Pub Date : 2020-07-05DOI: 10.23919/USNC/URSI49741.2020.9321618
D. Breton, C. Haedrich
Ground-to-ground radio links in urban environments rarely enjoy direct line-of-sight between terminals, and therefore in-canyon, over-rooftop, and scattering from distant structures become primary propagation modes. Because both rooftop diffraction and canyon propagation losses can be severe, and because the walls of deep urban canyons often occlude distant scatterers, the relative importance of these three propagation modes to a given urban channel is unclear. We present results of channel sounding measurements at 437 MHz for ground-to-ground links in Boston, Massachusetts, USA to quantify the importance of each propagation mode. Occupancy curves derived from our measured channels suggest that while canyon-mode propagation is dominant for short range urban links, the importance of the distant scatterer propagation mode increases with terminal separation distance, even when those scatterers are occluded from transmitter and/or receiver view. We present an urban channel model which evaluates the vertical profile of incident power on distant scatterers, even those that are occluded, and find that reasonable agreement can be obtained between measured and modeled channels only when occluded buildings are considered.
{"title":"Occluded Scatterers and the Urban Ground-to-ground Channel at Low UHF","authors":"D. Breton, C. Haedrich","doi":"10.23919/USNC/URSI49741.2020.9321618","DOIUrl":"https://doi.org/10.23919/USNC/URSI49741.2020.9321618","url":null,"abstract":"Ground-to-ground radio links in urban environments rarely enjoy direct line-of-sight between terminals, and therefore in-canyon, over-rooftop, and scattering from distant structures become primary propagation modes. Because both rooftop diffraction and canyon propagation losses can be severe, and because the walls of deep urban canyons often occlude distant scatterers, the relative importance of these three propagation modes to a given urban channel is unclear. We present results of channel sounding measurements at 437 MHz for ground-to-ground links in Boston, Massachusetts, USA to quantify the importance of each propagation mode. Occupancy curves derived from our measured channels suggest that while canyon-mode propagation is dominant for short range urban links, the importance of the distant scatterer propagation mode increases with terminal separation distance, even when those scatterers are occluded from transmitter and/or receiver view. We present an urban channel model which evaluates the vertical profile of incident power on distant scatterers, even those that are occluded, and find that reasonable agreement can be obtained between measured and modeled channels only when occluded buildings are considered.","PeriodicalId":443426,"journal":{"name":"2020 IEEE USNC-CNC-URSI North American Radio Science Meeting (Joint with AP-S Symposium)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128647455","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 : 2020-07-05DOI: 10.23919/USNC/URSI49741.2020.9321656
C. Suer, Y. Park, R. Lang, C. Haedrich, D. Breton
Coherent wave attenuation is calculated for a tree covered mountainside at P and L bands. The layer of trees is represented as a set of discrete scatterers such as trunks, branches, leaves and needles of different sizes and orientations. The ground surface along the sloping axis is characterized using Kirchhoffs method. Ground truth measurements are done to acquire information about the scatterers. The attenuation and conversion of different types of polarizations are inferred. The effects of these findings will be used to further solve the bistatic scattering problem for the given sample of random media.
{"title":"P and L Band Coherent Wave Propagation through a Tree Covered Mountainside","authors":"C. Suer, Y. Park, R. Lang, C. Haedrich, D. Breton","doi":"10.23919/USNC/URSI49741.2020.9321656","DOIUrl":"https://doi.org/10.23919/USNC/URSI49741.2020.9321656","url":null,"abstract":"Coherent wave attenuation is calculated for a tree covered mountainside at P and L bands. The layer of trees is represented as a set of discrete scatterers such as trunks, branches, leaves and needles of different sizes and orientations. The ground surface along the sloping axis is characterized using Kirchhoffs method. Ground truth measurements are done to acquire information about the scatterers. The attenuation and conversion of different types of polarizations are inferred. The effects of these findings will be used to further solve the bistatic scattering problem for the given sample of random media.","PeriodicalId":443426,"journal":{"name":"2020 IEEE USNC-CNC-URSI North American Radio Science Meeting (Joint with AP-S Symposium)","volume":"110 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122255351","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 : 2020-07-05DOI: 10.23919/USNC/URSI49741.2020.9321624
Yisong Zhang, J. Simpson, D. Welling, M. Liemohn
Space weather can affect the Earth over time spans of hours and days. However, time-stepping increments for FDTD models are typically on the order of a fraction of a second. This paper introduces a means of increasing the time stepping increment’s upper limit by artificially slowing down the speed of light. Numerically slowing down the speed of light is achieved by appropriately modifying the permittivity, permeability, and conductivity values in the model. Proof-of-concept results are provided to show that the method works well for homogeneous media.
