic significance of various learning elements; minimize the complexity of preparing exercises for training and control tests, and the complexity of training and control procedures for e-learning. The concept of a navigation model complies with the international e-learning standards SCORM and IMS, complements them with specific algorithms for aggregating learning objects (SCOs) into e-courses, and assists students in learning them. The mathematical substantiation of the navigation model makes it possible to automate the design of e-courses.

Integrated navigation model based on TDCP constrained algorithm

The velocity information estimated by the global navigation satellite system (GNSS) receiver is an important element for the dynamic alignment of the inertial navigation system, and it is of great significance to analyze it deeply and meticulously. A variety of GNSS velocity measurement models show different characteristics in a changeable environment, and this status quo is bound to break the monotonous situation in which the Doppler model is widely used. In this regard, this paper applies different GNSS velocity measurement models to strap-down interial navigation system (SINS) dynamic alignment and continuous observation. In addition, aiming at the shortcomings of the traditional time-differenced carrier phase (TDCP) algorithm, an optimization method is deduced from the formula level, and two effective constraint algorithms are given. Then, according to the vehicle test results, comprehensively compare the integrated navigation performance of various speed measurement models, and analyze the improvement effect of the proposed TDCP algorithm. This paper provides a summary for the comprehensive study of GNSS velocity measurement model and the application of optimized carrier phase to integrated navigation, which has certain practical value.

A Family-Based Mental Health Navigator Intervention for Youth in the Child Welfare System: Protocol for a Randomized Controlled Trial.

Youth in the child welfare system (child welfare-involved [CWI] youth) have high documented rates of mental health symptoms and experience significant disparities in mental health care services access and engagement. Adolescence is a developmental stage that confers increased likelihood of experiencing mental health symptoms and the emergence of disorders that can persist into adulthood. Despite a high documented need for evidence-based mental health services for CWI youth, coordination between child welfare and mental health service systems to increase access to care remains inadequate, and engagement in mental health services is low. Navigator models developed in the health care field to address challenges of service access, fragmentation, and continuity that affect the quality of care provide a promising approach to increase linkage to, and engagement in, mental health services for CWI youth. However, at present, there is no empirically supported mental health navigator model to address the unique and complex mental health needs of CWI youth and their families. Using a randomized controlled trial, this study aims to develop and test a foster care family navigator (FCFN) model to improve mental health service outcomes for CWI adolescents (aged 12-17 years). The navigator model leverages an in-person navigator and use of adjunctive digital health technology to engage with, and improve, care coordination, tracking, and monitoring of mental health service needs for CWI youth and families. In total, 80 caregiver-youth dyads will be randomized to receive either the FCFN intervention or standard of care (clinical case management services): 40 (50%) to FCFN and 40 (50%) to control. Qualitative exit interviews will inform the feasibility and acceptability of the services received during the 6-month period. The primary trial outcomes are mental health treatment initiation and engagement. Other pre- and postservice outcomes, such as proportion screened and time to screening, will also be evaluated. We hypothesize that youth receiving the FCFN intervention will have higher rates of mental health treatment initiation and engagement than youth receiving standard of care. We propose enrollment of 80 dyads by March 2024, final data collection by September 2024, and the publication of main findings in March 2025. After final data analysis and writing of the results, the resulting manuscripts will be submitted to journals for dissemination. This study will be the first to produce empirically driven conclusions and recommendations for implementing a family mental health navigation model for CWI youth with long-standing and unaddressed disparities in behavioral health services access. The study findings have potential to have large-scale trial applicability and be feasible and acceptable for eventual system implementation and adoption. ClinicalTrials.gov NCT04506437; https://www.clinicaltrials.gov/study/NCT04506437. DERR1-10.2196/49999.

Machine learning and UAV path following identification algorithm based on navigation spoofing

With advances in navigation technology, unmanned aerial vehicles (UAVs) have become widely available to all sections of society. Given the potential hazards of UAVs, as seen from their use in the Russia–Ukraine war and security incidents at London and Baghdad airports, counter-UAV technology is receiving unprecedented attention. This paper describes a method for taking control of a UAV indirectly by navigation spoofing and luring it to a designated area from which it can be neutralized. Our contributions are threefold: first, we analyze the limitations of the traditional state estimation and control model for UAV navigation based on the integration of the Global Navigation Satellite System and Inertial Navigation System. Second, we propose a particle hypothesis and planning model that is insensitive to the particular UAV navigation system, and decompose the UAV navigation spoofing process into two steps: identification and planning. Finally, to overcome the poor heading angle prediction accuracy of unary polynomial regression, we propose a support vector regression algorithm that improves the prediction accuracy from 1.5° to 1.01° under navigation spoofing. Experimental results using the proposed navigation spoofing method prove that machine learning offers significant advantages in modeless identification.

