面向空地协同应急的地表可通行性分析方法
|
摘要
在野外灾害应急与救援过程中,受灾害或复杂地表环境的影响,快速搜索一条到达目的地的可通行性路径是一项具有挑战性的任务。近些年随着空地协同技术的不断发展,利用无人机为地面车载系统提供其周边环境的数据并搜索可通行路径成为空地协同技术研发的一个热点。本文利用机器学习方法,结合基于无人机获取的野外可见光影像自动生成的数字表面模型(DSM)对该无人机可见光影像进行了详细分类,得到了较传统方法更好的分类结果;然后基于一般性规律,建立了地表可通行性判断条件与通行速度计算公式;再通过引入分类结果以及分类的不确定性因子,对A*算法进行了改进。基于改进的A*算法实现了实验区可通行性路径的快速搜索以及路径的可靠性评价,实验表明本文方法可得到更合理的结果。
In the process of disaster emergency and rescue, due to the impact of disaster and the complexity of the ground environment, it is a challenging task to plan a traversable route to the target place quickly. In recent years, with the rapid development of technologies in Air-ground Collaboration, using UAV to collect environmental data and plan a traversable route for the ground vehicle system has become a hot issue. In this research work, firstly, the visible images acquired by the UAV and the DSM automatically calculated from the visible images are used in Deep learning method to train a terrain surface classification model, a better result is extracted than the traditional remote sensing image classification methods; secondly, a judgement of trafficability for terrain surface based on general knowledge is established, and a formula for calculation of moving speed on different terrain surface is presented; thirdly, the classification result and its uncertainty are introduced in A* algorithm to polish up it, then the improved A* algorithm is used to accomplish a fast exploration of the traversable route in the experimental area, while the reliability evaluation of the route is also given as a reference for decision making in rescue process. Experiments show that our method can get more reasonable result.
In the process of disaster emergency and rescue, due to the impact of disaster and the complexity of the ground environment, it is a challenging task to plan a traversable route to the target place quickly. In recent years, with the rapid development of technologies in Air-ground Collaboration, using UAV to collect environmental data and plan a traversable route for the ground vehicle system has become a hot issue. In this research work, firstly, the visible images acquired by the UAV and the DSM automatically calculated from the visible images are used in Deep learning method to train a terrain surface classification model, a better result is extracted than the traditional remote sensing image classification methods; secondly, a judgement of trafficability for terrain surface based on general knowledge is established, and a formula for calculation of moving speed on different terrain surface is presented; thirdly, the classification result and its uncertainty are introduced in A* algorithm to polish up it, then the improved A* algorithm is used to accomplish a fast exploration of the traversable route in the experimental area, while the reliability evaluation of the route is also given as a reference for decision making in rescue process. Experiments show that our method can get more reasonable result.
引文
[1]Stentz T,Kelly A,Herman H,et al.Integrated air/ground vehicle system for semi-autonomous off-road navigation[J].Robotics Institute,2002(1):18.
[2]Stentz A,Kelly A,Rander P,et al.Realtime,multi-perspective perception for unmanned ground vehicles[C].AUVSI,2003:16.
[3]Kelly A,Stentz A,Amidi O,et al.Toward reliable off road autonomous vehicles operating in challenging environments[J].The International Journal of Robotics Research,2006,25(5-6):449-483.
[4]Grocholsky B,Bayraktar S,Kumar V,et al.Uav and ugv collaboration for active ground feature search and localization[C]//Proc.of the AIAA 3rd”Unmanned Unlimited”Technical Conference,2004:6565.
[5]Grocholsky B,Keller J,Kumar V,et al.Cooperative air and ground surveillance[J].IEEE Robotics&Automation Magazine,2006,13(3):16-25.
[6]Pippin C,Gray G,Matthews M,et al.The design of an airground research platform for cooperative surveillance[R].Tech.Rep.112010,Georgia Tech Research Institute(November 2010),2010.
[7]Cantelli L,Mangiameli M,Melita C D,et al.UAV/UGV cooperation for surveying operations in humanitarian demining[C]//Safety,Security,and Rescue Robotics(SSRR),2013 IEEEInternational symposium on.IEEE,2013:1-6.
[8]Jin Hyo Kim,Ji Wook Kwon,Jiwon Seo.Multi-UAV-based stereo vision system without GPS for ground obstacle mapping to assist path planning of UGV[J].Electronics Letters,2014,50(20):1431-1432
[9]Elias Mueggler,Matthias Faessler,Flavio Fontana,et al.Aerialguided navigation of a ground robot among movable obstacles[C].IEEE International Symposium on Safety,Security,and Rescue Robotics(SSRR),2014:1-8.
[10]Li J,Deng G,Luo C,et al.A hybrid path planning method in unmanned Air/Ground vehicle(UAV/UGV)cooperative systems[J].IEEE Transactions on Vehicular Technology,2016,65(12):9585-9596.
[11]Delmerico J,Mueggle E,Nitsch J,et al.Active autonomous aerial exploration for ground robot path planning[J].IEEERobotics and Automation Letters,2017,2(2):664-671.
[12]Wang L,Cheng D,Gao F,et al.A collaborative aerial-ground robotic system for fast exploration[J].Arxiv:Robotics,2018,(3):15-27.
[13]Vaduva C,Gavat I,Datcu M.Deep learning in very high resolution remote sensing image information mining communication concept[C]//Signal Processing Conference(EUSIPCO),2012 Proceedings of the 20th European.IEEE,2012:2506-2510.
[14]Chen Y,Lin Z,Zhao X,et al.Deep learning-based classification of hyperspectral data[J].IEEE Journal of Selected topics in applied earth observations and remote sensing,2014,7(6):2094-2107.
