Minimizing position uncertainty for under-ice autonomous underwater vehicles

Localization underwater has been known to be challenging due to the limited accessibility of the Global Positioning System (GPS) to obtain absolute positions. This becomes more severe in the under-ice environment since the ocean surface is covered with ice, making it more difficult to access GPS or to deploy localization infrastructure. In this paper, a novel solution that minimizes localization uncertainty and communication overhead of under-ice Autonomous Underwater Vehicles (AUVs) is proposed. Existing underwater localization solutions generally rely on reference nodes at ocean surface or on localization infrastructure to calculate positions, and they are not able to estimate the localization uncertainty, which may lead to the increase of localization error. In contrast, using the notion of external uncertainty (i.e., the position uncertainty as seen by others), our solution can characterize an AUV’s position with a probability model. This model is further used to estimate the uncertainty associated with our proposed localization techniques. Based on this uncertainty estimate, we further propose algorithms to minimize localization uncertainty and communication overhead. Our solution is emulated and compared against existing solutions, showing improved performance.     

水下机器人的神经网络自适应控制

研究了水下机器人神经网络直接自适应控制方法,采用Lyapunov稳定性理论,证明了存在有界外界干扰和有界神经网络逼近误差条件下,水下机器人控制系统的跟踪误差一致稳定有界.为了进一步验证该水控制方法的正确性和稳定性,利用水下机器人实验平台进行了动力定位实验、单自由度跟踪实验和水平面跟踪实验等验证实验.     

Proximal object and hazard detection for autonomous underwater vehicle with optical fibre sensors

This paper describes short range and tactile optical fibre sensors for marine applications. The sensors are designed for obstacle avoidance on unmanned underwater vehicles (UUVs) operating in confined spaces, but have other possible applications. The fibre sensors augment the sensory abilities derived from ultrasonic and other sensors employed for marine proximity measurement. Of particular interest is proximity detection in the “near” (less than 1 m) and tactile areas. The paper describes the basic principle of operation and alternative sensor configurations. Results are given based on laboratory tests and deployment on a mini autonomous submersible in a test pool.     

一种水下机器人运动的过程神经元控制

过程神经网络是传统神经网络的拓展, 增加了一个对于时间的聚合算子, 从而更好地模拟了生物神经元的信息处理机制. 这是由于水下机器人运动控制系统的输入、输出均是随时间连续变化的过程量. 结合S函数和预先规划思想, 建立水下机器人过程神经元的运动控制模型. 仿真试验证明,该新型控制模型, 对于水下机器人的运动非线性控制器具有设计简单、响应速度快、超调小、鲁棒性好等优点.     

一种基于混沌过程神经元水下机器人运动控制方法

由于系统的强非线性以及不确定性,同时考虑到港湾环境下水声信号的噪声大,水下机器人进行精确作业时的运动控制一直是其实用化过程中困挠人们的问题。过程神经网络是传统神经网络的拓展,它增加了一个对于时间的聚合算子,使网络同时具有时空二维信息处理能力,从而更好地模拟了生物神经元的信息处理机制。水下机器人运动控制系统的输入、输出均是随时间连续变化的过程量。在基本神经元模型上,结合S函数和预先规划思想,建立水下机器人过程神经元运动控制模型,参数学习过程中,将遍历性的渐变混沌噪声引入其中,增强控制器全局优化能力。仿真试验表明,该新型控制模型,对于水下机器人的运动非线性控制器具有设计简单、响应速度快、超调小、鲁棒性好等各种优点。     

A time differences of arrival‐based homing strategy for autonomous underwater vehicles

A new sensor‐based homing integrated guidance and control law is presented to drive an underactuated autonomous underwater vehicle (AUV) toward a fixed target, in 3‐D, using the information provided by an ultra‐short baseline (USBL) positioning system. The guidance and control law is first derived at a kinematic level, expressed on the space of the time differences of arrival (TDOAs), as directly measured by the USBL sensor, and assuming the plane wave approximation. Afterwards, the control law is extended for the dynamics of an underactuated AUV resorting to backstepping techniques. The proposed Lyapunov‐based control law yields almost global asymptotic stability (AGAS) in the absence of external disturbances and is further extended, keeping the same properties, to the case where known ocean currents affect the motion of the vehicle. Simulations are presented and discussed that illustrate the performance and behavior of the overall closed‐loop system in the presence of realistic sensor measurements and actuator saturation. Copyright © 2009 John Wiley & Sons, Ltd.     

水下机器人协同控制的TO模型区域划分定位

针对稀疏型水声传感器网络定位算法面临的定位覆盖率低和误差高的问题, 本文提出一种水下机器人协 同控制的截角八面体(TO)模型区域划分定位算法. 首先搭建定位系统模型, 提出TO模型满足三维目标区域划分原 则, 并证明其体积比相对最优; 然后设计TO模型最优区域划分方式, 提出最小值判定法进一步整合目标节点, 自主 水下机器人(AUVs)协同控制筛选包含目标节点的子区域; 通过分析通信半径和虚拟锚节点数量对实验结果的影响, 设置最优定位参数, 降低能耗和定位误差, 最后利用最小二乘法完成定位. 本文分别对定位覆盖率、子区域AUV路 径长度和定位精度进行了仿真实验, 结果表明, 相比于其他区域划分方案, 所提算法误差较小、定位覆盖率高且鲁 棒性强.     

