基于DSmT的移动机器人地图构建及传感器管理

针对智能移动机器人探测未知环境的问题,引入了一种新的信息融合方法DSmT(Dezert-Smarandache Theo-ry),采用栅格地图,并根据声纳在DSmT框架下的数学模型,利用经典DSm模型构造了一组能自动调节误差范围的声纳基本信度赋值函数(gbbaf),以处理未知环境下声纳获取的不确定和不精确信息,甚至于高冲突信息。提出了简单有效的传感器管理方法,完全消除了复杂环境下声波的多次反射和串扰现象。最后,用Pioneer 2-DX机器人分别进行了DSmT和DST(Dempster-Shafer Theory)两种算法的地图构建实验,并绘制了相应的二维基本信度赋值地图。将DSmT与DST构建出的环境地图做比较,充分验证了DSmT及提出的传感器管理方法在未知环境下的有效性,为处理动态高冲突信息提供了有力的理论依据。     

自抗扰控制技术在高低温环境模拟系统中的应用

针对现有健康状态评估方法主观性较强,准确度不高等问题,提出一种基于DSmT理论和模糊综合评判的健康状态评估模型。首先,确定评估对象的评价指标体系,对采集的原始数据进行预处理;然后利用模糊综合评判理论确定广义基本信度赋值;接着采用DSmT融合规则对广义基本信度赋值合成,得到健康状态等级。对于多级的系统评估,可将融合后的结果作为新的广义基本信度赋值进行DSmT融合。实例验证表明,该状态评估模型能够准确有效地实现对系统的健康状态评估,而且能够克服高冲突证据的融合问题,具有良好的应用价值。     

基于经典DSmT 的Sonar栅格地图创建

基于最近提出的一种在贝叶斯和DST扩展而来的信息融合算法DSmT(Dezert-Smarandache Theory),在实验的基础上,结合Sonar测量的基本特性,对静态结构化环境建模,并构造了广义基本信度赋值函数,利用经典DSm融合规则,融合每个栅格的声纳冗余信息,计算栅格占用的Bel.最后,以Pioneer Ⅱ 移动机器人作为试验平台,并在线对小型环境进行了3D栅格占用信度分布地图创建,其俯视图与实际2D地图中的物体外观轮廓及所在位置进行比较,其比较结果充分验证了算法的有效性,为进一步研究应用基于折扣理论的DSmT解决异类或同类非可靠多源信息融合,基于Hybrid DSmT的动态环境地图创建,以及多机器人联合创建地图和自定位奠定了坚实的基础.     

基于混合DSm模型的多机器人地图构建

针对多移动机器人探测静态未知环境,提出了多机器人的一类新的体系结构和机器人内部控制结构,采用分布式融合框架,引入了最近提出的一种在贝叶斯和DST扩展而来的信息融合理论DSmT,结合限制传播算法,建立了不精确传感器(声纳)的混合DSm模型.构造了基本信度赋值函数,计算每个栅格的基本信度值(gbba),有效地融合了多个移动机器人使用声纳获取到的不精确、不确定和高冲突环境信息.最后,以Pioneer 2-DXe机器人作为实验平台,将由混合DSm模型构建出的静态环境地图与实际环境布局做比较,并利用openGL绘制出三维置信度分布图,充分验证了所提出的算法和基于通信的多机器人系统的有效性.     

基于混合DSm模型的移动机器人动态环境地图构建

针对移动机器人探测动态未知环境的问题,引入了一种由贝叶斯理论和Dempster-Shafer证据理论（DST）扩展而来的新的信息融合方法——Dezert-Smarandache理论（DSmT）．采用栅格地图,并根据声纳的物理特性,在DSmT框架下建立了声纳的数学模型．运用DSmT中的高级模型,即混合DSm模型,构造了一组基本信度赋值函数（gbbaf）,用以处理动态环境下声纳获取的不确定和不精确信息,甚至于高冲突信息．借助Pioneer2-Dxe移动机器人分别进行了混合DSm模型和DST两种算法的地图构建实验,并绘制了相应的二维基本信度赋值地图．将由混合DSm模型与DST构建出的环境地图进行了比较,充分验证了混合DSm模型在未知动态环境下的有效性,为处理动态高冲突信息提供了有力的理论依据．     