Обобщенная модель автономной навигации полноприводного мобильного робота на пересеченной местности

Обобщенная модель автономной навигации полноприводного мобильного робота на пересеченной местности

Error suppression method of SINS/CNS integrated navigation based on diametrical stargazing

The strap-down inertial navigation system (SINS)/celestial navigation system (CNS) integrated navigation system has been extensively employed in the aerospace field owing to its unique advantages. The internal parameter errors of the star sensor constitute one of the primary factors that affect the navigation accuracy. In this paper, an error suppression method based on diametrical stargazing has been designed to suppress the navigation errors caused by the internal parameter errors of the star sensor. First, a SINS/CNS integrated navigation model based on Kalman filter has been established. Then the influence of the principal point error, image-plane tilt error, and distortions of the optical system on the measurement accuracy of the star sensor have been analyzed. Further, the process of introducing the errors into the measurement equation has been obtained. Finally, for multiple errors, two stargazing methods have been designed to suppress the impact of these errors on the navigation accuracy, whereas the constant errors of the inertial device can be modulated. The simulation results demonstrate that the proposed method can improve the positioning accuracy of the system, and verify the effectiveness of the algorithm.

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.

3D-printed helmet-type neuro-navigation approach (I-Helmet) for transcranial magnetic stimulation.

Neuro-navigation is a key technology to ensure the clinical efficacy of TMS. However, the neuro-navigation system based on positioning sensor is currently unable to be promoted and applied in clinical practice due to its time-consuming and high-cost. In the present study, we designed I-Helmet system to promote an individualized and clinically friendly neuro-navigation approach to TMS clinical application. I-Helmet system is based on C++ with a graphical user interface that allows users to design a 3D-printed helmet model for coil navigation. Besides, a dedicated coil positioning accuracy detection method was promoted based on three-dimensional (3D) printing and 3D laser scanning for evaluation. T1 images were collected from 24 subjects, and based on each image, phantom were created to simulate skin and hair. Six 3D-printed helmets with the head positioning hole enlarged by 0-5% tolerance in 1% increments were designed to evaluate the influences of skin, hair, and helmet-tolerance on the positioning accuracy and contact force of I-Helmet. Finally, I-Helmet system was evaluated by comparing its positioning accuracy with three skin hardnesses, three hair styles, three operators, and with or without landmarks. The accuracy of the proposed coil positioning accuracy detection method was about 0.30 mm in position and 0.22° in orientation. Skin and hair had significant influences on positioning accuracy (p < 0.0001), whereas different skin hardnesses, hair styles, and operators did not (p > 0.05). The tolerance of the helmet presented significant influences on positioning accuracy (p < 0.0001) and contact force (p < 0.0001). The positioning accuracy significantly increased (p < 0.0001) with landmark guided I-Helmet. 3D-printed helmet-type Neuro-navigation approach (I-Helmet) with 3% tolerance and landmarks met the positioning requirements for TMS in clinical practice with less than 5 N mean contact force, 3-5 mm positioning accuracy, 65.7 s mean operation time, and 50-yuan material cost. All the results suggest that the cost of I-Helmet system may be much less than the that of training clinical doctors to position the coil of TMS operation during short period of time.

A Bioinspired Navigation System for Multirotor UAV by Integrating Polarization Compass/Magnetometer/INS/GNSS

Stable and accurate heading angle estimation is challenging for low-cost small UAV navigation, especially in the presence of geomagnetic anomaly. In this paper, a bio-inspired integrated navigation system (BINS) is designed such that the environmental adaptability of the navigation system can be enhanced. A bio-inspired integrated navigation algorithm is presented to fuse the information of polarized skylight, geomagnetic field, inertia, and global navigation satellite system (GNSS). The relationship among ambient light intensity, degree of polarization, and accuracy of polarization compass is revealed. Thus, the integrated polarization compass can be adapted to more complex weather conditions. A tightly coupled integrated navigation model is developed for BINS to fuse the information of polarization compass and magnetometer. Within the developed scheme, the polarization compass can not only correct the heading error of the system, but also help the magnetometer reconstruct the actual local geomagnetic field distribution. The problem of inaccurate geomagnetic field model in the magnetic anomaly environment is addressed. Moreover, the Chi-Square test is used to detect polarization compass outliers such that the navigation modes can be switched promptly. Finally, the ground static experiments and UAV flight tests are carried out. In comparison of traditional polarization navigation and geomagnetic navigation, the proposed system effectively improves the heading accuracy.

