Fish image classification on small-scale datasets is a classical fine-grained classification problem with more challenging than common classification problems since popular Convolutional Neural Networks (CNNs) always need massive labeled images to achieve best effects. This paper presents a method for fine-grained fish image classification on small-scale dataset by improving transfer learning with Bilinear Convolutional Neural Networks (B-CNNs). Contrast to the popular CNNs for image classification such as VGG net or ResNet, our method is capable of classifying images with insufficient training data. We evaluate the performance of our method and state-of-the-art CNNs on a small-scale fine-grained dataset (Croatian fish dataset) with only about 10 samples per category. The experimental results show that our transfer learning method can greatly enhance the accuracy on four popular CNNs. In particular, the method on Inception-v4 with 78.25
{"title":"Transfer Learning for Small-Scale Fish Image Classification","authors":"Chenchen Qiu, J. Cui, Shaoyong Zhang, Chao Wang, Zhaorui Gu, Haiyong Zheng, Bing Zheng","doi":"10.1109/OCEANSKOBE.2018.8559473","DOIUrl":"https://doi.org/10.1109/OCEANSKOBE.2018.8559473","url":null,"abstract":"Fish image classification on small-scale datasets is a classical fine-grained classification problem with more challenging than common classification problems since popular Convolutional Neural Networks (CNNs) always need massive labeled images to achieve best effects. This paper presents a method for fine-grained fish image classification on small-scale dataset by improving transfer learning with Bilinear Convolutional Neural Networks (B-CNNs). Contrast to the popular CNNs for image classification such as VGG net or ResNet, our method is capable of classifying images with insufficient training data. We evaluate the performance of our method and state-of-the-art CNNs on a small-scale fine-grained dataset (Croatian fish dataset) with only about 10 samples per category. The experimental results show that our transfer learning method can greatly enhance the accuracy on four popular CNNs. In particular, the method on Inception-v4 with 78.25","PeriodicalId":441405,"journal":{"name":"2018 OCEANS - MTS/IEEE Kobe Techno-Oceans (OTO)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114585581","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 : 2018-05-28DOI: 10.1109/OCEANSKOBE.2018.8558859
Barbara Arbanas, T. Petrović, S. Bogdan
In this paper, we present a newly developed consensus protocol for an underwater multi-robot system using acoustic communication. The approach is motivated by a system in which robots, called aMussels, are dispersed in the environment and measure its properties over longer periods of time. They communicate via acoustic modems using a simple sequenced communication protocol. The classical average consensus protocol has been modified to suit these conditions. The performance of the algorithm is analyzed in a simulation testbed, where we examine its properties and show the convergence of the method. The protocol is then applied to the real system in an underwater experiment, where we demonstrate its convergence in realistic conditions.
{"title":"Consensus Protocol for Underwater Multi-Robot System Using Scheduled Acoustic Communication","authors":"Barbara Arbanas, T. Petrović, S. Bogdan","doi":"10.1109/OCEANSKOBE.2018.8558859","DOIUrl":"https://doi.org/10.1109/OCEANSKOBE.2018.8558859","url":null,"abstract":"In this paper, we present a newly developed consensus protocol for an underwater multi-robot system using acoustic communication. The approach is motivated by a system in which robots, called aMussels, are dispersed in the environment and measure its properties over longer periods of time. They communicate via acoustic modems using a simple sequenced communication protocol. The classical average consensus protocol has been modified to suit these conditions. The performance of the algorithm is analyzed in a simulation testbed, where we examine its properties and show the convergence of the method. The protocol is then applied to the real system in an underwater experiment, where we demonstrate its convergence in realistic conditions.","PeriodicalId":441405,"journal":{"name":"2018 OCEANS - MTS/IEEE Kobe Techno-Oceans (OTO)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117210884","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 : 2018-05-28DOI: 10.1109/OCEANSKOBE.2018.8559262
R. Petroccia, J. Sliwka, A. Grati, V. Grandi, P. Guerrini, A. Munafò, Marin Stipanov, J. Alves, R. Been
This paper presents the experimental activities performed by the NATO STO Centre for Maritime Research and Experimentation (CMRE) during the CommsNet17 trial where a persistent Underwater Acoustic Sensor Network (UASN) was deployed. The CommsNet17 trial was held from the 27th of November to the 6th of December in the Gulf of La Spezia (IT), close to the CMRE premises, using the CMRE Littoral Ocean Observatory Network (LOON) as one of its key components. A network consisting of up to eleven nodes was deployed, including static and mobile assets. Various aspects related to persistent UASNs were addressed, including autonomous and distributed network discovery and node configuration, node localisation and navigation, self-adjustment of the network topology in support to the assigned tasks, underwater docking, wireless battery recharging and data offloading. The collected results show that the employed solutions were able to successfully complete all these tasks, thus demonstrating the effective deployment of a persistent, distributed and ad-hoc UASN.
