基于模糊神经网络的水下机器人局部路径规划方法

该文探讨了基于模糊神经网络理论的实时局部路径规划问题,并提出了能实现模糊控制规则的基于强化学习的自学习和自调整的规划算法,设计了水下机器人实时运动规划器结构以及规划器操作过程和相应的算法。仿真实验结果验证了本文所提方法的有效性和可行性。     

基于改进人工势场法的水下滑翔机路径规划

研究了水下滑翔机在单个运动周期下躲避障碍物的路径规划问题,针对水下滑翔机运动特点,采用改进的人工势场法,规划出避障路径。首先,对传统的人工势场法进行改进,以克服局部极值与目标不可达问题,并引入速度势场函数,将静态势场转变为动态势场;然后,将水下滑翔机的运动特性及约束考虑进来,提出障碍物影响半径确定方法;之后,分析了定常海流对路径规划的影响。最后,以HUST-2号水下滑翔机为例在不同情况下进行仿真试验。结果表明,所用方法能使水下滑翔机成功避开水中静态与动态障碍物。     

基于RRT*算法的水下机器人全局路径规划方法

针对海洋环境的复杂性,考虑水下机器人能量的局限性,为减小洋流环境中作业全程水下机器人的能量消耗,以某水下机器人为研究对象,设计实现基于RRT*的路径最短和能耗最低的路径规划算法;并进行包括RRT*算法和RRT算法在复杂环境下的对比、不同洋流流速环境中水下机器人路径最短和能耗最低路径规划的仿真模拟。最后在水池中,利用实验室现有的水下机器人平台进行了真机实验。仿真测试和真机实验结果表明:所设计的基于RRT*的路径最短和能耗最低的路径规划算法可行有效。     

大型自主水下机器人全局静态与局部动态融合路径规划方法

针对大型自主水下机器人在做全局路径规划时面临环境建模复杂,算法求解能力弱以及面对局部动态障碍时自主性低,避障路径规划困难等问题,采用极坐标表示形成路径同心圆,在严格机动性约束下提出基于改进粒子群算法和速度障碍法的全局静态与局部动态相融合的路径规划方法。在极坐标表示的环境模型中,在全局静态规划中引入最优粒子"变异"过程提升算法求解能力;在局部动态规划中利用速度障碍法求解局部碰撞范围和安全路径区域以保证避障路径最优。实验结果表明,与传统粒子群和遗传算法相比,改进方法在全局静态规划中路径更短、求解能力更强,局部动态规划能够得到出最优避障路径。     

基于改进人工势场法的水下自主航行器路径规划

针对水下自主航行器(AUV)在水下复杂环境下的路径规划问题,提出改进人工势场法的路径规划算法.在传统人工势场法的基础上,提出考虑避障半径的斥力场函数方法,进一步优化规划路径.通过设置子目标点法和距离比较法,有效解决二维和三维空间下的局部最小值点问题和目标不可达问题.通过仿真对比验证提出的基于改进人工势场法的水下自主航行器路径规划算法的有效性.     

改进人工势场法的多移动机器人路径规划

针对未知环境下多移动机器人系统的路径规划问题,提出了一种新型的改进人工势场法.在吸引力、排斥力的基础上,按照统一规则对具有碰撞危险的多个机器人增加一旋转力,使它们同时右转(或左转),解决了传统势场法中多移动机器人多次避障甚至避障死锁的问题.仿真结果验证了方法的有效性.     

基于改进人工势场的避障路径规划策略研究

随着战争形势的无人化、多样化和跨域化发展,水下无人航行器作为各军事强国抢占水下作战域和海洋不对称作战优势的主要抓手,在未来战争中发挥着越来越重要的作用。由于海洋非结构化环境的复杂性,水下无人航行器在执行任务过程中,需能够在无人干预情况下进行系列操作和任务决策。本文针对水下无人航行器局部路径规划,提出基于速度矢量判断的改进人工势场法的避障航路规划策略,通过增加障碍物斥力场范围,强化目标点附近的引力场,优化障碍物斥力系数,并通过速度矢量判断旋转方向,使得水下无人航行器能够结合环境感知信息进行路径实时调整,最终能够达到安全快速避障。最后结合航行器流体动力与运动控制一体化仿真模型进行仿真分析,验证提出算法的有效性。     

