Pub Date : 2023-04-24DOI: 10.1109/PLANS53410.2023.10140096
Birendra Kujur, S. Khanafseh, B. Pervan
In this paper, we demonstrate the performance of the proposed optimal Inertial Navigation System (INS) monitor [19] using experimental setup that includes Global Navigation Satellite System (GNSS) spoofing scenarios using state-of-the-art GNSS spoofing software Skydel and real IMU data. Skydel is a software-based simulation platform which can generate GNSS radio frequency (RF) signals that can be fed into a receiver, using a Universal Software Radio Peripheral (USRP). The experimental setup includes GNSS, and Inertial Measurement Unit (IMU), dynamic data collection unit in a ground vehicle, which is used to generate the test trajectory for Skydel. Skydel is then used to generate authentic and spoofed signals which are then collected using a GNSS receiver. Along with the previously collected IMU data, the authentic and spoofed signals are used to validate the optimal INS monitor. A spoofer's uncertainty of user position (or position tracking error) is modeled as white Gaussian noise and added to the replica of authentic signal to form the spoofed signal. We show that the monitor is able to detect spoofer's tracking error even at decimeter level magnitudes. As a result, the conducted experiments demonstrate the monitor ability in detecting realistic GNSS spoofing events even with minimal tracking errors.
{"title":"Experimental Validation of Optimal INS Monitor against GNSS Spoofer Tracking Error Detection","authors":"Birendra Kujur, S. Khanafseh, B. Pervan","doi":"10.1109/PLANS53410.2023.10140096","DOIUrl":"https://doi.org/10.1109/PLANS53410.2023.10140096","url":null,"abstract":"In this paper, we demonstrate the performance of the proposed optimal Inertial Navigation System (INS) monitor [19] using experimental setup that includes Global Navigation Satellite System (GNSS) spoofing scenarios using state-of-the-art GNSS spoofing software Skydel and real IMU data. Skydel is a software-based simulation platform which can generate GNSS radio frequency (RF) signals that can be fed into a receiver, using a Universal Software Radio Peripheral (USRP). The experimental setup includes GNSS, and Inertial Measurement Unit (IMU), dynamic data collection unit in a ground vehicle, which is used to generate the test trajectory for Skydel. Skydel is then used to generate authentic and spoofed signals which are then collected using a GNSS receiver. Along with the previously collected IMU data, the authentic and spoofed signals are used to validate the optimal INS monitor. A spoofer's uncertainty of user position (or position tracking error) is modeled as white Gaussian noise and added to the replica of authentic signal to form the spoofed signal. We show that the monitor is able to detect spoofer's tracking error even at decimeter level magnitudes. As a result, the conducted experiments demonstrate the monitor ability in detecting realistic GNSS spoofing events even with minimal tracking errors.","PeriodicalId":344794,"journal":{"name":"2023 IEEE/ION Position, Location and Navigation Symposium (PLANS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114828245","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-24DOI: 10.1109/PLANS53410.2023.10139983
Jiachen Zhang, W. Wen, L. Hsu, Zhengxia Gong, Zhongzhe Su
Safety-critical localization is essential for unmanned autonomous systems. LiDAR localization gains great popularity in urban canyons due to its high ranging accuracy. Inheriting from the integrity monitoring theory for GNSS, safety-certifiable LiDAR localization first consists in fault detection and exclusion (FDE). In face of numerous LiDAR measurements, conventional chi-square test for FDE is computationally intractable. What's more, inliers could be mistakenly excluded without reconsideration. This paper proposes a computationally tractable and flexible FDE method. It's realized via outlier mitigation aided by graduated non-convexity (GNC) relaxation. The two novel loss functions truncated least square (TLS) and the Geman McClure (GM) are combined respectively. The outlier-mitigated planar-feature-based LiDAR localization is formulated with GNC and TLS or GM. More importantly, a triple-layer optimization method is proposed to solve the localization formulation. Besides the typical GNC relaxation, the control parameter is taken into consideration for tuning the outliers resistance degree. The outlier mitigated pose estimation and the weightings ranging from 0 to 1 for the exploited LiDAR measurements are finally produced. Extensive experiments of the proposed method is conducted on urban dataset. What's more, considering that TSL and GM provides distinct outlier mitigation patterns, the performances from them are investigated and compared.
