In the intelligent unmanned vehicles swarm systems, multi-sensor integrated navigation is commonly used to provide reliable positioning solutions. To correctly fuse the measurements from multiple sensors, their time should be precisely synchronized. Aiming at the problem of multi-sensor time synchronization, this paper proposes a time synchronization algorithm based on correlation analysis between velocities or accelerations measured by different sensors. Taking the Global Navigation Satellite System (GNSS) and the inertial navigation system (INS) as experimental object, a field test was conducted to verify the performance of the proposed algorithm in a GNSS/INS integrated navigation system. The results show that the algorithm can indeed calibrate the time transmission delay or clock source inconsistency between GNSS and INS within a certain range of misalignment. The accuracies of calibration are 95.71% and 99.64% under velocities and accelerations correlation analysis, respectively.
{"title":"A Time Synchronization Algorithm Based on Correlation Analysis in GNSS/INS Integrated Navigation","authors":"Haoli Zhang, Jiaju Guo, Xin Liu, Dezhong Zhou, Yanqing Hou","doi":"10.1109/ICUS55513.2022.9986909","DOIUrl":"https://doi.org/10.1109/ICUS55513.2022.9986909","url":null,"abstract":"In the intelligent unmanned vehicles swarm systems, multi-sensor integrated navigation is commonly used to provide reliable positioning solutions. To correctly fuse the measurements from multiple sensors, their time should be precisely synchronized. Aiming at the problem of multi-sensor time synchronization, this paper proposes a time synchronization algorithm based on correlation analysis between velocities or accelerations measured by different sensors. Taking the Global Navigation Satellite System (GNSS) and the inertial navigation system (INS) as experimental object, a field test was conducted to verify the performance of the proposed algorithm in a GNSS/INS integrated navigation system. The results show that the algorithm can indeed calibrate the time transmission delay or clock source inconsistency between GNSS and INS within a certain range of misalignment. The accuracies of calibration are 95.71% and 99.64% under velocities and accelerations correlation analysis, respectively.","PeriodicalId":345773,"journal":{"name":"2022 IEEE International Conference on Unmanned Systems (ICUS)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125660503","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 : 2022-10-28DOI: 10.1109/ICUS55513.2022.9987135
Yufeng Liao, Biyin Zhang, Yang Liu
The path planning of unmanned surface vehicle is the key to realize the intelligent driving of unmanned vehicle. Aiming at the problem of low search efficiency caused by the increase of vertex connections in the existing Visibility Graph, this paper presents a path planning algorithm based on Sparse Visibility Graph, which improves the planning efficiency of Visibility Graph by reducing the complexity of visibility graph and improving the search algorithm. Firstly, Sparse Visibility Graph construction is introduced, which reduces the complexity of the Visibility Graph by clipping unnecessary edges to reduce the degree of vertices. Secondly, the improved Lazy Theta* is introduced, the weighted valuation function is introduced to analyze the influence of the actual cost and the estimated cost on the planning effect. Aiming at the problem that the basic A * search path is constrained by the grid and the Theta* planning path is not optimal and the search efficiency is low. Through delayed the line-of-sight check and improvement in checking generations limits, the improved Lazy Theta* algorithm improves the efficiency of planning and the authenticity of the path. Finally, simulation experiments are carried out in a two-dimensional grid environment. The results show that, compared with the search algorithm based on Visibility Graph, the path planning based on Sparse Visibility Graph has a shorter search time, can achieve more efficient local path planning, and the path is more reasonable.
