动态环境下基于概率模型检测的路径规划方法

提出了一种动态复杂环境下采用概率模型检测技术进行路径规划的新方法。考虑到实际应用中机器人其移动行为总是受到外界因素的影响,将机器人移动行为看作一个不确定事件,提取环境中的影响因素,构建马尔可夫决策过程模型。采用时态逻辑语言描述机器人目标任务,表达复杂多样的需求行为。运用工具PRISM验证属性,得到满足任务需求的全局优化路径。另外,在全局路径的基础上提出了一种动态避障策略,实现避障局部规划的同时尽量保证机器人最大概率完成任务。通过理论和仿真实验结果证明该方法的正确性和有效性。     

一种动态环境下移动机器人的路径规划方法

本文提出了在动态环境中，移动机器人的一种路径规划方法，适用于环境中存 在已知和未知、静止和运动障碍物的复杂情况．采用链接图法建立了机器人工作空间模型， 整个系统由全局路径规划器和局部路径规划器两部分组成．在全局路径规划器中，应用遗传 算法规划出初步全局优化路径．在局部路径规划器中，设计了三种基本行为：跟踪全局路径 的行为、避碰的行为和目标制导的行为，采用基于行为的方法进一步优化路径．其中，避碰 的行为是通过强化学习得到的．仿真和实验结果表明所提方法简便可行，能够满足移动 机器人导航的高实时性要求．     

口腔机器人手术中移动路径控制方法设计

口腔手术机器人是口腔医学的智能体工具，为了提高口腔机器人手术中移动路径自动控制水平，提出基于视觉感知和运动学模型规划的口腔机器人手术中移动路径控制方法。构建口腔机器人手术中移动的运动学规划模型，采用视觉伺服控制的方法提取口腔机器人手术中移动过程中的视觉特征参数，采用神经网络学习方法实现对口腔机器人手术中移动过程中的路径空间参数捕获和避障参数拟合，采用路径边缘引导的方法，结合目标点测距和视觉特征点匹配方法，实现对口腔机器人手术中移动路径的自适应控制和空间移动规划设计。测试结果表明，采用该方法口腔机器人手术中移动路径控制的环境适应性较强，控制鲁棒性较高，提高了口腔机器人手术过程中的精准性。     

基于改进RRT算法的RoboCup机器人动态路径规划

针对足球机器人在场上采用反应式方法避障时存在的速度慢、效果差的问题,采用改进的快速扩展随机树(RRT)算法设计了一种能够适应机器人足球赛场动态移动障碍环境的路径规划器.首先,引入基本的快速扩展随机树算法,针对其随机性强、路径过长的缺点,提出了以一定概率选择目标点、增加引力分量以及路径平滑处理等改进方式;引入路径缓存区以及动态扩展随机树的方法来解决移动障碍物环境中的路径规划问题.复杂障碍物环境中的仿真实验表明,改进的规划路径长度比基本快速扩展随机树算法所得路径缩短约20%.最终将策略移植到实体NAO机器人上参加RoboCup比赛,取得世界八强的成绩.     

改进蚁群算法在迷宫路径寻优中的应用研究

针对机器人在障碍环境下寻找最优路径的问题,提出了一种动态环境下的机器人路径规划的仿生算法.该算法采用栅格法对场地建模,并模拟蚂蚁的觅食行为,由多只蚂蚁协作完成最优路径的搜索.搜索过程采用了概率搜索策略和自适应调整信息素的方法,使得搜索策略更有效.仿真实验结果表明,在场地复杂的情况下,该算法可以有效地规划出最优路径.     

复杂环境下基于贝叶斯决策的机器人路径规划

提出了一种基于贝叶斯决策的机器人路径规划蚂蚁算法,该算法在路径节点选择方式上采用贝叶斯模型,通过后验概率对候选节点进行评估,解决了用传统蚂蚁算法进行路径规划时容易陷入局部最优的问题。仿真实验表明,机器人应用该算法可在复杂障碍环境下快速规划出一条全局优化避障路径。     