基于神经网络的欠驱动水下机器人三维同步跟踪和镇定控制

研究了欠驱动水下机器人的三维同步跟踪和镇定控制问题,并考虑了模型参数不确定性、未知外界干扰和输入饱和限制的影响.针对不同类型期望轨迹的特性,构造了新的辅助虚拟信号以实现对欠驱动方向的控制.基于反步法和Lyapunov直接法,设计了一种饱和自适应统一动力学控制律,使得AUV的状态误差最终收敛至零点附近的有界区域内,其中未...     

A two‐step control approach for docking of autonomous underwater vehicles

This paper presents a novel integrated guidance and control strategy for docking of autonomous underwater vehicles. The approach to the base, and hence the control design, is divided in two steps: (i) in the first, at higher speed, the vehicle dynamics is assumed to be underactuated, and an appropriate control law is derived to steer the vehicle towards the final docking path, achieving convergence to zero of the appropriate error variables for almost all initial conditions; (ii) in the second stage, at low speed, the vehicle is assumed to be fully actuated, and a robust control law is designed that achieves convergence to zero of the appropriate error variables for all initial conditions, in the presence of parametric model uncertainty. Simulations are presented illustrating the performance of the proposed controllers, including model uncertainty and sensor noise. Copyright © 2014 John Wiley & Sons, Ltd.     

The evolution of different neuronal control structures for autonomous agents

The use of evolutionary methods to generate controllers for real-world autonomous agents has attracted recent attention. Most of the pertinent research has employed genetic algorithms or variations thereof. Recent research has indicated that the presence of epistasis drastically slows down genetic algorithms. For this reason, this paper uses a different evolutionary method, evolution strategies, for the evolution of various (complex) neuronal control architectures for mobile robots inspired by Braitenberg vehicles. In these experiments, the evolution strategy accelerates the development process by more than an order of magnitude (a few hours compared to more than two days). Furthermore, the evolution strategy yields the same efficacy when applied to receptive-field controllers that require many more parameters than Braitenberg controllers. This dramatic speedup is very important, since the development process is to be done in real robots.     

垂直面欠驱动自治水下机器人定深问题的自适应输出反馈控制

针对垂直面欠驱动自治水下机器人(AUV)定深控制问题,本文仅使用可测量的深度和纵摇角信息,基于反步法设计自适应输出反馈控制器.为此首先设计观测器,实现不可测纵摇角速度反馈;再利用径向基神经网络对不确定水动力系数和纵荡、垂荡及纵摇角速度耦合产生的非线性结构进行补偿;采用自适应策略对纵荡和垂荡速度形成的有界干扰进行抑制.本文采用AUV一阶非完整模型,不以线性化为目的,放宽了纵摇角只能在小范围内变化的限制.最后通过理论证明和仿真实验表明该方法能够实现AUV深度和姿态控制,对未建模非线性动态和有界扰动具有很强的自适应性和鲁棒性.     

Seafloor map generation for autonomous underwater vehicle navigation

Elevation map generation is an essential component of any autonomous underwater vehicle designed to navigate close to the seafloor because elevation maps are used for obstacle avoidance, path planning and self localization. We present an algorithm for the reconstruction of elevation maps of the seafloor from side-scan sonar backscatter images and sparse bathymetric points co-registered within the image. Given the trajectory for the underwater vehicle, the reconstruction is corrected for the attitude of the side-scan sonar during the image generation process. To perform reconstruction, an arbitrary but computable scattering model is assumed for the seafloor backscatter. The algorithm uses the sparse bathymetric data to generate an initial estimate for the elevation map which is then iteratively refined to fit the backscatter image by minimizing a global error functional. Concurrently, the parameters of the scattering model are determined on a coarse grid in the image by fitting the assumed scattering model to the backscatter data. The reconstruction is corrected for the movement of the sensor by initially doing local reconstructions in sensor coordinates and then transforming the local reconstructions to a global coordinate system using vehicle attitude and performing the reconstruction again. We demonstrate the effectiveness of our algorithm on synthetic and real data sets. Our algorithm is shown to decrease the average elevation error when compared to real bathymetry from 4.6 meters for the initial surface estimate to 1.6 meters for the final surface estimate from a survey taken of the Juan de Fuca Ridge.     