{"title":"USV Path Planning Based on Sparse Visibility Graph","authors":"Yufeng Liao, Biyin Zhang, Yang Liu","doi":"10.1109/ICUS55513.2022.9987135","DOIUrl":"https://doi.org/10.1109/ICUS55513.2022.9987135","url":null,"abstract":"The path planning of unmanned surface vehicle is the key to realize the intelligent driving of unmanned vehicle. Aiming at the problem of low search efficiency caused by the increase of vertex connections in the existing Visibility Graph, this paper presents a path planning algorithm based on Sparse Visibility Graph, which improves the planning efficiency of Visibility Graph by reducing the complexity of visibility graph and improving the search algorithm. Firstly, Sparse Visibility Graph construction is introduced, which reduces the complexity of the Visibility Graph by clipping unnecessary edges to reduce the degree of vertices. Secondly, the improved Lazy Theta* is introduced, the weighted valuation function is introduced to analyze the influence of the actual cost and the estimated cost on the planning effect. Aiming at the problem that the basic A * search path is constrained by the grid and the Theta* planning path is not optimal and the search efficiency is low. Through delayed the line-of-sight check and improvement in checking generations limits, the improved Lazy Theta* algorithm improves the efficiency of planning and the authenticity of the path. Finally, simulation experiments are carried out in a two-dimensional grid environment. The results show that, compared with the search algorithm based on Visibility Graph, the path planning based on Sparse Visibility Graph has a shorter search time, can achieve more efficient local path planning, and the path is more reasonable.","PeriodicalId":345773,"journal":{"name":"2022 IEEE International Conference on Unmanned Systems (ICUS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131079436","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 : 2022-10-28DOI: 10.1109/ICUS55513.2022.9986968
Lin Zhou, Jinbiao Zhu, Jihao Yang, Jie Geng
Hyperspectral image classification is a prominent topic in many remote sensing applications, but the limited number of manually annotated samples leads to performance bottlenecks. To resolve this issue, a data augmentation and spatial-spectral residual framework is proposed for hyperspectral image classification using limited samples. Firstly, an unsupervised pseudo-sample generation method is proposed to augment the sample set, and the generalization capability of the model is improved by mixup operations. Then, to adequately extract the spatial-spectral features of hyperspectral images, a spatial-spectral residual framework is designed to improve the classification performance of the model. The qualitative and quantitative experiments were carried out on Indian Pines dataset to validate the effectiveness of the model.
{"title":"Data Augmentation and Spatial-Spectral Residual Framework for Hyperspectral Image Classification Using Limited Samples","authors":"Lin Zhou, Jinbiao Zhu, Jihao Yang, Jie Geng","doi":"10.1109/ICUS55513.2022.9986968","DOIUrl":"https://doi.org/10.1109/ICUS55513.2022.9986968","url":null,"abstract":"Hyperspectral image classification is a prominent topic in many remote sensing applications, but the limited number of manually annotated samples leads to performance bottlenecks. To resolve this issue, a data augmentation and spatial-spectral residual framework is proposed for hyperspectral image classification using limited samples. Firstly, an unsupervised pseudo-sample generation method is proposed to augment the sample set, and the generalization capability of the model is improved by mixup operations. Then, to adequately extract the spatial-spectral features of hyperspectral images, a spatial-spectral residual framework is designed to improve the classification performance of the model. The qualitative and quantitative experiments were carried out on Indian Pines dataset to validate the effectiveness of the model.","PeriodicalId":345773,"journal":{"name":"2022 IEEE International Conference on Unmanned Systems (ICUS)","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132102502","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 : 2022-10-28DOI: 10.1109/ICUS55513.2022.9986843
Duo Zhang, Qi Zhang
The small-sized liquid damping time delay mechanism has the problems of large computational volume in the dynamic CFD simulation. Based on the establishment of the dynamic model of the damper, this paper analyzes the relationship between the liquid damping force and the influence factors such as the orifice hole length, temperature, and piston movement speed through the combination of CFX static simulation and CFD dynamic simulation. The parametric representation of damping force and influencing factors is obtained by data fitting and correction. The fitting model is tested under multiple working conditions, which shows the effectiveness of the method.