{"title":"LiDAR Feature Outlier Mitigation Aided by Graduated Non-convexity Relaxation for Safety-critical Localization in Urban Canyons","authors":"Jiachen Zhang, W. Wen, L. Hsu, Zhengxia Gong, Zhongzhe Su","doi":"10.1109/PLANS53410.2023.10139983","DOIUrl":"https://doi.org/10.1109/PLANS53410.2023.10139983","url":null,"abstract":"Safety-critical localization is essential for unmanned autonomous systems. LiDAR localization gains great popularity in urban canyons due to its high ranging accuracy. Inheriting from the integrity monitoring theory for GNSS, safety-certifiable LiDAR localization first consists in fault detection and exclusion (FDE). In face of numerous LiDAR measurements, conventional chi-square test for FDE is computationally intractable. What's more, inliers could be mistakenly excluded without reconsideration. This paper proposes a computationally tractable and flexible FDE method. It's realized via outlier mitigation aided by graduated non-convexity (GNC) relaxation. The two novel loss functions truncated least square (TLS) and the Geman McClure (GM) are combined respectively. The outlier-mitigated planar-feature-based LiDAR localization is formulated with GNC and TLS or GM. More importantly, a triple-layer optimization method is proposed to solve the localization formulation. Besides the typical GNC relaxation, the control parameter is taken into consideration for tuning the outliers resistance degree. The outlier mitigated pose estimation and the weightings ranging from 0 to 1 for the exploited LiDAR measurements are finally produced. Extensive experiments of the proposed method is conducted on urban dataset. What's more, considering that TSL and GM provides distinct outlier mitigation patterns, the performances from them are investigated and compared.","PeriodicalId":344794,"journal":{"name":"2023 IEEE/ION Position, Location and Navigation Symposium (PLANS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123045238","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-24DOI: 10.1109/PLANS53410.2023.10139979
Uthman Olawoye, Jason N. Gross
This paper explores the use of applying a deep learning approach for 3D object detection to compute the relative position of an Unmanned Aerial Vehicle (UAV) from an Unmanned Ground Vehicle (UGV) equipped with a LiDAR sensor in a GPS Denied environment. This was achieved by evaluating the LiDAR sensor's data through a 3D detection algorithm (PointPillars). The PointPillars algorithm incorporates a column voxel point-cloud representation and a 2D Convolutional Neural Network (CNN) to generate distinctive point-cloud features representing the object to be identified, in this case, the UAV. The current localization method utilizes point-cloud segmentation, Euclidean clustering, and predefined heuristics to obtain the relative position of the UAV. Results from the two methods were then compared to a reference truth solution.
{"title":"UAV Position Estimation Using a LiDAR-based 3D Object Detection Method","authors":"Uthman Olawoye, Jason N. Gross","doi":"10.1109/PLANS53410.2023.10139979","DOIUrl":"https://doi.org/10.1109/PLANS53410.2023.10139979","url":null,"abstract":"This paper explores the use of applying a deep learning approach for 3D object detection to compute the relative position of an Unmanned Aerial Vehicle (UAV) from an Unmanned Ground Vehicle (UGV) equipped with a LiDAR sensor in a GPS Denied environment. This was achieved by evaluating the LiDAR sensor's data through a 3D detection algorithm (PointPillars). The PointPillars algorithm incorporates a column voxel point-cloud representation and a 2D Convolutional Neural Network (CNN) to generate distinctive point-cloud features representing the object to be identified, in this case, the UAV. The current localization method utilizes point-cloud segmentation, Euclidean clustering, and predefined heuristics to obtain the relative position of the UAV. Results from the two methods were then compared to a reference truth solution.","PeriodicalId":344794,"journal":{"name":"2023 IEEE/ION Position, Location and Navigation Symposium (PLANS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128998042","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-24DOI: 10.1109/PLANS53410.2023.10139930
Michele Brizzi, A. Neri
A novel technique for building a detailed Digital Map of the rail environment has been implemented and tested. It is based on the fusion of multiple rides acquired by means of on-board GNSS receivers, IMUs and visual sensors from trains in commercial operation. More in detail, an efficient incremental procedure for processing the collected data to reduce the computational complexity and storage requirements has been proposed. We demonstrate that the proposed system is able to obtain accurate track geometry and trackside objects position information while being robust to signal degradations typical of the rail environment.