{"title":"Deployment of a Persistent Underwater Acoustic Sensor Network: The CommsNet17 Experience","authors":"R. Petroccia, J. Sliwka, A. Grati, V. Grandi, P. Guerrini, A. Munafò, Marin Stipanov, J. Alves, R. Been","doi":"10.1109/OCEANSKOBE.2018.8559262","DOIUrl":"https://doi.org/10.1109/OCEANSKOBE.2018.8559262","url":null,"abstract":"This paper presents the experimental activities performed by the NATO STO Centre for Maritime Research and Experimentation (CMRE) during the CommsNet17 trial where a persistent Underwater Acoustic Sensor Network (UASN) was deployed. The CommsNet17 trial was held from the 27th of November to the 6th of December in the Gulf of La Spezia (IT), close to the CMRE premises, using the CMRE Littoral Ocean Observatory Network (LOON) as one of its key components. A network consisting of up to eleven nodes was deployed, including static and mobile assets. Various aspects related to persistent UASNs were addressed, including autonomous and distributed network discovery and node configuration, node localisation and navigation, self-adjustment of the network topology in support to the assigned tasks, underwater docking, wireless battery recharging and data offloading. The collected results show that the employed solutions were able to successfully complete all these tasks, thus demonstrating the effective deployment of a persistent, distributed and ad-hoc UASN.","PeriodicalId":441405,"journal":{"name":"2018 OCEANS - MTS/IEEE Kobe Techno-Oceans (OTO)","volume":"128 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129384780","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 : 2018-05-28DOI: 10.1109/OCEANSKOBE.2018.8559246
Zhenyu Jia, Fei Yuan, Keyu Chen, En Cheng, Jianghui Li
Underwater acoustic (UWA) channels are characterized by narrow-band, time-varying, and strong multipath interference. At the same time, Doppler effect also has a significant impact in UWA communications due to low sound speed in water. Therefore, traditional wireless multiple access technology cannot be used directly underwater. However, by combining some existing technologies in underwater communications, the traditional multiple access method can be improved to enable underwater multiple access. In this paper, we introduced chirp signals as carriers, which is widely used in UWA communications. Combined with code division multiple access (CDMA) and orthogonal frequency division multiple access (OFDMA) techniques, a multi-carrier chirp signal based CDMA method is proposed, which improves the reliability and spectrum efficiency of underwater multiple access. The simulation results show that the proposed method has good Bit error rate (BER) performance. Further, the method is also able to achieve good BER performance under the influence of Doppler effect because chirp signals is capable of resisting Doppler effects.