{"title":"Improving the Efficiency of Maxwell’s Equations FDTD Modeling for Space Weather Applications by Scaling the Speed of Light","authors":"Yisong Zhang, J. Simpson, D. Welling, M. Liemohn","doi":"10.23919/USNC/URSI49741.2020.9321624","DOIUrl":"https://doi.org/10.23919/USNC/URSI49741.2020.9321624","url":null,"abstract":"Space weather can affect the Earth over time spans of hours and days. However, time-stepping increments for FDTD models are typically on the order of a fraction of a second. This paper introduces a means of increasing the time stepping increment’s upper limit by artificially slowing down the speed of light. Numerically slowing down the speed of light is achieved by appropriately modifying the permittivity, permeability, and conductivity values in the model. Proof-of-concept results are provided to show that the method works well for homogeneous media.","PeriodicalId":443426,"journal":{"name":"2020 IEEE USNC-CNC-URSI North American Radio Science Meeting (Joint with AP-S Symposium)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122332974","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 : 2020-07-05DOI: 10.23919/USNC/URSI49741.2020.9321692
Qiqi Dai, B. Wen, Y. Lee, A. Yucel, Genevieve Ow, Mohamed Lokman Mohd Yusof
This paper proposes a deep learning-based approach for rapidly detecting the defects inside tree trunks via ground penetrating radar (GPR) technology. In this approach, GPR measurements are performed centimeters-away from the surface of tree trunk on a straight trajectory. The n the B-scans obtained from GPR measurements are processed via a deep learning algorithm to detect the defects inside the tree trunks, classify their types, and estimate their sizes/severities. An open-source finite-difference time-domain (FDTD) simulator is used to produce a large set of B-scans from random realizations of realistic 2D tree trunk cross-sections without and with different size of defects (cavities, decays, and cracks). The data set is then used to train and test a six-layer convolutional neural network (CNN) with drop-out layers and weight regularization to avoid overfitting. Our preliminary results show that the testing accuracy of the CNN algorithm is more than 90%. The testing results demonstrate that the current methodology al lows accurately detecting the types and sizes of defects inside tree trunks to monitor the health condition of trees.
{"title":"A Deep Learning-Based Methodology for Rapidly Detecting the Defects inside Tree Trunks via GPR","authors":"Qiqi Dai, B. Wen, Y. Lee, A. Yucel, Genevieve Ow, Mohamed Lokman Mohd Yusof","doi":"10.23919/USNC/URSI49741.2020.9321692","DOIUrl":"https://doi.org/10.23919/USNC/URSI49741.2020.9321692","url":null,"abstract":"This paper proposes a deep learning-based approach for rapidly detecting the defects inside tree trunks via ground penetrating radar (GPR) technology. In this approach, GPR measurements are performed centimeters-away from the surface of tree trunk on a straight trajectory. The n the B-scans obtained from GPR measurements are processed via a deep learning algorithm to detect the defects inside the tree trunks, classify their types, and estimate their sizes/severities. An open-source finite-difference time-domain (FDTD) simulator is used to produce a large set of B-scans from random realizations of realistic 2D tree trunk cross-sections without and with different size of defects (cavities, decays, and cracks). The data set is then used to train and test a six-layer convolutional neural network (CNN) with drop-out layers and weight regularization to avoid overfitting. Our preliminary results show that the testing accuracy of the CNN algorithm is more than 90%. The testing results demonstrate that the current methodology al lows accurately detecting the types and sizes of defects inside tree trunks to monitor the health condition of trees.","PeriodicalId":443426,"journal":{"name":"2020 IEEE USNC-CNC-URSI North American Radio Science Meeting (Joint with AP-S Symposium)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127302404","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 : 2020-07-05DOI: 10.23919/USNC/URSI49741.2020.9321631
Chenglong Wang, Junchao Ji, Xidong Wu, Jinfang Zhou
This paper presents a two-dimensional (2D) photonic crystal structure based on graphene surface plasmon polaritons (SPP) for terahertz wavelength division multiplexing (WDM) applications. By etching a periodic array of equal-diameter cylindrical holes with the same height in the ground, a periodic SPP effective index profile of 2D photonic crystal can be created on graphene with a single gate voltage between graphene and ground. Based on this uneven ground structure, a photonic crystal multimode interference (MMI) WDM has been demonstrated. Simulation results show that the designed device exhibits isolations of higher than 17.30 dB and bandwidth of 0.06 THz at both frequencies.