A multi-objective optimisation strategy for ice navigation under ship safety-following scenarios

Shipping activities in the Northern Sea Route (NSR) are increasing. However, due to the complex sea ice conditions, ships still need assistance of icebreakers when sailing through most Arctic waters. Icebreaker escort operations are important to ensure the safe passage of ships through ice-covered waters, but the speed of following ships will be limited, thus creating challenges for NSR sailing economics and emission reduction issues. Optimizing the sailing speed of following ships can reduce NSR ship emissions and improve navigation economy while safeguarding their sailing safety. In this paper, a green, economical and safe optimization approach under ship-following navigation has been developed by drawing on similarities with the car-following phenomenon. Then, three ship-following scenarios are designed for simulating the actual ice navigation conditions. Finally, a cost-eco-efficiency multi-objective optimisation model for ice navigation is established, and the performance of the proposed model is demonstrated by the navigation case of “Tian You”. The results show that the proposed model can effectively reduce navigation costs and control navigation emissions while ensuring navigation safety. This study also demonstrates that there is still room for a 20% improvement in the optimised speed and a 4.8% reduction in the ship-following distance compared to the actual situation.

Remapping in a recurrent neural network model of navigation and context inference.

Neurons in navigational brain regions provide information about position, orientation, and speed relative to environmental landmarks. These cells also change their firing patterns ('remap') in response to changing contextual factors such as environmental cues, task conditions, and behavioral states, which influence neural activity throughout the brain. How can navigational circuits preserve their local computations while responding to global context changes? To investigate this question, we trained recurrent neural network models to track position in simple environments while at the same time reporting transiently-cued context changes. We show that these combined task constraints (navigation and context inference) produce activity patterns that are qualitatively similar to population-wide remapping in the entorhinal cortex, a navigational brain region. Furthermore, the models identify a solution that generalizes to more complex navigation and inference tasks. We thus provide a simple, general, and experimentally-grounded model of remapping as one neural circuit performing both navigation and context inference.

Remapping in a recurrent neural network model of navigation and context inference

Neurons in navigational brain regions provide information about position, orientation, and speed relative to environmental landmarks. These cells also change their firing patterns (‘remap’) in response to changing contextual factors such as environmental cues, task conditions, and behavioral states, which influence neural activity throughout the brain. How can navigational circuits preserve their local computations while responding to global context changes? To investigate this question, we trained recurrent neural network models to track position in simple environments while at the same time reporting transiently-cued context changes. We show that these combined task constraints (navigation and context inference) produce activity patterns that are qualitatively similar to population-wide remapping in the entorhinal cortex, a navigational brain region. Furthermore, the models identify a solution that generalizes to more complex navigation and inference tasks. We thus provide a simple, general, and experimentally-grounded model of remapping as one neural circuit performing both navigation and context inference.

Time-Driven and Privacy-Preserving Navigation Model for Vehicle-to-Vehicle Communication Systems

Effective time-driven navigation is an operative way to alleviate traffic congestion, which is also a challenging problem in the Internet of Vehicles context. Most existing centralized navigation systems often cannot react promptly to real-time local traffic situations, while most existing distributed navigation systems are vulnerable to privacy attacks. To overcome these drawbacks, we propose a learning model that provides a provable guarantee of vehicles' privacy while still enabling efficient navigation under real-time traffic conditions. The proposed model adopts a novel multi-agent system with customized differentially private mechanisms. To verify the effectiveness and stability of our approach, we implement the proposed method on CARLA, which is an autonomous driving simulator. In four experimental tasks with varying parameters, we demonstrate fully that our proposed method outperforms other benchmarks.

Low-Cost Onboard Multisensor State Estimator Based on Global Mixed Sampling Optimization Filter

Long-term convergent and accurate state estimation( attitude, velocity, and position) is critical to unmanned aerial vehicles (UAVs). However, the measurement accuracy and noise of multiple low-cost sensors onboard are poorer compared to high-precision sensors. To improve the solution accuracy and reliability of UAVs state estimation based on the low-cost sensors, a two-step multi-sensor fusion estimator (federated mixed-sampling Kalman filter, FMSKF) is proposed. Firstly, a multi-sensor integrated navigation model array is designed depending on the different sensors, including the gyroscope, accelerometer, magnetometer, GPS module, and barometer. Then, a three-stage mixed sampling Kalman filter is proposed to obtain the local convergent state vector, which is difficult to meet the flight accuracy requirements. Therefore, a multi-sensor global information optimization filter is proposed to obtain the global convergent navigation solution parameters. Finally, the simulation and flight experimental results and detailed analysis demonstrate that the proposed algorithm can improve the state estimation accuracy, filtering robustness, and obtain desirable navigation performance.