{"title":"Multi-ride Fusion for Rail Digital Map Construction","authors":"Michele Brizzi, A. Neri","doi":"10.1109/PLANS53410.2023.10139930","DOIUrl":"https://doi.org/10.1109/PLANS53410.2023.10139930","url":null,"abstract":"A novel technique for building a detailed Digital Map of the rail environment has been implemented and tested. It is based on the fusion of multiple rides acquired by means of on-board GNSS receivers, IMUs and visual sensors from trains in commercial operation. More in detail, an efficient incremental procedure for processing the collected data to reduce the computational complexity and storage requirements has been proposed. We demonstrate that the proposed system is able to obtain accurate track geometry and trackside objects position information while being robust to signal degradations typical of the rail environment.","PeriodicalId":344794,"journal":{"name":"2023 IEEE/ION Position, Location and Navigation Symposium (PLANS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127596734","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-24DOI: 10.1109/PLANS53410.2023.10139969
Sharbel E. Kozhaya, Haitham Kanj, Z. M. Kassas
A novel blind spectral approach is proposed for blind beacon estimation, Doppler tracking, and opportunistic positioning with unknown low Earth orbit (LEO) satellite signals. The framework is agnostic to the modulation and multiple access scheme adopted by LEO satellites. First, an analytical derivation of the received signal frequency spectrum is presented, which accounts for the highly dynamic channel between the LEO satellite and a terrestrial receiver. Second, a frequency domain-based blind Doppler discriminator is proposed. Third, a Kalman filter (KF)-based Doppler tracking algorithm is developed. Fourth, a blind beacon estimation framework for LEO satellites is proposed and its convergence properties are studied. Simulation results are presented showing successful beacon estimation and Doppler tracking of Starlink LEO satellites transmitting 5G orthogonal division multiple access (OFDM) signals. Experimental results are presented demonstrating the efficacy of the proposed framework on multi-constellation LEO satellites, namely OneWeb, Starlink, Orbcomm, and Iridium NEXT. Despite adopting different modulation and multiple access transmission schemes, the proposed framework is capable of successfully estimating the beacon and tracking the Doppler, in a blind fashion, of 8 LEO satellites (2 OneWeb, 4 Starlink, 1 Iridium NEXT, and 1 Orbcomm) over a period of about 560 seconds with Hz-level accuracy. The produced Doppler measurements were fused through a nonlinear least-squares estimator to localize a stationary receiver to an unprecedented level of accuracy. Starting with an initial estimate about 3,600 km away, a final three-dimensional (3-D) position error of 5.8 m and 2-D position error of 5.1 m was achieved. Aside from achieving this unprecedented accuracy, these results represent the first successful opportunistic tracking of unknown OneWeb LEO signals and their exploitation for positioning.