{"title":"Research on Code Division Multiple Access Based on Chirp Multi-Carrier","authors":"Zhenyu Jia, Fei Yuan, Keyu Chen, En Cheng, Jianghui Li","doi":"10.1109/OCEANSKOBE.2018.8559246","DOIUrl":"https://doi.org/10.1109/OCEANSKOBE.2018.8559246","url":null,"abstract":"Underwater acoustic (UWA) channels are characterized by narrow-band, time-varying, and strong multipath interference. At the same time, Doppler effect also has a significant impact in UWA communications due to low sound speed in water. Therefore, traditional wireless multiple access technology cannot be used directly underwater. However, by combining some existing technologies in underwater communications, the traditional multiple access method can be improved to enable underwater multiple access. In this paper, we introduced chirp signals as carriers, which is widely used in UWA communications. Combined with code division multiple access (CDMA) and orthogonal frequency division multiple access (OFDMA) techniques, a multi-carrier chirp signal based CDMA method is proposed, which improves the reliability and spectrum efficiency of underwater multiple access. The simulation results show that the proposed method has good Bit error rate (BER) performance. Further, the method is also able to achieve good BER performance under the influence of Doppler effect because chirp signals is capable of resisting Doppler effects.","PeriodicalId":441405,"journal":{"name":"2018 OCEANS - MTS/IEEE Kobe Techno-Oceans (OTO)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129469202","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 : 2018-05-28DOI: 10.1109/OCEANSKOBE.2018.8558802
José Pinto, R. Mendes, J. da Silva, J. Dias, J. D. de Sousa
This paper presents a novel approach for studying riverine plumes. It combines remote-sensing with autonomous in-situ observations performed by a set of unmanned vehicles (underwater and aerial). While an aerial drone is remotely piloted to detect the plume from air, multiple underwater vehicles autonomously detect the plume edge and map their structure by traversing it multiple times, autonomously. This approach has been used multiple times to acquire the 3D structure of the Douro river plume, in Porto.
{"title":"Multiple Autonomous Vehicles Applied to Plume Detection and Tracking","authors":"José Pinto, R. Mendes, J. da Silva, J. Dias, J. D. de Sousa","doi":"10.1109/OCEANSKOBE.2018.8558802","DOIUrl":"https://doi.org/10.1109/OCEANSKOBE.2018.8558802","url":null,"abstract":"This paper presents a novel approach for studying riverine plumes. It combines remote-sensing with autonomous in-situ observations performed by a set of unmanned vehicles (underwater and aerial). While an aerial drone is remotely piloted to detect the plume from air, multiple underwater vehicles autonomously detect the plume edge and map their structure by traversing it multiple times, autonomously. This approach has been used multiple times to acquire the 3D structure of the Douro river plume, in Porto.","PeriodicalId":441405,"journal":{"name":"2018 OCEANS - MTS/IEEE Kobe Techno-Oceans (OTO)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130885450","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 : 2018-05-28DOI: 10.1109/OCEANSKOBE.2018.8559461
Yishan Su, Yongpeng Zuo, Yun Li, Zhigang Jin, Xiaomei Fu
During the past decade, the research of underwater acoustic sensors network have gained significant progress. With the deployment of UASNs, the acquisition and long-range transmission of underwater data becomes a challenge. Traditionally, the data from underwater sensors can be sent to terrestrial control center mainly by VHF (very high frequency) or satellite communication systems. However, a VHF communication system can only cover the area with a distance of 15-30km and a satellite communication system may cost a lot especially for a long-time-deployed sensor network. To effectively evaluate the performance of underwater networks and further transmit underwater data from open sea, in this paper, we develop an underwater data acquisition and transmission testbed based on BeiDou Satellite System (BDS) and underwater acoustic communication technology. With the help of the proposed testbed, we can conduct various tests including protocols evaluations and underwater application developments.
{"title":"An Underwater Data Acquisition and Transmission Testbed Based on BeiDou Satellite System (BDS) and Underwater Acoustic Communication Technology","authors":"Yishan Su, Yongpeng Zuo, Yun Li, Zhigang Jin, Xiaomei Fu","doi":"10.1109/OCEANSKOBE.2018.8559461","DOIUrl":"https://doi.org/10.1109/OCEANSKOBE.2018.8559461","url":null,"abstract":"During the past decade, the research of underwater acoustic sensors network have gained significant progress. With the deployment of UASNs, the acquisition and long-range transmission of underwater data becomes a challenge. Traditionally, the data from underwater sensors can be sent to terrestrial control center mainly by VHF (very high frequency) or satellite communication systems. However, a VHF communication system can only cover the area with a distance of 15-30km and a satellite communication system may cost a lot especially for a long-time-deployed sensor network. To effectively evaluate the performance of underwater networks and further transmit underwater data from open sea, in this paper, we develop an underwater data acquisition and transmission testbed based on BeiDou Satellite System (BDS) and underwater acoustic communication technology. With the help of the proposed testbed, we can conduct various tests including protocols evaluations and underwater application developments.","PeriodicalId":441405,"journal":{"name":"2018 OCEANS - MTS/IEEE Kobe Techno-Oceans (OTO)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132724360","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 : 2018-05-28DOI: 10.1109/OCEANSKOBE.2018.8559237
A. Want, R. Harris, J. Porter
Generating electricity from marine energy is a globally important industry and forms a vital part of several nations' objectives to decarbonise energy production. In the UK, the resource-rich waters around Scotland have led to world-leading developments of facilities and supporting infrastructure used in deploying and testing wave and tidal energy technologies. A major concern to industries working in the marine environment is biofouling on submerged structures, including energy converters and measurement instrumentation. In this study, marine energy devices, infrastructure, and instrumentation were surveyed to characterise biofouling organisms. Fouling communities varied between deployment habitats; key organisms were identified allowing recommendations for scheduling device maintenance and preventing spread of invasive species. A method to measure biofouling impact on hydrodynamic response is described and applied to data from a wave-monitoring buoy deployed at a test site in Scotland. Results are discussed in relation to measurement accuracy of resources for power generation. Further applications are suggested for future testing in other scenarios including habitats use for extracting tidal energy.