NAVIGATION STRATEGY FOR MOBILE ROBOT BASED ON COMPUTER VISION AND YOLOV5 NETWORK IN THE UNKNOWN ENVIRONMENT

Intelligent mobile robots must possess the ability to navigate in complex environments. The field of mobile robot navigation is continuously evolving, with various technologies being developed. Deep learning has gained attention from researchers, and numerous navigation models utilizing deep learning have been proposed. In this study, the YOLOv5 model is utilized to identify objects to aid the mobile robot in determining movement conditions. However, the limitation of deep learning models being trained on insufficient data, leading to inaccurate recognition in unforeseen scenarios, is addressed by introducing an innovative computer vision technology that detects lanes in real-time. Combining the deep learning model with computer vision technology, the robot can identify different types of objects, allowing it to estimate distance and adjust speed accordingly. Additionally, the paper investigates the recognition reliability in varying light intensities. The findings of this study offer promising directions for future breakthroughs in mobile robot navigation

(Un)equal demands and opportunities: Conceptualizing student navigation in undergraduate engineering programs

AbstractBackgroundIt is well known that earning a bachelor's degree in engineering is a demanding task, but ripe with opportunity. For students from historically excluded demographic groups, this task is exacerbated by oppressive circumstances. Although considerable research has documented how student outcomes differ across demographic groups, much less is known about the dynamic processes that marginalize some students.PurposeThe purpose of this article is to propose a conceptual model of student navigation in the context of undergraduate engineering programs. Our goal is to illustrate how localized, structural features unjustly shape the demands and opportunities encountered by students and influence how they respond.Scope/MethodWe developed our model using an iterative, four‐stage process. This process included (1) clarifying the purpose of the development process; (2) identifying concepts and insights from prior research; (3) synthesizing the concepts and insights into propositions; and (4) visualizing the suspected relationships between the salient constructs in the propositions.ResultsOur model focuses on the dynamic interactions between the characteristics of students, the embedded contexts in which they are situated, and the support infrastructure of their learning environment.ConclusionThe resulting model illustrates the influence of structural features on how students a) respond to demands and opportunities and b) navigate obstacles present in the learning environment. Although its focus is on marginalized students in undergraduate engineering programs, the model may be applicable to STEM higher education more broadly.

Addressing social isolation and loneliness in older adults at the health plan level: A quality improvement project

The aim of this quality improvement project was to develop and implement a strategy to address social isolation and loneliness in older adults at the health plan level. Validated tools were utilized to identify health plan members experiencing social isolation and/or loneliness. Identified members were offered a home-based intervention which consisted of a befriending model provided through in-home caregiver visits (n = 26) as well as a social prescribing (navigation) model (n = 23) to connect members to community resources. The quality improvement project demonstrated that validated assessment tools can successfully identify older health plan members experiencing social isolation and/or loneliness. In addition, the project results indicated that befriending through in-home caregiver visits and social navigation can improve social isolation and loneliness scores. Additional studies are needed to determine the significance of these improvements and if social navigation adds any additional benefit compared to in-home caregiver visits alone.

Dynamic trajectory planning for ships in dense environment using collision grid with deep reinforcement learning

In areas with high maritime traffic, ship safety is of utmost importance when validating autonomous navigation models. While exact methods exist for specific situations, they are inadequate in a global context. This study employs an approximate deep reinforcement learning method to solve the navigation problem in a dense environment with numerous static and moving obstacles. Our model prioritizes ship integrity by enabling the agent to dynamically adapt its kinematics to its surroundings to reach a designated goal without colliding with obstacles. To achieve this, we incorporate collision grids in the form of danger zones as input to our model and train it using the proximal policy optimization algorithm. Additionally, we propose implementing the International Regulations for Preventing Collisions at Sea (COLREGs) in the collision grid as these navigation rules are necessary to obtain realistic behavior of an autonomous agent. The agent’s performance is evaluated on a set of randomly generated scenarios operating in an environment complexity similar to the one used during training. These tests demonstrate that this type of data structure allows a trained agent to navigate in a dense environment while adhering to the COLREGs with a success rate of 94.69%.