{"title":"Multi-Constellation Blind Beacon Estimation, Doppler Tracking, and Opportunistic Positioning with OneWeb, Starlink, Iridium NEXT, and Orbcomm LEO Satellites","authors":"Sharbel E. Kozhaya, Haitham Kanj, Z. M. Kassas","doi":"10.1109/PLANS53410.2023.10139969","DOIUrl":"https://doi.org/10.1109/PLANS53410.2023.10139969","url":null,"abstract":"A novel blind spectral approach is proposed for blind beacon estimation, Doppler tracking, and opportunistic positioning with unknown low Earth orbit (LEO) satellite signals. The framework is agnostic to the modulation and multiple access scheme adopted by LEO satellites. First, an analytical derivation of the received signal frequency spectrum is presented, which accounts for the highly dynamic channel between the LEO satellite and a terrestrial receiver. Second, a frequency domain-based blind Doppler discriminator is proposed. Third, a Kalman filter (KF)-based Doppler tracking algorithm is developed. Fourth, a blind beacon estimation framework for LEO satellites is proposed and its convergence properties are studied. Simulation results are presented showing successful beacon estimation and Doppler tracking of Starlink LEO satellites transmitting 5G orthogonal division multiple access (OFDM) signals. Experimental results are presented demonstrating the efficacy of the proposed framework on multi-constellation LEO satellites, namely OneWeb, Starlink, Orbcomm, and Iridium NEXT. Despite adopting different modulation and multiple access transmission schemes, the proposed framework is capable of successfully estimating the beacon and tracking the Doppler, in a blind fashion, of 8 LEO satellites (2 OneWeb, 4 Starlink, 1 Iridium NEXT, and 1 Orbcomm) over a period of about 560 seconds with Hz-level accuracy. The produced Doppler measurements were fused through a nonlinear least-squares estimator to localize a stationary receiver to an unprecedented level of accuracy. Starting with an initial estimate about 3,600 km away, a final three-dimensional (3-D) position error of 5.8 m and 2-D position error of 5.1 m was achieved. Aside from achieving this unprecedented accuracy, these results represent the first successful opportunistic tracking of unknown OneWeb LEO signals and their exploitation for positioning.","PeriodicalId":344794,"journal":{"name":"2023 IEEE/ION Position, Location and Navigation Symposium (PLANS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127873185","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-24DOI: 10.1109/PLANS53410.2023.10139942
Sarah Hasnain, Michael Berkson, Sharon Maguire, Evan Sun, Katie Zaback
The increasing number of private and public actors interested in space-based missions has driven need for greater flexibility and reliability in regards to navigation. Autonomous navigation in space will reduce reliance on ground-based systems and high operational costs due to crowded communication networks. Further, there is a clear need for autonomous navigation solutions in GPS-denied environments, as well as deep-space regions in which traditional GPS methods are infeasible. One promising approach for achieving autonomous navigation in the dynamic landscape of space is X-ray pulsar-based navigation (XNAV). XNAV capitalizes on the periodicity of pulsar-emitted X-rays for positioning, navigation, as well as determining and responding to timing error (PNT). In this paper, a novel, flexible pulsar simulation framework for the testing, and validation of XNAV systems is presented. Pulsar-Leveraged Autonomous Navigation Testbed System (PLANTS) is a low-cost software-hardware hybrid testbed for XNAV PNT solutions. PLANTS simulates high-fidelity pulsar X-ray events along desired flight trajectories over a user-defined mission timeline, which can be used to optimize XNAV hardware and mission planning components (such as spacecraft attitude and X-ray detector orientation planning, based on output pulsar viewing schedules and angles over time). Ultimately, this testbed provides a flexible platform for a wide array of future XNAV research and development efforts aimed at the goal of mission-readiness and sustained space operations. The goal of the PLANTS framework is to develop a system for XNAV project teams which is cost-efficient, algorithm-agnostic (i.e. supports interoperability with current and emerging software toolkits), and incorporates hardware-in-the-loop (HWIL). This paper describes the first iteration of PLANTS, which leverages software-defined radios (SDRs), coupled with a number of software utilities including the Python-based PINT pulsar timing software package. Initial results exhibit successful outputs of pulsar data extraction, transformation, and loading (ETL), flight plans, timing models, and light curves portraying photon arrival events. The future of XNAV will require the development of effective, intelligent navigation algorithms and accessible testing facilities with HWIL. The PLANTS framework meets these needs and empowers advancement of the state-of-the-art in autonomous space navigation.