小样本条件下基于属性权重Shapley值分配的粗糙集决策模型

小样本条件下,根据粗糙集理论构建的决策规则受数据来源偶然性误差影响较大,个别数据样本难以反映真实知识关系.为解决小样本条件下粗糙集决策规则可信度未知的问题,提出信息区分量、属性影响方向等概念,运用Shapley值法进行属性权重分配,求取每个属性对决策结果的影响方向,进而得出决策规则的参考信度,以寻求真实可信且适合工程实际的决策规则.实例分析论证了所提方法的可行性以及对数据来源误差的分辨能力.     

一种新的GBPA生成方法及其在模式识别中的应用

广义证据理论是一种在不完备识别框架中处理多传感器信息融合问题的实用方法.由于时代环境的影响,人们的认知存在局限性,难免会将不完备的识别框架认为是完备的,经典证据理论在这种情况下并不完全适用.因此,根据广义证据理论提出一种新的广义基本概率赋值(generalized basic probability assignment,GBPA)生成方法.该方法首先根据训练数据分别构造样本类别和测试样本的广义三角模糊数模型;然后通过计算样本和类别间的广义三角模糊距离生成GBPA;最后使用广义组合规则融合所有证据并得出最终的结论.Iris数据集的实验结果表明所提方法合理有效,即使在样本不足的情况下仍有较高的分类精度.     

基于D-S信息融合的水下机器人地图构建

针对未知动态环境中自治水下机器人(Autonomous Underwater Vehicle,AUV)的路径规划问题,给出一种基 于D-S (Shafer-Dempster)信息融合的水下栅格地图构建算法.首先通过建立一个声纳传感器模型,将声纳数据转换成栅格的信度函数分配值;接着应用D-S证据理论信息融合算法更新地图数据,从而构建出水下动态栅格地图;最后通过真实地图与融合构建地图比较,说明D-S融合算法在地图构建中的可行性.     

D-S证据理论在目标识别中的应用

D-S证据理论的应用过程与事件密切相关,针对目标识别这一特定领城,研究了证据理论中基本概率赋值的获取、组合规则及决策规则等问题.识别实例中,利用灰关联分析法来处理基本概率赋值的问题,对多传感器基本概率赋值进行组合后,最终通过决策完成对2种车型的类型识别,经计算仿真表明,D-S证据理论在目标识别中具有一定的有效性和优越性.     

DSmT框架下的自适应通用分配法则

针对Dempster-Shafer证据理论（DST）及Dezert-Smarandache证据理论（DSmT）均无法处理不确定信息的问题,定义了辨识框架中的不确定因子,通过深入分析比较DSmT框架下的各个冲突分配法则（PCR）,提出了一种基于PCR2的自适应通用分配法则（AUPR）,并根据声纳的数学模型构造了一组新的声纳信度赋值函数（gbbaf）,用以描述声纳获取的不确定和不精确信息,甚至于高冲突信息。最后,以Pioneer 2-DXe机器人为实验平台,绘制了实验场景的各种信度分布图。实验结果充分验证了所提方法的有效性和实用性,为信息融合理论中如何处理不确定信息提供了有力的理论依据。     

基于决策树的羽流追踪机器人自主决策方法研究

针对无法获得可靠羽流流向信息不利于实现羽流追踪的问题，提出了一种基于决策树的羽流追踪移动机器人自主决策方法。该方法通过移动机器人两侧的浓度传感器采集到的浓度信息，利用追踪的行为规则建立决策树模型，获得行为决策信息，使机器人高效地追踪到羽流并精确地定位。由于浓度变化关系蕴含了羽流的流向及流速信息，从而取代了传统方法中流向及流速传感器。在扩散环境下，通过移动机器人羽流追踪实验，实现了良好的源定位效果。     

Human–robot communication for collaborative decision making — A probabilistic approach