{"title":"A terahertz photonic crystal wavelength division multiplexer based on graphene surface plasmon polaritons","authors":"Chenglong Wang, Junchao Ji, Xidong Wu, Jinfang Zhou","doi":"10.23919/USNC/URSI49741.2020.9321631","DOIUrl":"https://doi.org/10.23919/USNC/URSI49741.2020.9321631","url":null,"abstract":"This paper presents a two-dimensional (2D) photonic crystal structure based on graphene surface plasmon polaritons (SPP) for terahertz wavelength division multiplexing (WDM) applications. By etching a periodic array of equal-diameter cylindrical holes with the same height in the ground, a periodic SPP effective index profile of 2D photonic crystal can be created on graphene with a single gate voltage between graphene and ground. Based on this uneven ground structure, a photonic crystal multimode interference (MMI) WDM has been demonstrated. Simulation results show that the designed device exhibits isolations of higher than 17.30 dB and bandwidth of 0.06 THz at both frequencies.","PeriodicalId":443426,"journal":{"name":"2020 IEEE USNC-CNC-URSI North American Radio Science Meeting (Joint with AP-S Symposium)","volume":"559 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131515279","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 : 2020-07-05DOI: 10.23919/USNC/URSI49741.2020.9321628
Hussain Ali, Tarig Ballal, T. Al-Naffouri, M. Sharawi
DOA estimation in the presence of coherent sources using a small number of snapshots faces the challenge of rank deficiency of the received signal covariance matrix. When the covariance matrix is rank deficient, only the pseudo inverse of the covariance matrix can be computed, which can give undesirable results. Traditionally, regularized least-squares (RLS) algorithms are used to tackle estimation problems in systems with ill-conditioned or rank deficient matrices. In this work, we combine the Capon beamformer with the RLS framework to develop a DOA estimation method for scenarios with rank deficient covariance matrices. Simulation results demonstrate the effectiveness of the proposed approach.
{"title":"DOA Estimation with a Rank-deficient Covariance matrix: A Regularized Least-squares approach","authors":"Hussain Ali, Tarig Ballal, T. Al-Naffouri, M. Sharawi","doi":"10.23919/USNC/URSI49741.2020.9321628","DOIUrl":"https://doi.org/10.23919/USNC/URSI49741.2020.9321628","url":null,"abstract":"DOA estimation in the presence of coherent sources using a small number of snapshots faces the challenge of rank deficiency of the received signal covariance matrix. When the covariance matrix is rank deficient, only the pseudo inverse of the covariance matrix can be computed, which can give undesirable results. Traditionally, regularized least-squares (RLS) algorithms are used to tackle estimation problems in systems with ill-conditioned or rank deficient matrices. In this work, we combine the Capon beamformer with the RLS framework to develop a DOA estimation method for scenarios with rank deficient covariance matrices. Simulation results demonstrate the effectiveness of the proposed approach.","PeriodicalId":443426,"journal":{"name":"2020 IEEE USNC-CNC-URSI North American Radio Science Meeting (Joint with AP-S Symposium)","volume":"134 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128637300","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 : 2020-07-05DOI: 10.23919/USNC/URSI49741.2020.9321666
Yanyan Hu, Yuchen Jin, Xuqing Wu, Jiefu Chen
Deep-learning techniques have been playing an increasingly important role for scientific modeling and simulations. Recent advances in high-performance tensor processing hardware and software are also providing new opportunities for accelerated linear algebra calculations. In this paper, we exploit a trainable recurrent neural network (RNN) to formulate the electromagnetic propagation and solve the Maxwell's equations on one of the most state-of-the-art differentiable programming platforms—Pytorch. Due to the specific performance-focused design of PyTorch, the computation efficiency is substantially improved compared to Matlab. Moreover, RNN-based implementation possesses potential advantages of leveraging the differentiable programming platform for varied applications that iterate around forward modeling, for example, uncertainty quantification, optimization, and inversion. Numerical simulation demonstrates the effectiveness of our method.