{"title":"Pulsar-Leveraged Autonomous Navigation Testbed System (PLANTS): A Low-Cost Software-Hardware Hybrid Testbed for Pulsar-based Autonomous Navigation (XNAV) Positioning, Navigation, and Timing (PNT) Solutions","authors":"Sarah Hasnain, Michael Berkson, Sharon Maguire, Evan Sun, Katie Zaback","doi":"10.1109/PLANS53410.2023.10139942","DOIUrl":"https://doi.org/10.1109/PLANS53410.2023.10139942","url":null,"abstract":"The increasing number of private and public actors interested in space-based missions has driven need for greater flexibility and reliability in regards to navigation. Autonomous navigation in space will reduce reliance on ground-based systems and high operational costs due to crowded communication networks. Further, there is a clear need for autonomous navigation solutions in GPS-denied environments, as well as deep-space regions in which traditional GPS methods are infeasible. One promising approach for achieving autonomous navigation in the dynamic landscape of space is X-ray pulsar-based navigation (XNAV). XNAV capitalizes on the periodicity of pulsar-emitted X-rays for positioning, navigation, as well as determining and responding to timing error (PNT). In this paper, a novel, flexible pulsar simulation framework for the testing, and validation of XNAV systems is presented. Pulsar-Leveraged Autonomous Navigation Testbed System (PLANTS) is a low-cost software-hardware hybrid testbed for XNAV PNT solutions. PLANTS simulates high-fidelity pulsar X-ray events along desired flight trajectories over a user-defined mission timeline, which can be used to optimize XNAV hardware and mission planning components (such as spacecraft attitude and X-ray detector orientation planning, based on output pulsar viewing schedules and angles over time). Ultimately, this testbed provides a flexible platform for a wide array of future XNAV research and development efforts aimed at the goal of mission-readiness and sustained space operations. The goal of the PLANTS framework is to develop a system for XNAV project teams which is cost-efficient, algorithm-agnostic (i.e. supports interoperability with current and emerging software toolkits), and incorporates hardware-in-the-loop (HWIL). This paper describes the first iteration of PLANTS, which leverages software-defined radios (SDRs), coupled with a number of software utilities including the Python-based PINT pulsar timing software package. Initial results exhibit successful outputs of pulsar data extraction, transformation, and loading (ETL), flight plans, timing models, and light curves portraying photon arrival events. The future of XNAV will require the development of effective, intelligent navigation algorithms and accessible testing facilities with HWIL. The PLANTS framework meets these needs and empowers advancement of the state-of-the-art in autonomous space navigation.","PeriodicalId":344794,"journal":{"name":"2023 IEEE/ION Position, Location and Navigation Symposium (PLANS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121346867","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-24DOI: 10.1109/PLANS53410.2023.10139960
M. Haag, Mats Martens, Kevin Kotinkar, Jakob Dommaschk
This paper describes a basic framework for cognitive and collaborative navigation of small Unmanned Aerial Vehicles (sUAVs) with a focus on operation in challenging environments where GNSS performance may be deteriorated or even unavailable. The basic framework is based on a dynamic decision system where swarm members, a.k.a. agents, collect local sensor data and data from other agents in the swarm, to estimate the absolute and relative pose state of the swarm and its members and, hence, get better situational awareness to make decision that maintain safety but also satisfy the mission objectives. The paper discusses one possible way to integrate this swarm information using factor graphs and non-linear solvers. Simulation results will show the initial effectiveness of this method within the current architecture. The paper will, furthermore, describe the hardware and software architecture of the TU Berlin swarm test sUAVs and focus on the common GNSS, IMU, range radio board (SwarmEx) that forms the common core of the platforms' sensor payloads. Some initial results of the range radio performance will be included as well. Finally, the flight test environment will be described.
{"title":"Flight Test Setup for Cooperative Swarm Navigation in Challenging Environments using UWB, GNSS, and Inertial Fusion","authors":"M. Haag, Mats Martens, Kevin Kotinkar, Jakob Dommaschk","doi":"10.1109/PLANS53410.2023.10139960","DOIUrl":"https://doi.org/10.1109/PLANS53410.2023.10139960","url":null,"abstract":"This paper describes a basic framework for cognitive and collaborative navigation of small Unmanned Aerial Vehicles (sUAVs) with a focus on operation in challenging environments where GNSS performance may be deteriorated or even unavailable. The basic framework is based on a dynamic decision system where swarm members, a.k.a. agents, collect local sensor data and data from other agents in the swarm, to estimate the absolute and relative pose state of the swarm and its members and, hence, get better situational awareness to make decision that maintain safety but also satisfy the mission objectives. The paper discusses one possible way to integrate this swarm information using factor graphs and non-linear solvers. Simulation results will show the initial effectiveness of this method within the current architecture. The paper will, furthermore, describe the hardware and software architecture of the TU Berlin swarm test sUAVs and focus on the common GNSS, IMU, range radio board (SwarmEx) that forms the common core of the platforms' sensor payloads. Some initial results of the range radio performance will be included as well. Finally, the flight test environment will be described.","PeriodicalId":344794,"journal":{"name":"2023 IEEE/ION Position, Location and Navigation Symposium (PLANS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126978558","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-24DOI: 10.1109/PLANS53410.2023.10140065
Taehoon Lee, Byungjin Lee, Jae-Ryong Yun, S. Sung
In this paper, we propose a method to integrate data from Inertial Navigation System (INS), Magnetic Pose Estimation System (MPS), and Laser Imaging Detection and Ranging (LiDAR) using a Federated Kalman Filter (FKF). We adaptively adjusted the information sharing factor using the Mahalanobis distance to maintain navigation performance in indoor environments with mirrors that contaminate LiDAR measurements. By adaptively adjusting the information sharing factor, we can adjust the weight of each local filter. To validate navigation performance, we conducted UGV driving tests in various indoor environments. We conducted experiments by driving a UGV on a course with a diameter of 3.6 meters. UGVs are equipped with LiDAR, MPS receivers, and IMUs to measure data. We used four 1-meter diameter MPS coils. An optical motion capture device, the Optitrack, was used as reference data.