{"title":"Parametric Approximation Method for Dynamic Process of Small-sized Damping Time Delay Mechanism","authors":"Duo Zhang, Qi Zhang","doi":"10.1109/ICUS55513.2022.9986843","DOIUrl":"https://doi.org/10.1109/ICUS55513.2022.9986843","url":null,"abstract":"The small-sized liquid damping time delay mechanism has the problems of large computational volume in the dynamic CFD simulation. Based on the establishment of the dynamic model of the damper, this paper analyzes the relationship between the liquid damping force and the influence factors such as the orifice hole length, temperature, and piston movement speed through the combination of CFX static simulation and CFD dynamic simulation. The parametric representation of damping force and influencing factors is obtained by data fitting and correction. The fitting model is tested under multiple working conditions, which shows the effectiveness of the method.","PeriodicalId":345773,"journal":{"name":"2022 IEEE International Conference on Unmanned Systems (ICUS)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130899135","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 : 2022-10-28DOI: 10.1109/ICUS55513.2022.9987086
Ping Zheng, Yuezu Lv
With increasing uncertainties in power systems, reinforcement learning evolves as a promising approach for decision and control problems. This paper focuses on the unit commitment and dispatch problem, with startup and shutdown power trajectories involved, investigating it via reinforcement learning. First, we convert the problem into a Markov decision process, where constraints are tackled by projections and elaborate reward. Then, to cope with discrete commitment actions and continuous power outputs simultaneously, a discrete-continuous reinforcement learning algorithm is proposed by combining deep Q-network with soft actor-critic algorithm. Finally, numerical examples are done, verifying the effectiveness of the presented algorithm.
{"title":"A Discrete-Continuous Reinforcement Learning Algorithm for Unit Commitment and Dispatch Problem","authors":"Ping Zheng, Yuezu Lv","doi":"10.1109/ICUS55513.2022.9987086","DOIUrl":"https://doi.org/10.1109/ICUS55513.2022.9987086","url":null,"abstract":"With increasing uncertainties in power systems, reinforcement learning evolves as a promising approach for decision and control problems. This paper focuses on the unit commitment and dispatch problem, with startup and shutdown power trajectories involved, investigating it via reinforcement learning. First, we convert the problem into a Markov decision process, where constraints are tackled by projections and elaborate reward. Then, to cope with discrete commitment actions and continuous power outputs simultaneously, a discrete-continuous reinforcement learning algorithm is proposed by combining deep Q-network with soft actor-critic algorithm. Finally, numerical examples are done, verifying the effectiveness of the presented algorithm.","PeriodicalId":345773,"journal":{"name":"2022 IEEE International Conference on Unmanned Systems (ICUS)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116140023","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 : 2022-10-28DOI: 10.1109/ICUS55513.2022.9986846
Sainan Li, Anxue Guo, Hann-Tzong Su, Tianfeng Fan, H. Tan
In order to form stable aggregation and hovering states near the target in reconnaissance or jamming tasks, this paper proposes a swarm strategy for unmanned aerial vehicles based on target guidance control. Firstly, establish the interaction rules and the swarm aggregation control model. Furthermore, considering different initial states of them, the target guidance control model is designed from the normal and tangential directions respectively. Finally, combining with the swarm aggregation and the target tendency, the proposed strategy is verified by simulation studies.
{"title":"Swarming Strategy of Unmanned Aerial Vehicles Based on Target Guidance Control","authors":"Sainan Li, Anxue Guo, Hann-Tzong Su, Tianfeng Fan, H. Tan","doi":"10.1109/ICUS55513.2022.9986846","DOIUrl":"https://doi.org/10.1109/ICUS55513.2022.9986846","url":null,"abstract":"In order to form stable aggregation and hovering states near the target in reconnaissance or jamming tasks, this paper proposes a swarm strategy for unmanned aerial vehicles based on target guidance control. Firstly, establish the interaction rules and the swarm aggregation control model. Furthermore, considering different initial states of them, the target guidance control model is designed from the normal and tangential directions respectively. Finally, combining with the swarm aggregation and the target tendency, the proposed strategy is verified by simulation studies.","PeriodicalId":345773,"journal":{"name":"2022 IEEE International Conference on Unmanned Systems (ICUS)","volume":"352 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115978247","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 : 2022-10-28DOI: 10.1109/ICUS55513.2022.9986558
Di Xu, Yanchen Liu, Zhijun Liu, Zhengjie Wang
The unmanned aerial vehicle (UAV) swarm cannot receive GNSS signals in the navigation denial environment, and the loss of GNSS signal calibration fusion of the positional data will result in excessive errors leading to the loss of the swarm's positioning capability. To address the problem of GNSS signal loss and navigation failure in this rejection environment, a cooperative control method based on inter-UAV communication range observation information is proposed. Firstly this method obtains the distance information between nodes within the cluster through inter-UAV communication. Then the range information and part of the inertial data are used to establish the relative coordinate system and the information of the UAV centralized configuration under the relative coordinate system, and then the barometric altitude information and magnetic compass information of each UAV are used to maintain the UAV heading, altitude and cluster configuration, so as to improve the survivability of the UAV swarm in the navigation denial environment. Finally, the effectiveness of the method is experimentally verified.