{"title":"Biodiversity Characterisation of Fouling Communities and Their Hydrodynamic Consequences on Marine Renewable Energy Infrastructure in the UK","authors":"A. Want, R. Harris, J. Porter","doi":"10.1109/OCEANSKOBE.2018.8559237","DOIUrl":"https://doi.org/10.1109/OCEANSKOBE.2018.8559237","url":null,"abstract":"Generating electricity from marine energy is a globally important industry and forms a vital part of several nations' objectives to decarbonise energy production. In the UK, the resource-rich waters around Scotland have led to world-leading developments of facilities and supporting infrastructure used in deploying and testing wave and tidal energy technologies. A major concern to industries working in the marine environment is biofouling on submerged structures, including energy converters and measurement instrumentation. In this study, marine energy devices, infrastructure, and instrumentation were surveyed to characterise biofouling organisms. Fouling communities varied between deployment habitats; key organisms were identified allowing recommendations for scheduling device maintenance and preventing spread of invasive species. A method to measure biofouling impact on hydrodynamic response is described and applied to data from a wave-monitoring buoy deployed at a test site in Scotland. Results are discussed in relation to measurement accuracy of resources for power generation. Further applications are suggested for future testing in other scenarios including habitats use for extracting tidal energy.","PeriodicalId":441405,"journal":{"name":"2018 OCEANS - MTS/IEEE Kobe Techno-Oceans (OTO)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133563856","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 : 2018-05-28DOI: 10.1109/OCEANSKOBE.2018.8559047
Xue Cheng, Yingmin Wang
Direction of Arrival (DOA) for multiple-input multiple-output sonar is considered. A high dimensional matrix by matched filter can be obtained in the receive array, when we directly apply traditional algorithm to MIMO Sonar. This matrix loss the original incoherence, and produces higher computational complexity. In this paper, reduced-complex multiple signal classification (MUSIC) based on Toeplitz reconstruction (TRC-MUSIC) is proposed. This algorithm can reduce the dimension of operation effectively, avoid degrees of freedom (DOFs) and virtual array aperture loss by traditional decorrelation algorithm. We can obtain better location performance in the situation of low signal-to-noise ratio (SNR) and lower snapshots. Simulation results show the usefulness of the proposed algorithm. The TRC-MUSIC method can provide more accurate location and better stability, suppress noise effectively in low snapshots.