{"title":"Solving Time Domain Electromagnetic Problems using a Differentiable Programming Platform","authors":"Yanyan Hu, Yuchen Jin, Xuqing Wu, Jiefu Chen","doi":"10.23919/USNC/URSI49741.2020.9321666","DOIUrl":"https://doi.org/10.23919/USNC/URSI49741.2020.9321666","url":null,"abstract":"Deep-learning techniques have been playing an increasingly important role for scientific modeling and simulations. Recent advances in high-performance tensor processing hardware and software are also providing new opportunities for accelerated linear algebra calculations. In this paper, we exploit a trainable recurrent neural network (RNN) to formulate the electromagnetic propagation and solve the Maxwell's equations on one of the most state-of-the-art differentiable programming platforms—Pytorch. Due to the specific performance-focused design of PyTorch, the computation efficiency is substantially improved compared to Matlab. Moreover, RNN-based implementation possesses potential advantages of leveraging the differentiable programming platform for varied applications that iterate around forward modeling, for example, uncertainty quantification, optimization, and inversion. Numerical simulation demonstrates the effectiveness of our method.","PeriodicalId":443426,"journal":{"name":"2020 IEEE USNC-CNC-URSI North American Radio Science Meeting (Joint with AP-S Symposium)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117058344","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 : 2020-07-05DOI: 10.23919/USNC/URSI49741.2020.9321610
Anik Naha Biswas, Yee Hui Lee, Shilpa Manandhar
In this paper, the behavioral change in the atmospheric delay gradient during a rainfall event has been analyzed. The horizontal gradient of the delay results from the azimuthal asymmetry of the atmospheric refractivity. Resultant gradient estimated from two components (Gradient in the north direction and gradient in the east direction) manifests a consistent trend before precipitation. The gradient vector exhibits an increasing magnitude and a change in direction with rainfall in comparison to a non-rainy event. So, atmospheric delay gradient can be considered as a useful feature for rainfall forecasting by virtue of its clear pattern before rainfall which can be used to improve the prediction accuracy of rainfall event.
{"title":"Analysis of GPS Gradient Parameters for Rainfall Prediction","authors":"Anik Naha Biswas, Yee Hui Lee, Shilpa Manandhar","doi":"10.23919/USNC/URSI49741.2020.9321610","DOIUrl":"https://doi.org/10.23919/USNC/URSI49741.2020.9321610","url":null,"abstract":"In this paper, the behavioral change in the atmospheric delay gradient during a rainfall event has been analyzed. The horizontal gradient of the delay results from the azimuthal asymmetry of the atmospheric refractivity. Resultant gradient estimated from two components (Gradient in the north direction and gradient in the east direction) manifests a consistent trend before precipitation. The gradient vector exhibits an increasing magnitude and a change in direction with rainfall in comparison to a non-rainy event. So, atmospheric delay gradient can be considered as a useful feature for rainfall forecasting by virtue of its clear pattern before rainfall which can be used to improve the prediction accuracy of rainfall event.","PeriodicalId":443426,"journal":{"name":"2020 IEEE USNC-CNC-URSI North American Radio Science Meeting (Joint with AP-S Symposium)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114398193","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 : 2020-07-05DOI: 10.23919/USNC/URSI49741.2020.9321611
H. Krishnan, J. Kent, J. Dowell, Adam P Bearsdley, J. Bowman, G. Taylor, Nithyanandhan Thyagarajan, D. Jacobs
In this paper, we describe our efforts towards the development of a real-time radio imaging correlator for the Long-Wavelength Array station in Sevilleta, New Mexico. We briefly discuss the direct-imaging algorithm and present the architecture of the GPU implementation. We describe the code-level modifications carried out for one of the modules in the algorithm that improves GPU-memory management and highlight the performance improvements achieved through it. We emphasize our ongoing efforts in tuning the overall run-time duration of the correlator which in turn is expected to increase the operating bandwidth in order to address the demands of wide-band capability for radio transient science.