A waypoint-tracking controller for a bionic autonomous underwater vehicle with two pectoral fins

This study aims to develop a waypoint-tracking control system for a biomimetic underwater vehicle (BUV). The BUV is propelled by wide paired pectoral foils, and each pectoral foil is driven by three independent fin rays. To simplify the control strategy, the maximum flapping amplitude of the pectoral fin is used to control the forward velocity, and a turning factor is defined for the manoeuvre control. Several swimming experiments are carried out to investigate the influence of the control parameters on the swimming performance of the prototype. Based on the results of the swimming experiments, a waypoint-tracking control system is proposed, which contains two layers: the velocity control layer and the heading angle control layer. A subdivision control method is adopted by the velocity control layer to get the maximum flapping amplitude. The fuzzy control method is employed by the heading angle control layer to obtain the turning factor for steering motion. Several waypoint-tracking experiments are carried out to verify effectiveness of the control system. The results show that the prototype can automatically reach the target area with the designed control system, even though the waypoints are arranged or randomly given.     

Vision-based autonomous tracking and landing of a fully-actuated rotorcraft

Autonomous landing is a challenging phase of flight for an aerial vehicle, especially when attempting to land on a moving target. This paper presents vision-based tracking and landing of a fully-actuated tilt-augmented quadrotor on a moving target. A fully-actuated vehicle allows higher freedom in terms of control design and a larger flight envelope since the position and attitude states are decoupled. An adaptive control law is designed to track a moving target with only relative position information from a camera. Low-cost hardware is used, and experiments are carried out to validate the proposed methodology for targets moving at realistic speeds.     

不确定海流环境下水下机器人最优时间路径规划

为解决海流预测不精确条件下,现有基于确定性海流路径规划算法鲁棒性差和规划的路径有可能为不可行路径的问题,本文提出一种基于区间优化的水下机器人(AUV)最优时间路径规划算法.该算法采用双层架构,外层用蚁群系统算法(ACS)寻找由起点至终点的候选路径;内层以区间海流为环境模型,计算候选路径航行时间上下限,并分别通过区间序关系和基于可靠性的区间可能度模型将航行时间区间转换为确定性评价函数,并将评价函数值作为候选路径适应度值返回到外层算法.仿真结果表明,相对于确定海流场路径规划方案,提出的方案增强了路径规划器的鲁棒性并解决了结果路径不可行问题.     

A data‐driven particle filter for terrain based navigation of sensor‐limited autonomous underwater vehicles

In this article a new Data‐Driven formulation of the Particle Filter framework is proposed. The new formulation is able to learn an approximate proposal distribution from previous data. By doing so, the need to explicitly model all the disturbances that might affect the system is relaxed. Such characteristics are particularly suited for Terrain Based Navigation for sensor‐limited AUVs, where typical scenarios often include non‐negligible sources of noise affecting the system, which are unknown and hard to model. Numerical results are presented that demonstrate the superior accuracy, robustness and efficiency of the proposed Data‐Driven approach.     

Robust control of variable speed autonomous underwater vehicle

Set point tracking control of autonomous underwater vehicle (AUV) via robust model predictive control (RMPC) is considered. Input-constrained RMPC with integral action, which has been developed in our previous work, is used to control the AUV in this study. In order to derive a RMPC control rule, non-linear dynamics of AUV with six degree of freedom is linearized at certain operating points. So, horizontal and vertical plane dynamics of system are represented by linear models which have polytopic uncertainties. Since the derived control rule will be used in real time, the computation time should be reduced. To overcome this computational time problem and get rid of trial–error step of Algorithm 1, a new algorithm is proposed here. The simulations are carried out using the control rule based on this algorithm and these results are presented.     

Underwater guidance of distributed autonomous underwater vehicles using one leader

Consider the case where autonomous underwater vehicles (AUVs) are deployed to monitor a 3D underwater environment. This paper tackles the problem of guiding all AUVs to the destination while not colliding with a priori unknown 3D obstacles. Suppose that among all AUVs, only the leader AUV has an ability of locating itself, while accessing a destination location. A follower, an AUV that is not a leader, has no sensors for locating itself. Every follower can only measure the relative position of its neighbor AUVs utilizing its sonar sensors. Our paper addresses distributed controls, so that multiple followers track the leader while preserving communication connectivity. We design controls, so that all AUVs reach the destination safely, while maintaining connectivity in cluttered 3D environments. To the best of our knowledge, our article is novel in developing 3D underwater guidance controls, so that all AUVs equipped with sonar sensors are guided to reach a destination in a priori unknown cluttered environments. MATLAB simulations are used to validate the proposed guidance scheme in underwater environments with many obstacles.     

Neural networks to determine task oriented dexterity indices for an underwater vehicle-manipulator system

A method for the fast approximation of dexterity indices for given underwater vehicle-manipulator systems (UVMS) configurations is presented. Common underwater tasks are associated with two well-known dexterity indices and two types of neural networks are designed and trained to approximate each one of them. The method avoids the lengthy calculation of the Jacobian, its determinant and the computationally expensive procedure of singular value decomposition required to compute the dexterity indices. It provides directly and in a considerably reduced computational time the selected dexterity index value for the given configuration of the system. The full kinematic model of the UVMS is considered and the NN training dataset is formulated by the conventional calculation of the selected dexterity indices. A comparison between the computational cost of the analytical calculation of the indices and their approximation by the two NN is presented for the validation of the proposed approach. This paper contributes mainly on broadening the applications of NN to a problem of high complexity and of high importance for UVMS high performance intervention.