基于改进DQN算法的机器人路径规划

传统深度Q网络(DQN)算法通过融合深度神经网络和强化学习方法,解决了Q-learning算法在应对复杂环境时出现的维数灾难问题,被广泛应用于移动机器人的路径规划,但传统DQN算法的网络收敛速度较慢,路径规划效果较差,难以在较少的训练回合内获取最优路径。为了解决上述问题,提出一种改进的ERDQN算法。通过记录重复状态出现的频率,利用该频率重新计算Q值,使得在网络训练的过程中一种状态重复出现的次数越多,下一次出现该状态的概率越低,从而提高机器人对环境的探索能力,在一定程度上降低了网络收敛于局部最优的风险,减少了网络收敛的训练回合。根据机器人移动方向和机器人与目标点的距离,重新设计奖励函数。机器人在靠近目标点时能够获得正奖励,远离目标点时能够获得负奖励,并通过当前机器人的移动方向和机器人与目标点的距离调整奖励的绝对值,从而使机器人能够在避开障碍物的前提下规划出更优路径。实验结果表明,与DQN算法相比,ERDQN算法的平均得分提高了18.9%,规划出的路径长度和回合数减少了约20.1%和500。上述结果证明了ERDQN算法能够有效提高网络收敛速度及路径规划性能。     

基于A*改进算法的机器人移动路径优化仿真

针对传统机器人移动路径分析方法中存在撞击率高的问题,引入A*改进算法对机器人移动路径进行优化.首先利用栅格单位搭建机器人移动环境模型,并对栅格模型进行优化处理,分别在无障碍、静态障碍以及动态障碍三种移动环境下进行路径规划,通过A*改进算法获取机器人移动原始路径.在此基础上,计算路径优化约束条件,对原始路径进行平滑优化处理,输出机器人移动路径的优化结果,完成移动路径优化.经过仿真对比实验可知,使用A*改进算法进行机器人移动路径优化,路径长度和时间均能得到减少,且撞击率和路径转弯次数也明显得到改善.     

改进萤火虫算法求解多机器人路径规划

为解决多机器人在静态环境中的路径规划问题,以路径长度为优化目标模型,并针对此模型设计了多机器人萤火虫算法(MR-FA)。首先,考虑到路径安全性对环境中的障碍物采取扩张操作,设计初始化规则以提高生成初始种群的效率;其次,根据算法的连续性原理及特点,设计个体等长策略将维度不一致的个体转变为等维度个体以便于萤火虫的移动更新,并对移动更新后的不可行解采取路径修正策略;然后对规划出的每个机器人的移动路径进行碰撞检测,同时针对机器人不同的碰撞情况设计相应的避碰策略,即暂停—回退策略(PFS)、局部路径重规划策略(LPRS);最后,为验证MR-FA的有效性,在三组环境中进行仿真实验并与其他三种算法进行对比,综合得出MR-FA在解决多机器人路径规划时更有优势。     

移动机器人路径规划的仿真研究

研究移动机器人路径优化问题,由于移动机器人寻优中存在定位稳定性和避障准确性问题,机器人路径规划不仅找到一条无碰撞、安全的移动机器人路径,而且要求路径尽可能最短。传统单一栅格法和遗传算法搜索最优路径效率低,难以全局最优路径。为了获得机器人全局最优路径,提出一种栅格法和混沌遗传算法相融合的移动机器人路径规划方法。首先采用栅格法对移动路径进行规划,作为遗传算法的初始种群,采用遗传算法进一步寻找最优路径。最后对移动机器人路径规划进行仿真,结果表明,混合算法可以很好地避免障碍物,快速找到一条机器人最优移动路径,十分适合于复杂环境路径规划。     

Algebraic theory of probabilistic processes

In this paper we extend de Nicola and Hennessy’s testing theory to deal with probabilities. We say that two processes are testing equivalent if the probabilities with which they pass any test are equal. We present three alternative semantic views of our testing equivalence. First, we introduce adequate extensions of acceptance sets (inducing an operational characterization) and acceptance trees (inducing a denotational semantics). We also present a sound and complete axiomatization of our testing equivalence. So, this paper represents a complete study of the adaptation of the classical testing theory for probabilistic processes.     

基于元组存在性的概率数据模型研究

随着信息与通讯技术的快速发展,数据管理正面临着越来越多的挑战,其中之一就是数据的不确定性。提出一种基于元组存在性的概率数据模型相似文献   

Methods for Quantitative Analysis of Probabilistic Pushdown Automata

We present several constructions and techniques which have recently been used to tackle the prob- lems of qualitative/quantitative analysis of probabilistic pushdown automata.     