{"title":"INS/MPS/LiDAR Integrated Navigation System Using Federated Kalman Filter in an Indoor Environment","authors":"Taehoon Lee, Byungjin Lee, Jae-Ryong Yun, S. Sung","doi":"10.1109/PLANS53410.2023.10140065","DOIUrl":"https://doi.org/10.1109/PLANS53410.2023.10140065","url":null,"abstract":"In this paper, we propose a method to integrate data from Inertial Navigation System (INS), Magnetic Pose Estimation System (MPS), and Laser Imaging Detection and Ranging (LiDAR) using a Federated Kalman Filter (FKF). We adaptively adjusted the information sharing factor using the Mahalanobis distance to maintain navigation performance in indoor environments with mirrors that contaminate LiDAR measurements. By adaptively adjusting the information sharing factor, we can adjust the weight of each local filter. To validate navigation performance, we conducted UGV driving tests in various indoor environments. We conducted experiments by driving a UGV on a course with a diameter of 3.6 meters. UGVs are equipped with LiDAR, MPS receivers, and IMUs to measure data. We used four 1-meter diameter MPS coils. An optical motion capture device, the Optitrack, was used as reference data.","PeriodicalId":344794,"journal":{"name":"2023 IEEE/ION Position, Location and Navigation Symposium (PLANS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133159153","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-24DOI: 10.1109/PLANS53410.2023.10140086
Taylor Knuth, P. Groves
This paper introduces an IMU based context machine learning algorithm for terrain topography classification. Four different terrains are considered: concrete, pebble, sand, and grass. The grass terrain is further split into two separate classes based off moisture content of the grass, wet and dry. Separate terrain topography datasets are created by walking on different terrains and logging the data. The subject has been equipped with an IMU attached on the surface of the shoe above the toes. Data is collected and stored via a Bluetooth smartphone controller over multiple recording sessions. Acceleration, angular rate, and magnetic field were recorded. The recorded data is extracted in two second sliding window intervals, whereupon the magnitude of the sensor outputs, in three dimensions, is calculated. A low-pass band filter is also applied to the magnitude for the acceleration, angular rate, and magnetic field data. The magnitude output is processed in the time domain to calculate variance, energy, kurtosis, range, skewness, and the zero-crossing rate. The magnitude data is converted into the frequency domain and the peak magnitude and its corresponding frequency in the sliding window are determined. A set of 44 features is extracted from each window and then tested and trained to classify terrain topography using five different machine learning methods: Artificial Neural Network, Decision Tree, k-Nearest Neighbor, Naive-Bayes, and Support Vector Machine. The 44-feature set is optimized using a wrapper selection algorithm for the Decision Tree and k-Nearest Neighbor algorithms. The results show that by utilizing sensor data from an IMU in combination with machine learning methods a terrain topography classification algorithm can accurately predict various terrains over which the user traverses.