{"title":"Development of an Optimized Real-Time Radio Transient Imager for LWA-SV","authors":"H. Krishnan, J. Kent, J. Dowell, Adam P Bearsdley, J. Bowman, G. Taylor, Nithyanandhan Thyagarajan, D. Jacobs","doi":"10.23919/USNC/URSI49741.2020.9321611","DOIUrl":"https://doi.org/10.23919/USNC/URSI49741.2020.9321611","url":null,"abstract":"In this paper, we describe our efforts towards the development of a real-time radio imaging correlator for the Long-Wavelength Array station in Sevilleta, New Mexico. We briefly discuss the direct-imaging algorithm and present the architecture of the GPU implementation. We describe the code-level modifications carried out for one of the modules in the algorithm that improves GPU-memory management and highlight the performance improvements achieved through it. We emphasize our ongoing efforts in tuning the overall run-time duration of the correlator which in turn is expected to increase the operating bandwidth in order to address the demands of wide-band capability for radio transient science.","PeriodicalId":443426,"journal":{"name":"2020 IEEE USNC-CNC-URSI North American Radio Science Meeting (Joint with AP-S Symposium)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114884145","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 : 2020-07-05DOI: 10.23919/USNC/URSI49741.2020.9321625
Yibing Li, Sitong Zhang, Fang Ye, T. Jiang, Yingsong Li
The path planning of Unmanned Aerial Vehicle (UAV) is a critical component of rescue operation. As impacted by the continuity of the task space and the high dynamics of the aircraft, conventional approaches cannot find the optimal control strategy. Accordingly, in this study, a deep reinforcement learning (DRL)-based UAV path planning method is proposed, enabling the UAV to complete the path planning in a 3D continuous environment. The deep deterministic policy gradient (DDPG) algorithm is employed to enable UAV to autonomously make decisions. Besides, to avoid obstacles, the concepts of connected area and threat function are proposed and adopted in the reward shaping. Lastly, an environment with static obstacles is built, and the agent is trained using the proposed method. As has been proved by the experiments, the proposed algorithm can fit a range of scenarios.
{"title":"A UAV Path Planning Method Based on Deep Reinforcement Learning","authors":"Yibing Li, Sitong Zhang, Fang Ye, T. Jiang, Yingsong Li","doi":"10.23919/USNC/URSI49741.2020.9321625","DOIUrl":"https://doi.org/10.23919/USNC/URSI49741.2020.9321625","url":null,"abstract":"The path planning of Unmanned Aerial Vehicle (UAV) is a critical component of rescue operation. As impacted by the continuity of the task space and the high dynamics of the aircraft, conventional approaches cannot find the optimal control strategy. Accordingly, in this study, a deep reinforcement learning (DRL)-based UAV path planning method is proposed, enabling the UAV to complete the path planning in a 3D continuous environment. The deep deterministic policy gradient (DDPG) algorithm is employed to enable UAV to autonomously make decisions. Besides, to avoid obstacles, the concepts of connected area and threat function are proposed and adopted in the reward shaping. Lastly, an environment with static obstacles is built, and the agent is trained using the proposed method. As has been proved by the experiments, the proposed algorithm can fit a range of scenarios.","PeriodicalId":443426,"journal":{"name":"2020 IEEE USNC-CNC-URSI North American Radio Science Meeting (Joint with AP-S Symposium)","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126317173","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：