{"title":"A Reduced-Complex Method Based on Toeplitz Reconstruction for Direction of Arrival Estimation in Multiple-Input Multiple-Output Sonar","authors":"Xue Cheng, Yingmin Wang","doi":"10.1109/OCEANSKOBE.2018.8559047","DOIUrl":"https://doi.org/10.1109/OCEANSKOBE.2018.8559047","url":null,"abstract":"Direction of Arrival (DOA) for multiple-input multiple-output sonar is considered. A high dimensional matrix by matched filter can be obtained in the receive array, when we directly apply traditional algorithm to MIMO Sonar. This matrix loss the original incoherence, and produces higher computational complexity. In this paper, reduced-complex multiple signal classification (MUSIC) based on Toeplitz reconstruction (TRC-MUSIC) is proposed. This algorithm can reduce the dimension of operation effectively, avoid degrees of freedom (DOFs) and virtual array aperture loss by traditional decorrelation algorithm. We can obtain better location performance in the situation of low signal-to-noise ratio (SNR) and lower snapshots. Simulation results show the usefulness of the proposed algorithm. The TRC-MUSIC method can provide more accurate location and better stability, suppress noise effectively in low snapshots.","PeriodicalId":441405,"journal":{"name":"2018 OCEANS - MTS/IEEE Kobe Techno-Oceans (OTO)","volume":"100 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130513769","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 : 2018-05-28DOI: 10.1109/OCEANSKOBE.2018.8559092
C. Chi, H. Vishnu, K. T. Beng
Broadband acoustic Doppler current profilers (ADCPs) are instruments that are widely used in underwater observation. However, conventional broadband ADCPs face the problem of being limited in accuracy due to velocity ambiguity. This paper proposes a method based on using orthogonal coprime pulse pairs and the robust Chinese remainder theorem to mitigate this limitation. The proposed method breaks through the limit of velocity ambiguity of conventional broadband ADCP and improves the performance significantly. The simulations show that our proposed method can decrease the standard deviation of current velocity measurement by nearly three time, when compared to the conventional method.
{"title":"Improving Broadband Acoustic Doppler Current Profiler with Orthogonal Coprime Pulse Pairs and Robust Chinese Remainder Theorem","authors":"C. Chi, H. Vishnu, K. T. Beng","doi":"10.1109/OCEANSKOBE.2018.8559092","DOIUrl":"https://doi.org/10.1109/OCEANSKOBE.2018.8559092","url":null,"abstract":"Broadband acoustic Doppler current profilers (ADCPs) are instruments that are widely used in underwater observation. However, conventional broadband ADCPs face the problem of being limited in accuracy due to velocity ambiguity. This paper proposes a method based on using orthogonal coprime pulse pairs and the robust Chinese remainder theorem to mitigate this limitation. The proposed method breaks through the limit of velocity ambiguity of conventional broadband ADCP and improves the performance significantly. The simulations show that our proposed method can decrease the standard deviation of current velocity measurement by nearly three time, when compared to the conventional method.","PeriodicalId":441405,"journal":{"name":"2018 OCEANS - MTS/IEEE Kobe Techno-Oceans (OTO)","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131315548","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 : 2018-05-28DOI: 10.1109/OCEANSKOBE.2018.8559094
Juhwan Kim, Seokyong Song, Minsung Sung, Son-cheol Yu
We conducted efficient underwater manipulation that can lift heavy weight or perform simple tasks. We developed manipulation purposed small agent vehicle with gripper, flops and sensors. With the winch system, the lifting force of weight is empowered. The agent has independent navigation and object detection abilities, then only winding up winch signal is needed. This system is useful for use in areas where large underwater robots are difficult to access. We called this system as agent vehicle manipulation system. Finally, we accomplished the stand alone experiment of manipulation system in a water tank. The agent autonomously found such object, and lifted up to surface. The system verified the results of manipulation that can successfully bring the object.
{"title":"Development of Manipulation Purpose Small Agent Vehicle for UUVs","authors":"Juhwan Kim, Seokyong Song, Minsung Sung, Son-cheol Yu","doi":"10.1109/OCEANSKOBE.2018.8559094","DOIUrl":"https://doi.org/10.1109/OCEANSKOBE.2018.8559094","url":null,"abstract":"We conducted efficient underwater manipulation that can lift heavy weight or perform simple tasks. We developed manipulation purposed small agent vehicle with gripper, flops and sensors. With the winch system, the lifting force of weight is empowered. The agent has independent navigation and object detection abilities, then only winding up winch signal is needed. This system is useful for use in areas where large underwater robots are difficult to access. We called this system as agent vehicle manipulation system. Finally, we accomplished the stand alone experiment of manipulation system in a water tank. The agent autonomously found such object, and lifted up to surface. The system verified the results of manipulation that can successfully bring the object.","PeriodicalId":441405,"journal":{"name":"2018 OCEANS - MTS/IEEE Kobe Techno-Oceans (OTO)","volume":"225 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132293257","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：