[15]Chen Y,Jiang H,Li C,et al.Deep feature extraction and classification of hyperspectral images based on convolutional neural networks[J].IEEE Transactions on Geoscience and Remote Sensing,2016,54(10):6232-6251.
[16]Wang L,Zhang J,Liu P,et al.Spectral-spatial multi-featurebased deep learning for hyperspectral remote sensing image classification[J].Soft Computing,2017,21(1):213-221.
[17]Hu W,Huang Y,Wei L,et al.Deep convolutional neural networks for hyperspectral image classification[J].Journal of Sensors,2015,(4):258619.
[18]Makantasis K,Karantzalos K,Doulamis A,et al.Deep supervised learning for hyperspectral data classification through convolutional neural networks[C]//Geoscience and Remote Sensing Symposium(IGARSS),2015 IEEE International.IEEE,2015:4959-4962.
[19]吕启,窦勇,牛新,等.基于DBN模型的遥感图像分类[J].计算机研究与发展,2014,51(9):1911-1918.LüQi,Dou Yong,Niu Xin,et al.Remote sensing image classification based on DBN model[J]Journal of Computer Research and Development,2014,51(9):1911-1918.
[20]刘大伟,韩玲,韩晓勇.基于深度学习的高分辨率遥感影像分类研究[J].光学学报,2016(4):298-306.Liu Dawei,Han Ling,Han Xiaoyong.High spatial resolution remote sensing image classification based on deep learning[J].Acta Optica Sinica,2016(4):298-306.
[2]Stentz A,Kelly A,Rander P,et al.Realtime,multi-perspective perception for unmanned ground vehicles[C].AUVSI,2003:16.
[3]Kelly A,Stentz A,Amidi O,et al.Toward reliable off road autonomous vehicles operating in challenging environments[J].The International Journal of Robotics Research,2006,25(5-6):449-483.
[4]Grocholsky B,Bayraktar S,Kumar V,et al.Uav and ugv collaboration for active ground feature search and localization[C]//Proc.of the AIAA 3rd”Unmanned Unlimited”Technical Conference,2004:6565.
[5]Grocholsky B,Keller J,Kumar V,et al.Cooperative air and ground surveillance[J].IEEE Robotics&Automation Magazine,2006,13(3):16-25.
[6]Pippin C,Gray G,Matthews M,et al.The design of an airground research platform for cooperative surveillance[R].Tech.Rep.112010,Georgia Tech Research Institute(November 2010),2010.
[7]Cantelli L,Mangiameli M,Melita C D,et al.UAV/UGV cooperation for surveying operations in humanitarian demining[C]//Safety,Security,and Rescue Robotics(SSRR),2013 IEEEInternational symposium on.IEEE,2013:1-6.
[8]Jin Hyo Kim,Ji Wook Kwon,Jiwon Seo.Multi-UAV-based stereo vision system without GPS for ground obstacle mapping to assist path planning of UGV[J].Electronics Letters,2014,50(20):1431-1432
[9]Elias Mueggler,Matthias Faessler,Flavio Fontana,et al.Aerialguided navigation of a ground robot among movable obstacles[C].IEEE International Symposium on Safety,Security,and Rescue Robotics(SSRR),2014:1-8.
[10]Li J,Deng G,Luo C,et al.A hybrid path planning method in unmanned Air/Ground vehicle(UAV/UGV)cooperative systems[J].IEEE Transactions on Vehicular Technology,2016,65(12):9585-9596.
[11]Delmerico J,Mueggle E,Nitsch J,et al.Active autonomous aerial exploration for ground robot path planning[J].IEEERobotics and Automation Letters,2017,2(2):664-671.
[12]Wang L,Cheng D,Gao F,et al.A collaborative aerial-ground robotic system for fast exploration[J].Arxiv:Robotics,2018,(3):15-27.
[13]Vaduva C,Gavat I,Datcu M.Deep learning in very high resolution remote sensing image information mining communication concept[C]//Signal Processing Conference(EUSIPCO),2012 Proceedings of the 20th European.IEEE,2012:2506-2510.
[14]Chen Y,Lin Z,Zhao X,et al.Deep learning-based classification of hyperspectral data[J].IEEE Journal of Selected topics in applied earth observations and remote sensing,2014,7(6):2094-2107.
[15]Chen Y,Jiang H,Li C,et al.Deep feature extraction and classification of hyperspectral images based on convolutional neural networks[J].IEEE Transactions on Geoscience and Remote Sensing,2016,54(10):6232-6251.
[16]Wang L,Zhang J,Liu P,et al.Spectral-spatial multi-featurebased deep learning for hyperspectral remote sensing image classification[J].Soft Computing,2017,21(1):213-221.
[17]Hu W,Huang Y,Wei L,et al.Deep convolutional neural networks for hyperspectral image classification[J].Journal of Sensors,2015,(4):258619.
[18]Makantasis K,Karantzalos K,Doulamis A,et al.Deep supervised learning for hyperspectral data classification through convolutional neural networks[C]//Geoscience and Remote Sensing Symposium(IGARSS),2015 IEEE International.IEEE,2015:4959-4962.
[19]吕启,窦勇,牛新,等.基于DBN模型的遥感图像分类[J].计算机研究与发展,2014,51(9):1911-1918.LüQi,Dou Yong,Niu Xin,et al.Remote sensing image classification based on DBN model[J]Journal of Computer Research and Development,2014,51(9):1911-1918.
[20]刘大伟,韩玲,韩晓勇.基于深度学习的高分辨率遥感影像分类研究[J].光学学报,2016(4):298-306.Liu Dawei,Han Ling,Han Xiaoyong.High spatial resolution remote sensing image classification based on deep learning[J].Acta Optica Sinica,2016(4):298-306.