基于改进蚁群算法的水下机器人路径规划算法

针对当前基本蚁群算法应用于水下机器人全局路径规划时存在路径搜索速度慢、容易陷入局部最优等问题,对其进行优化,提出一种改进蚁群算法。首先,改进算法引入A*算法作为新的初始路径搜索策略提高初始解的质量,加快算法收敛速度;针对特殊环境下算法容易陷入局部最优的问题做出优化,引入狼群分配策略进行蚂蚁回退。此外,对距离启发函数做出改进,综合考虑当前节点和下一节点以及下一节点和目标节点之间的距离,提高了算法搜索效率;提出一种信息素动态自适应更新策略,加快了算法前期搜寻效率,同时又扩大了算法后期搜寻范围。最后,以三次B样条法为基础引入路径平滑操作,去除规划路径结果中的冗余节点,减少了水下机器人移动过程中的能耗。仿真结果表明,和基本蚁群算法相比,改进算法不仅能取得更短、能耗更低的最优路径,收敛速度也更快。     

应用GA,PSO,ACO算法进行水下自主运载器路径规划(英文)

In this paper,an underwater vehicle was modeled with six dimensional nonlinear equations of motion,controlled by DC motors in all degrees of freedom.Near-optimal trajectories in an energetic environment for underwater vehicles were computed using a numerical solution of a nonlinear optimal control problem(NOCP).An energy performance index as a cost function,which should be minimized,was defined.The resulting problem was a two-point boundary value problem(TPBVP).A genetic algorithm(GA),particle swarm optimization(PSO),and ant colony optimization(ACO) algorithms were applied to solve the resulting TPBVP.Applying an Euler-Lagrange equation to the NOCP,a conjugate gradient penalty method was also adopted to solve the TPBVP.The problem of energetic environments,involving some energy sources,was discussed.Some near-optimal paths were found using a GA,PSO,and ACO algorithms.Finally,the problem of collision avoidance in an energetic environment was also taken into account.     

基于蚁群优化的AUV全局路径规划研究(英文)

路径规划是自主式水下潜器(AUV)导航研究的重要课题,AUV可用于未知环境如海洋空间探测.在大范围海洋环境中,应用蚁群优化原理对自主式水下潜器的全局路径规划问题进行了研究.引入栅格建模方法建立了蚁群可视图模型,设计了蚁群信息素更新规则;给出了蚁群全局路径规划的操作步骤;针对蚁群规划路径不平滑问题,设计了切割算予和插点算子.仿真实验结果表明,蚁群全局规划算法非常适合于求解复杂环境中的规划问题,规划时间短、路径平滑,其原型系统可应用于非结构化无人环境监测.     

Local path planning method of the self-propelled model based on reinforcement learning in complex conditions

Conducting hydrodynamic and physical motion simulation tests using a large-scale self-propelled model under actual wave conditions is an important means for researching environmental adaptability of ships. During the navigation test of the self-propelled model, the complex environment including various port facilities, navigation facilities, and the ships nearby must be considered carefully, because in this dense environment the impact of sea waves and winds on the model is particularly significant. In order to improve the security of the self-propelled model, this paper introduces the Q learning based on reinforcement learning combined with chaotic ideas for the model＇s collision avoidance, in order to improve the reliability of the local path planning. Simulation and sea test results show that this algorithm is a better solution for collision avoidance of the self navigation model under the interference of sea winds and waves with good adaptability.     

Coordinate control of initiative mating device for autonomous underwater vehicle based on TDES

A novel initiative mating device, which has four 2-degree manipulators around the mating skirt, is proposed to mate between a skirt of AUV （autonomons underwater vehicle） and a disabled submarine. The primary function of the device is to keep exact mating between skirt and disabled submarine in a badly sub sea environment. According to the characteristic of rescue, an automaton model is brought forward to describe the mating proceed between AUV and manipulators. The coordinated control is implemented by the TDES （time discrete event system）. After taking into account the time problem, it is a useful method to control mating by sinmlation testing. The result shows that it reduces about 70 seconds after using intelligent co-ordinate control based on TDES through the whole mating procedure.     