{"title":"IMU Based Context Detection of Changes in the Terrain Topography","authors":"Taylor Knuth, P. Groves","doi":"10.1109/PLANS53410.2023.10140086","DOIUrl":"https://doi.org/10.1109/PLANS53410.2023.10140086","url":null,"abstract":"This paper introduces an IMU based context machine learning algorithm for terrain topography classification. Four different terrains are considered: concrete, pebble, sand, and grass. The grass terrain is further split into two separate classes based off moisture content of the grass, wet and dry. Separate terrain topography datasets are created by walking on different terrains and logging the data. The subject has been equipped with an IMU attached on the surface of the shoe above the toes. Data is collected and stored via a Bluetooth smartphone controller over multiple recording sessions. Acceleration, angular rate, and magnetic field were recorded. The recorded data is extracted in two second sliding window intervals, whereupon the magnitude of the sensor outputs, in three dimensions, is calculated. A low-pass band filter is also applied to the magnitude for the acceleration, angular rate, and magnetic field data. The magnitude output is processed in the time domain to calculate variance, energy, kurtosis, range, skewness, and the zero-crossing rate. The magnitude data is converted into the frequency domain and the peak magnitude and its corresponding frequency in the sliding window are determined. A set of 44 features is extracted from each window and then tested and trained to classify terrain topography using five different machine learning methods: Artificial Neural Network, Decision Tree, k-Nearest Neighbor, Naive-Bayes, and Support Vector Machine. The 44-feature set is optimized using a wrapper selection algorithm for the Decision Tree and k-Nearest Neighbor algorithms. The results show that by utilizing sensor data from an IMU in combination with machine learning methods a terrain topography classification algorithm can accurately predict various terrains over which the user traverses.","PeriodicalId":344794,"journal":{"name":"2023 IEEE/ION Position, Location and Navigation Symposium (PLANS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132076379","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-24DOI: 10.1109/PLANS53410.2023.10140121
Max Hofacker, H. G. Martinez, Martin Seidl, Fran Domazetović, Larissa Balestrero Machado, T. Pany, R. Forstner
This paper presents an UAV emulation system allowing early hardware-in-the-loop testing for Terrain-Relative-Navigation (TRN) and autonomous guidance algorithm development in context of spacecraft landing on asteroids. The capabilities of this system are shown within the scope of an flight campaign in which a Light Detection And Ranging (LiDAR) only odometry navigation, hazard detection and avoidance system was implemented and tested. Furthermore, a special focus on a new asteroid analogue environment is given. The implemented TRN algorithms are based on the result of an Iterative Closest Point (ICP) algorithm and the adopted use of LiDAR range measurements as altimeter source. A Linear Kalman Filter (LKF) performs the necessary sensor fusion taking into account spacecraft control and asteroid environment forces. The TRN system is inspired by the NASA's MAVeN (minimal augmented state algorithm for vision-based navigation) algorithm used as TRN algorithm on the Mars UAV Ingenuity [24].
{"title":"LiDAR-Based Autonomous Landing on Asteroids: Algorithms, Prototyping and End-to-End Testing with a UAV-Based Satellite Emulator","authors":"Max Hofacker, H. G. Martinez, Martin Seidl, Fran Domazetović, Larissa Balestrero Machado, T. Pany, R. Forstner","doi":"10.1109/PLANS53410.2023.10140121","DOIUrl":"https://doi.org/10.1109/PLANS53410.2023.10140121","url":null,"abstract":"This paper presents an UAV emulation system allowing early hardware-in-the-loop testing for Terrain-Relative-Navigation (TRN) and autonomous guidance algorithm development in context of spacecraft landing on asteroids. The capabilities of this system are shown within the scope of an flight campaign in which a Light Detection And Ranging (LiDAR) only odometry navigation, hazard detection and avoidance system was implemented and tested. Furthermore, a special focus on a new asteroid analogue environment is given. The implemented TRN algorithms are based on the result of an Iterative Closest Point (ICP) algorithm and the adopted use of LiDAR range measurements as altimeter source. A Linear Kalman Filter (LKF) performs the necessary sensor fusion taking into account spacecraft control and asteroid environment forces. The TRN system is inspired by the NASA's MAVeN (minimal augmented state algorithm for vision-based navigation) algorithm used as TRN algorithm on the Mars UAV Ingenuity [24].","PeriodicalId":344794,"journal":{"name":"2023 IEEE/ION Position, Location and Navigation Symposium (PLANS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133176436","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：