摘要：自治水下机器人（AutonomousUnderwaterVehicle,AUV）作为一种高技术手段在海洋开发和利用领域的作用不容小觑，深海搜寻作为AUV的一项核心技术，有重要的工程意义。针对AUV在三维水下环境中的目标搜索问题，提出了一种基于生物启发神经网络的目标搜索算法，目的是为了提高目标搜索的效率。由于水下环境复杂、单个AUV受到动力能源、搜寻范围的限制，为了使AUV在水下能高效地搜寻到目标且能够有效避开障碍物，对水下搜寻策略的研究显得十分重要。首先，根据三维水下环境构建生物启发神经网络；其次，将环境中未搜索区域定义为神经网络的激励输入，全局吸引AUV；同时，将环境中的障碍物定义为抑制输入，局部排斥AUV；最后，AUV根据神经网络中神经元的活性值规划搜索路径。为了验证本文所提算法的有效性，分别对搜索过程中AUV是否发生故障的情况下目标搜索进行了仿真。仿真结果表明本文所提算法搜索效率高，搜索路径重复率低。91338

毕业论文关键词：生物启发神经网络，自治水下机器人，目标搜索，避障

Abstract：Autonomous Underwater Robot (Autonomous Underwater, Vehicle, AUV) is a kind of high technology in the field of marine development and utilization of role should not be underestimated,  deep-sea  search  as  a  core  technology  of  AUV,  has  important   engineering

significance。 Aiming at the problem of autonomous underwater vehicle in three dimensional underwater environment，a new algorithm based on biologically inspired neural network model

for autonomous underwater vehicle (AUV) is proposed。 The purpose is to improve  the efficiency of target search。 Due to the complexity of the underwater environment, a single AUV is the power source, search range restriction, in order to enable AUV to efficiently search to the target and can effectively avoid the obstacles in the water, is very important to study the underwater search strategy is。 Firstly, a biologically inspired neural network is constructed according to the three-dimensional underwater environment。 Secondly, the search region in the environment is defined as the input of the neural network, the global attractor AUV。 Finally, according to the activity value of neurons in the neural network, AUV plans  search  path。 In order to verify the effectiveness of the proposed algorithm, the target search is simulated in the case of AUV in the search process。 The simulation results show that the proposed algorithm has high search efficiency and low repetition rate。

Keywords: Biologically inspired neural network, Autonomous Underwater Vehicle, Target search, Obstacle avoidance

目 录

1绪论 3

1。1课题背景与意义 3

1。2自治水下机器人（AUV） 3

1。3目标搜索算法 3

2基于生物启发神经网络目标搜索算法 7

2。1生物启发神经网络原理 7

2。2三维生物启发神经网络模型 8

2。3三维神经网络的稳定性分析 10

2。4多AUV协作搜索路径选择方法 11

3。三维水下环境中多AUV协作目标搜索 12

3。1仿真参数设置 12

3。2三维水下环境中目标搜索仿真 12

3。2。1无故障情况下搜索多目标