{"title":"Cooperative Multi-UAVs Configuration Maintenance Based on Inter-aircraft Ranging in Navigation Denial Environment","authors":"Di Xu, Yanchen Liu, Zhijun Liu, Zhengjie Wang","doi":"10.1109/ICUS55513.2022.9986558","DOIUrl":"https://doi.org/10.1109/ICUS55513.2022.9986558","url":null,"abstract":"The unmanned aerial vehicle (UAV) swarm cannot receive GNSS signals in the navigation denial environment, and the loss of GNSS signal calibration fusion of the positional data will result in excessive errors leading to the loss of the swarm's positioning capability. To address the problem of GNSS signal loss and navigation failure in this rejection environment, a cooperative control method based on inter-UAV communication range observation information is proposed. Firstly this method obtains the distance information between nodes within the cluster through inter-UAV communication. Then the range information and part of the inertial data are used to establish the relative coordinate system and the information of the UAV centralized configuration under the relative coordinate system, and then the barometric altitude information and magnetic compass information of each UAV are used to maintain the UAV heading, altitude and cluster configuration, so as to improve the survivability of the UAV swarm in the navigation denial environment. Finally, the effectiveness of the method is experimentally verified.","PeriodicalId":345773,"journal":{"name":"2022 IEEE International Conference on Unmanned Systems (ICUS)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131441714","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 : 2022-10-28DOI: 10.1109/icus55513.2022.9987194
Zhenwei Ma, Lin Chen, Jinbo Wang, Hongbo Chen
This paper proposes a robust adaptive global control approach for quadrotor aircraft formation system based on RBF neural network against parametric uncertainties and bounded external disturbances for the quadrotor aircraft sys-tem. The actual controller consists of neural network controller in the approximate domain and robust controller outside the approximate domain. In order to ensure that all the signals of the closed-loop system are globally consistent and ultimately bounded, a smooth switching function is introduced to realize the smooth switching among controllers. What is more, Lyapunov function and Barbalat lemma are used to prove the stability of the nonlinear quadrotor aircraft formation system, and analyze the system stability strictly. Finally, we apply the proposed controller in MATLAB and Simulink software platforms, and analyze the numerical results.
{"title":"Trajectory Tracking Control for Quadrotor Formation Subject to Environmental and Model Uncertainties","authors":"Zhenwei Ma, Lin Chen, Jinbo Wang, Hongbo Chen","doi":"10.1109/icus55513.2022.9987194","DOIUrl":"https://doi.org/10.1109/icus55513.2022.9987194","url":null,"abstract":"This paper proposes a robust adaptive global control approach for quadrotor aircraft formation system based on RBF neural network against parametric uncertainties and bounded external disturbances for the quadrotor aircraft sys-tem. The actual controller consists of neural network controller in the approximate domain and robust controller outside the approximate domain. In order to ensure that all the signals of the closed-loop system are globally consistent and ultimately bounded, a smooth switching function is introduced to realize the smooth switching among controllers. What is more, Lyapunov function and Barbalat lemma are used to prove the stability of the nonlinear quadrotor aircraft formation system, and analyze the system stability strictly. Finally, we apply the proposed controller in MATLAB and Simulink software platforms, and analyze the numerical results.","PeriodicalId":345773,"journal":{"name":"2022 IEEE International Conference on Unmanned Systems (ICUS)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127790389","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 : 2022-10-28DOI: 10.1109/ICUS55513.2022.9986888
Juntao Zhao, Xiaochuan Luo
A three-dimensional (3D) path planning method based on the improved mayfly algorithm (IMA) is proposed in this paper for the unmanned aerial vehicle (UAV) path planning problem under the condition of diverse static features and obstacle threats. Firstly, the 3D flight area environment model with obstacles is built. Then, the path planning method is developed, which can increase the global search capability by keeping population diversity with the improved Tent chaotic map, and balance the global and local searching capability through incorporating the dynamic adaptive inertia weight into the algorithm. In addition, Gaussian mutation strategy is used to increase the solution accuracy and the ability of the algorithm jumping out from the local optimum. Finally, the optimal collision-free flight path is obtained by smoothing the planned path using the cubic B-spline curve. Results show that the developed algorithm can plan a smooth flight path, and avoid obstacle threats.