Review and prospects of underwater glider path planning;

The underwater glider (UG) is a new type of underwater vehicle driven by buoyancy, which has the advantages of low energy consumption, high efficiency, long-endurance, low cost, reusability and so on. The UG can also meet the needs of long-term and large-scale ocean observation and exploration. As an observation platform, the UG needs to carry out path planning and correction continuously in the early stages and during missions in order to better serve the requirements of ocean observation and exploration. First, this paper summarizes the relevant literatures on path planning research methods for UG in recent years. UG path planning algorithms are mainly divided into three categories: traditional algorithms, intelligent optimization algorithms and multi-algorithm fusions. Combined with practical application, the performance of different path planning algorithms is compared. The key technologies of UG path planning, such as environment reconstruction, environment perception, intelligent decision-making and underwater positioning, are then summarized. Finally, the development direction of UG multi-algorithm integration, multi-glider cooperation, multi-dimensional integration of spatiotemporal constraints and high-precision in complex and unsteady environments are prospected. © 2023 The Author(s).     

Design and evaluation of a hierarchical control architecture for an autonomous underwater vehicle

This paper researches on a kind of control architecture for autonomous undelwater vehicle （AUV）. After describing the hybrid property of the AUV control system, we present the hierarchical AUV control architecture. The architecture is organized in three layers： mission layer, task layer and execution layer. State supervisor and task coordinator are two key modules handling discrete events, so we describe these two modules in detail. Finally, we carried out a series of tests to verify this architecture The test results show that the AUV can perform autonomous missions effectively and safely. We can conclude the control architecture is valid and practical.     

一种混合优化的全局路径规划算法

针对全局路径规划研究中遗传算法存在搜索范围广而导致收敛速度慢的问题,本文提出一种混合优化的全局路径规划方法,完成对图像读取、处理后使用A*算法预处理缩小可行区域从而提高收敛速度。所提出的混合优化规划方法主要优化遗传算法的初始种群,在不影响最终路线的情况下,缩小初始种群的搜索范围,提高算法进行全局路径规划的速度,快速有效的规划出全局路线。另外本文给出一种评价体系对规划结果进行定量的避障评价,评价结果能够以数值形式对规划结果进行综合评价,评价结果显示通过混合优化算法规划出的路径具有更佳的安全性。     

水下航行体水动力参数智能辨识方法研究

通过水下航行体的状态方程和试验观测方程,利用智能辨识技术对水下航行体的模拟运动数据进行了仿真辨识,求得了10个水动力参数.结果表明,智能算法简单有效,对目标函数没有可微性和连续性要求,避免了复杂的梯度矩阵计算,适合在复杂的非线性水动力系统辨识中应用.     

一种基于拓展支持向量机的无人船路径规划方法

为有效解决无人船实现智能航行面临的路径规划问题,提出一种基于拓展支持向量机(SVM)的路径规划方法。该方法在传统SVM路径规划方法的基础上,考虑任务途中障碍物的规避风险,将工作任务的起止点作为约束条件对传统方法进行拓展,并针对不同的应用情景给出线性路径规划和非线性路径规划的求解方法。同时,为提高路径规划的效率,从SVM的原理出发,提出一种训练集裁剪的方法,大幅度减少利用SVM求解路径时所需的训练样本个数。此外,通过仿真验证该方法的可行性和有效性。     

Application of A* algorithm for real-time path re-planning of an unmanned surface vehicle avoiding underwater obstacles

This paper describes path re-planning techniques andunderwater obstacle avoidance for unmanned surface vehicle （USV）based on multi-beam forward looking sonar （FLS）. Near-optimalpaths in static and dynamic environments with underwaterobstacles are computed using a numerical solution procedure basedon an A algorithm. The USV is modeled with a circular shape in 2degrees of freedom （surge and yaw）. In this paper, two-dimensional（2-D） underwater obstacle avoidance and the robust real-time pathre-planning technique for actual USV using multi-beam FLS aredeveloped. Our real-time path re-planning algorithm has beentested to regenerate the optimal path for several updated frames inthe field of view of the sonar with a proper update frequency of theFLS. The performance of the proposed method was verifiedthrough simulations, and sea experiments. For simulations, theUSV model can avoid both a single stationary obstacle, multiplestationary obstacles and moving obstacles with the near-optimaltrajectory that are performed both in the vehicle and the worldreference frame. For sea experiments, the proposed method for anunderwater obstacle avoidance system is implemented with a USVtest platform. The actual USV is automatically controlled andsucceeded in its real-time avoidance against the stationary underseaobstacle in the field of view of the FLS together with the GlobalPositioning System （GPS） of the USV.