复杂环境下UUV完全遍历路径规划方法
|
摘要
针对复杂环境下无人水下航行器(UUV)完全遍历路径规划方法的不足,文中基于蚁群算法和生物激励神经网络,提出了一种高覆盖率、低重复率的完全遍历路径规划方法。该方法基于栅格法和生物激励神经网络进行环境建模,将神经元活性值引入蚂蚁转移概率公式中,既克服了蚁群算法需要对环境提前扫描学习,运算复杂的不足,又避免了生物激励神经网络随机性强,重复率高的缺陷。仿真试验表明,文中方法不仅有效实现了复杂环境下UUV完全遍历路径规划,而且能够以最短路线跳出死角,具有覆盖率大、重复率小,实用性强的优点。该研究可为进一步开展动态环境中的UUV完全遍历路径规划提供参考。
A new method with high coverage rate and low repetition rate is presented to solve the complete coverage path planning problem for an unmanned underwater vehicle(UUV) under complex environment. The method integrates ant colony algorithm and biologically inspired neural network. In this method, environment is modeled with grid cell and biologically inspired neural network, and the neuronic activity is introduced into the ant transition probability formula. As a result, the shortcomings of ant colony algorithm, e.g., scanning and learning environment in advance are needed and computation is complicated, are overcome, and the defects of high complexity, high randomness and high repetition rate of the biologically inspired neural network are avoided. Simulation experiment in complex environment shows that complete coverage path planning of UUV can be efficiently implemented by the proposed method, and the method can get rid of the dead corner in the shortest route. The present method has higher coverage rate and lower repetition rate. This research may provide a reference for further improvement of the UUV complete coverage path planning in dynamic environment.
A new method with high coverage rate and low repetition rate is presented to solve the complete coverage path planning problem for an unmanned underwater vehicle(UUV) under complex environment. The method integrates ant colony algorithm and biologically inspired neural network. In this method, environment is modeled with grid cell and biologically inspired neural network, and the neuronic activity is introduced into the ant transition probability formula. As a result, the shortcomings of ant colony algorithm, e.g., scanning and learning environment in advance are needed and computation is complicated, are overcome, and the defects of high complexity, high randomness and high repetition rate of the biologically inspired neural network are avoided. Simulation experiment in complex environment shows that complete coverage path planning of UUV can be efficiently implemented by the proposed method, and the method can get rid of the dead corner in the shortest route. The present method has higher coverage rate and lower repetition rate. This research may provide a reference for further improvement of the UUV complete coverage path planning in dynamic environment.
引文
[1]Choset H.Coverage for Robotics——A Survey of Recent Results[J].Annals of Mathematics&Artificial Intelligence,2001,31(1-4):113-126.
[2]De Carvalho R N,Vidal H A,Vieira P,et al.Complete Coverage Path Planning and Guidance for Cleaning Robots[C]//IEEE International Symposium on Industrial Electronics:Portugal,1997(2):677-682.
[3]Acar E U,Choset H.Robust Sensor——Based Coverage of Unstructured Environments[C]//Proceedings of the2001 IEEE/RSJ on Intelligent Robots and Systems.Maui:IEEE,2001:61-68.
[4]陈鹏,李彩虹.移动机器人混合式全遍历覆盖路径规划算法[J].山东理工大学学报:自然科学版,2013,27(5):22-27.Chen Peng,Li Cai-hong.A Hybrid Algorithm of Complete Coverage Path Planning for Mobile Robot[J].Journal of Shandong University of Technology(Natural Science Edition),2013,27(5):22-27.
[5]朱博,邓三鹏,王英飞,等.基于生物激励神经网络的移动机器人遍历路径规划[J].装备制造技术,2014(12):30-32.
[6]邱雪娜,刘士荣,宋加涛,等.不确定动态环境下移动机器人的完全遍历路径规划[J].机器人,2006,28(6):586-592.Qiu Xue-na,Liu Shi-rong,Song Jia-tao,et al.A Complete Coverage Path Planning Method for Mobile Robots in Uncertain Dynamic Environments[J].Robot,2006,28(6):586-592.