Humans and robots need to exchange information if the objective is to achieve a task collaboratively. Two questions are considered in this paper: what and when to communicate. To answer these questions, we developed a human–robot communication framework which makes use of common probabilistic robotics representations. The data stored in the representation determines what to communicate, and probabilistic inference mechanisms determine when to communicate. One application domain of the framework is collaborative human–robot decision making: robots use decision theory to select actions based on perceptual information gathered from their sensors and human operators. In this paper, operators are regarded as remotely located, valuable information sources which need to be managed carefully. Robots decide when to query operators using Value-Of-Information theory, i.e. humans are only queried if the expected benefit of their observation exceeds the cost of obtaining it. This can be seen as a mechanism for adjustable autonomy whereby adjustments are triggered at run-time based on the uncertainty in the robots’ beliefs related to their task. This semi-autonomous system is demonstrated using a navigation task and evaluated by a user study. Participants navigated a robot in simulation using the proposed system and via classical teleoperation. Results show that our system has a number of advantages over teleoperation with respect to performance, operator workload, usability, and the users’ perception of the robot. We also show that despite these advantages, teleoperation may still be a preferable driving mode depending on the mission priorities.     

Remote control of space robots

A method for the remote control of a space robot is proposed for the case of large delays in the transmission of control signals from the Earth to the local robot control system and in feedback signals. The method involves the use of the model of the space robot and its current environment with the simulation of gravity conditions at the ground control center. In this model environment, the operator should carry out the required actions by controlling the space robot model in the master-slave mode using an arm with six degrees of freedom capable of reflecting the interaction force of a model robot working tool with models of the objects of the environment. The arm movement trajectory and the law of time variation of the reflected interaction force vector are program-based for the local space robot control system and should be executed by it upon reception from the ground control center. The robot’s possible erroneous actions generated by the inevitable inaccuracy of the environment model are compensated by the proposed method of programmed trajectory correction. In accordance with it, in order to generate correction signals, additional information received from different sensors is used. These sensors can be installed on both the model and space robot itself. This information includes data on the mutual position of a robot’s working tool and models of the objects of the environment, as well as on the interaction forces between them. The paper presents a detailed theoretical justification of the proposed approach and experimental results that confirm the theoretical conclusions.     

传感器网络中利用反演集合估计的机器人定位方法

针对大型协作环境中移动机器人的全局定位问题,提出根据机器人车载传感器、环境传感器以及其他机器人的实时数据估计移动机器人的位置。首先,提出的方法整合大量不同类型传感器,从最简单传感器到最复杂传感器;然后,考虑了测量值数量可变、通用测角测量、受容错约束的测量统计知识等约束条件,将非线性边界误差估计问题看作一种反演集合。最后处理特定类型的异常值和不精确环境下的模型误差。完成了误差和异常值的处理,就基本上获得了定位图,解决了移动机器人的定位问题。提出的方法利用实物实验进行验证。测试区域装备有多个传感器、固定在墙顶部的摄像机以及位于机器人上的可见标记。实验结果表明提出的方法在协作环境中具有明显优势,处理异常值更加可靠。     

多传感器信息融合在移动机器人定位中的应用

机器人自定位是实现自主导航的关键问题之一。为了满足机器人在导航时精确定位的要求,提出一种基于多传感器信息融合的自定位算法。根据对机器人运动机构的分析和运动机构间的刚体约束,建立起机器人的运动学模型;由传感器的工作原理建立里程计和超声波传感器的观测模型;利用扩展卡尔曼滤波(EKF)算法将里程计和超声波传感器采集的数据进行融合;最后,由匹配的环境特征对机器人的位置进行修正,得到精确的位置估计。实验结果表明:该算法明显地消除了里程计的累计误差,有效地提高了定位精度。     

An integrated MCDM method for robot selection under interval-valued Pythagorean uncertain linguistic environment

Robots have received considerable attention in many manufacturing companies due to their great capabilities and characteristics. Selecting an appropriate robot for a specific application can be regarded as a challenging multicriteria decision-making problem. Furthermore, decision makers are inclined to represent their opinions by using linguistic terms owing to their ambiguous thinking. In this regard, we put forward a novel robot selection model by integrating quality function development (QFD) theory and qualitative flexible multiple criteria method (QUALIFLEX) under interval-valued Pythagorean uncertain linguistic context. For the developed model, the evaluations given by decision makers are presented as interval-valued Pythagorean uncertain linguistic sets for dealing with the uncertainty and vagueness of decision makers’ information. An extended QFD method is used for determining criteria weights from the perspective of customers. A modified QUALIFLEX technique based on closeness degree is utilized to generate the ranking order of alternative robots and determine the most suitable one. Finally, an empirical example of an auto manufacturing company is applied to clarify the effectiveness and accuracy of the proposed robot selection approach.     