{"title":"Three-dimensional Path Planning for Unmanned Aerial Vehicle (UAV) Based on Improved Mayfly Algorithm","authors":"Juntao Zhao, Xiaochuan Luo","doi":"10.1109/ICUS55513.2022.9986888","DOIUrl":"https://doi.org/10.1109/ICUS55513.2022.9986888","url":null,"abstract":"A three-dimensional (3D) path planning method based on the improved mayfly algorithm (IMA) is proposed in this paper for the unmanned aerial vehicle (UAV) path planning problem under the condition of diverse static features and obstacle threats. Firstly, the 3D flight area environment model with obstacles is built. Then, the path planning method is developed, which can increase the global search capability by keeping population diversity with the improved Tent chaotic map, and balance the global and local searching capability through incorporating the dynamic adaptive inertia weight into the algorithm. In addition, Gaussian mutation strategy is used to increase the solution accuracy and the ability of the algorithm jumping out from the local optimum. Finally, the optimal collision-free flight path is obtained by smoothing the planned path using the cubic B-spline curve. Results show that the developed algorithm can plan a smooth flight path, and avoid obstacle threats.","PeriodicalId":345773,"journal":{"name":"2022 IEEE International Conference on Unmanned Systems (ICUS)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134274797","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 : 2022-10-28DOI: 10.1109/ICUS55513.2022.9986900
Chengcheng Wang, Yulong Wang, Chen Peng
The purpose of this paper is to discuss distributed cooperative target tracking for a multi-unmanned surface vehicle (multi-USV) system. The cooperative target tracking problem is formulated as a multi-USV learning problem. Based on this formulation, a multi-USV distributed cooperative target tracking (MUTT) algorithm is proposed. To avoid the collisions between USVs during the tracking process, an additional safety layer is introduced. Some safety signals are constructed based on USVs' states. By correcting actions through the trained safety layer, USVs can avoid collisions reasonably. Moreover, for the sake of demonstrating the effectiveness of the proposed MUTT algorithm in target tracking, reward functions and mission scenarios are well constructed. Furthermore, a comparison of the MUTT algorithm and Multi-Agent Deep Deterministic Policy Gradient (MADDPG) algorithm is given. The obtained results manifest that the proposed MUTT algorithm provides safe policies for multi-USV cooperative target tracking tasks.
{"title":"Multi-USV Deep Reinforcement Learning for Distributed Cooperative Target Tracking","authors":"Chengcheng Wang, Yulong Wang, Chen Peng","doi":"10.1109/ICUS55513.2022.9986900","DOIUrl":"https://doi.org/10.1109/ICUS55513.2022.9986900","url":null,"abstract":"The purpose of this paper is to discuss distributed cooperative target tracking for a multi-unmanned surface vehicle (multi-USV) system. The cooperative target tracking problem is formulated as a multi-USV learning problem. Based on this formulation, a multi-USV distributed cooperative target tracking (MUTT) algorithm is proposed. To avoid the collisions between USVs during the tracking process, an additional safety layer is introduced. Some safety signals are constructed based on USVs' states. By correcting actions through the trained safety layer, USVs can avoid collisions reasonably. Moreover, for the sake of demonstrating the effectiveness of the proposed MUTT algorithm in target tracking, reward functions and mission scenarios are well constructed. Furthermore, a comparison of the MUTT algorithm and Multi-Agent Deep Deterministic Policy Gradient (MADDPG) algorithm is given. The obtained results manifest that the proposed MUTT algorithm provides safe policies for multi-USV cooperative target tracking tasks.","PeriodicalId":345773,"journal":{"name":"2022 IEEE International Conference on Unmanned Systems (ICUS)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115030681","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：