[7]段海滨,王道波,于秀芬.蚁群算法的研究现状及其展望[J].中国工程科学,2007,9(2):98-102.Duan Hai-bin,Wang Dao-bo,Yu Xiu-fen.Ant Colony Algorithm:Survey and Prospect[J].Engineering Science,2007,9(2):98-102.
[8]张赤斌,王兴松.基于蚁群算法的完全遍历路径规划研究[J].中国机械工程,2008,19(16):1945-1949.Zhang Chi-bin,Wang Xing-song.Complete Coverage Path Planning Based on Ant Colony Algorithm[J].China Mechanical Engineering,2008,19(16):1945-1949.
[9]Yang S X,Meng M.Neural Network Approaches to Dynamic Collision-free Trajectory Generation[J].IEEE Transactions on Systems Man&Cybernetics Part B Cybernetics A Publication of the IEEE Systems Man&Cybernetics Society,2001,31(3):302-318.
[10]Bernabucci I,Conforto S,Capozza M,et al.A Biologically Inspired Neural Network Controller for Ballistic Arm Movements[J].Journal of Neuro engineering&Rehabilitation,2007,4(1):1-17.
[2]De Carvalho R N,Vidal H A,Vieira P,et al.Complete Coverage Path Planning and Guidance for Cleaning Robots[C]//IEEE International Symposium on Industrial Electronics:Portugal,1997(2):677-682.
[3]Acar E U,Choset H.Robust Sensor——Based Coverage of Unstructured Environments[C]//Proceedings of the2001 IEEE/RSJ on Intelligent Robots and Systems.Maui:IEEE,2001:61-68.
[4]陈鹏,李彩虹.移动机器人混合式全遍历覆盖路径规划算法[J].山东理工大学学报:自然科学版,2013,27(5):22-27.Chen Peng,Li Cai-hong.A Hybrid Algorithm of Complete Coverage Path Planning for Mobile Robot[J].Journal of Shandong University of Technology(Natural Science Edition),2013,27(5):22-27.
[5]朱博,邓三鹏,王英飞,等.基于生物激励神经网络的移动机器人遍历路径规划[J].装备制造技术,2014(12):30-32.
[6]邱雪娜,刘士荣,宋加涛,等.不确定动态环境下移动机器人的完全遍历路径规划[J].机器人,2006,28(6):586-592.Qiu Xue-na,Liu Shi-rong,Song Jia-tao,et al.A Complete Coverage Path Planning Method for Mobile Robots in Uncertain Dynamic Environments[J].Robot,2006,28(6):586-592.
[7]段海滨,王道波,于秀芬.蚁群算法的研究现状及其展望[J].中国工程科学,2007,9(2):98-102.Duan Hai-bin,Wang Dao-bo,Yu Xiu-fen.Ant Colony Algorithm:Survey and Prospect[J].Engineering Science,2007,9(2):98-102.
[8]张赤斌,王兴松.基于蚁群算法的完全遍历路径规划研究[J].中国机械工程,2008,19(16):1945-1949.Zhang Chi-bin,Wang Xing-song.Complete Coverage Path Planning Based on Ant Colony Algorithm[J].China Mechanical Engineering,2008,19(16):1945-1949.
[9]Yang S X,Meng M.Neural Network Approaches to Dynamic Collision-free Trajectory Generation[J].IEEE Transactions on Systems Man&Cybernetics Part B Cybernetics A Publication of the IEEE Systems Man&Cybernetics Society,2001,31(3):302-318.
[10]Bernabucci I,Conforto S,Capozza M,et al.A Biologically Inspired Neural Network Controller for Ballistic Arm Movements[J].Journal of Neuro engineering&Rehabilitation,2007,4(1):1-17.