Simultaneous localization and map building for mobile robotnavigation

Inexpensive ultrasonic sensors, incremental encoders, and grid-based probabilistic modeling are used for improved robot navigation in indoor environments. For model-building, range data from ultrasonic sensors are constantly sampled and a map is built and updated immediately while the robot is travelling through the workspace. The local world model is based on the concept of an occupancy grid. The world model extracted from the range data is based on the geometric primitive of line segments. For the extraction of these features, methods such as the Hough transform and clustering are utilized. The perceived local world model along with dead-reckoning and ultrasonic sensor data are combined using an extended Kalman filter in a localization scheme to estimate the current position and orientation of the mobile robot, which is subsequently fed to the map-building algorithm. Implementation issues and experimental results with the Nomad 150 mobile robot in a real-world indoor environment (office space) are presented     

动态多行人环境中基于最优交互避碰的机器人导航

在动态的多行人环境中,服务机器人仅依赖于自身传感器、以第一人称视角自主导航时. 机器人自主定位的不确定性以及对周围行人运动状态估计的不确定性均增加,这给机器人导航决策带来了困难. 为解决这个问题,提出一种基于最优交互避碰的机器人自主导航法. 本方法采用一种改进的粒子PHD滤波法即NP-PHDF法跟踪多个行人的状态. NP-PHDF法结合了卡尔曼粒子滤波及PHD滤波优点,因此它可以跟踪数目变化的多个目标,能够跟踪突然的加减速以及急转弯运动,并且能够抵抗遮挡. 同时,与基于粒子滤波的机器人自主定位法类似,NP-PHDF法使得行人运动状态的不确定性能够以粒子的分布来度量. 为降低状态估计的不确定性,本文提出一种“圈粒子”的粒子圈存法从粒子的分布中提取机器人和行人的真实状态. 算法的有效性在实际场景实验中得到了验证.     

A learning fuzzy algorithm for motion planning of mobile robots

A sensor-based fuzzy algorithm is proposed to navigate a mobile robot in a 2-dimensional unknown environment filled with stationary polygonal obstacles. When the robot is at the starting point, vertices of the obstacles that are visible from the robot are scanned by the sensors and the one with the highest priority is chosen. Here, priority is an output fuzzy variable whose value is determined by fuzzy rules. The robot is then navigated from the starting point to the chosen vertex along the line segment connecting these two points. Taking the chosen vertex as the new starting point, the next navigation decision is made. The navigation process will be repeated until the goal point is reached.In implementation of fuzzy rules, the ranges of fuzzy variables are parameters to be determined. In order to evaluate the effect of different range parameters on the navigation algorithm, the total traveling distance of the robot is defined as the performance index first. Then a learning mechanism, which is similar to the simulated annealing method in the neural network theory, is presented to find the optimal range parameters which minimize the performance index. Several simulation examples are included for illustration.     

Mobile robot relocation from echolocation constraints

This paper presents a method for relocation of a mobile robot using sonar data. The process of determining the pose of a mobile robot with respect to a global reference frame in situations where no a priori estimate of the robot's location is available is cast as a problem of searching for correspondences between measurements and an a priori map of the environment. A physically-based sonar sensor model is used to characterize the geometric constraints provided by echolocation measurements of different types of objects. Individual range returns are used as data features in a constraint-based search to determine the robot's position. A hypothesize and test technique is employed in which positions of the robot are calculated from all possible combinations of two range returns that satisfy the measurement model. The algorithm determines the positions which provide the best match between the range returns and the environment model. The performance of the approach is demonstrated using data from both a single scanning Polaroid sonar and from a ring of Polaroid sonar sensors