Probabilistic verification of communication protocols

Complete behavior of a communication protocol can be very large. It is worth investigating whether partial exploration of the behavior generates reasonable results. We present such a procedure which performs partial exploration using most-probable-first search. Some of the ideas used in this procedure are based on a convolutional decoding procedure due to Jelinek and a performance evaluation procedure due to Rudin. Multiple trees of protocol behavior are constructed. Some results on estimating the probability of encountering an unexplored state in a finite run of a protocol are also presented. Nicholas F. Maxemchuk received the B.S.E.E. degree from the City College of New York, NY, and the M.S.E.E. and Ph.D. degrees from the University of Pennsylvania, Philadelphia. He is the Head of the Distributed Systems Research Department at AT & T Bell Laboratories, Murray Hill, NJ, and has been at AT & T Bell Laboratories since 1976. Prior to joining Bell Laboratories he was at the RCA David Sarnoff Research Center in Princeton, NJ for eight years. Dr. Maxemchuk has been on the adjunct faculties of Columbia University and the University of Pennsylvania. He has been an advisor to the United Nations on data networking and has been on networking panels for the US Air Force and DARPA. He has served as the Editor for Data Communications for the IEEE Transactions on Communications, as a Guest Editor for the IEEE Journal on Selected Areas in Communications, and has been on the program committee for numerous conferences and workshops. He was awarded the RCA Laboratories Outstanding Achievement Award, the Bell Laboratories Distinguished Technical Staff Award, and the IEEE's 1985 and 1987 Leonard G. Abraham Prize Paper Award. Krishan Sabnani received a BSEE degree from Indian Institute of Technology, New Delhi, India and a PhD degree from Columbia University, New York, NY. In 1981, he joined AT & T Bell Laboratories after graduating from Columbia University. He is currently working in the Distributed Systems Research Department of AT & T Bell Laboratories. His major area of interest is communication protocols. Dr. Sabnani was a co-chairman of the Eighth International Symposium on Protocol Specification, Testing, and Verification held in Atlantic City, NJ during June 1988. He is currently an editor of the IEEE Transactions on Communications and of the IEEE Transactions on Computers. He has served on the program committees of several conferences. He is also a guest editor of two special issues of the Journal on Selected Areas in Communications (JSAC) and the Computer Networks and ISDN Systems Journal, respectively.     

A unified view of class-selection with probabilistic classifiers

The possibility of selecting a subset of classes instead of one unique class for assignation is of great interest in many decision making systems. Selecting a subset of classes instead of singleton allows to reduce the error rate and to propose a reduced set to another classifier or an expert. This second step provides additional information, and therefore increases the quality of the result. In this paper, a unified view of the problem of class-selection with probabilistic classifiers is presented. The proposed framework, based on the evaluation of the probabilistic equivalence, allows to retrieve class-selective frameworks that have been proposed in the literature. We also describe an approach in which the decision rules are compared by the help of a normalized area under the error/selection curve. It allows to get a relative independence of the performance of a classifier without reject option, and thus a reliable class-selection decision rule evaluation. The power of this generic proposition is demonstrated by evaluating and comparing it to several state of the art methods on nine real world datasets, and four different probabilistic classifiers.     

A formal verification framework for SysML activity diagrams

SysML activity diagrams are OMG/INCOSE standard diagrams used for modeling and specifying probabilistic systems. They support systems composition by call behavior and send/receive artifacts. For verification, the existing approaches dedicated to these diagrams are limited to a restricted set of artifacts. In this paper, we propose a formal verification framework for these diagrams that supports the most important artifacts. It is based on mapping a composition of SysML activity diagrams to the input language of the probabilistic symbolic model checker called “PRISM”. To prove the soundness of our mapping approach, we capture the underlying semantics of both the SysML activity diagrams and their generated PRISM code. We found that the probabilistic equivalence relation between both semantics preserve the satisfaction of the system requirements. Finally, we demonstrate the effectiveness of our approach by presenting real case studies.     

True-concurrency probabilistic models: Markov nets and a law of large numbers

We introduce the model of Markov nets, a probabilistic extension of safe Petri nets under the true-concurrency semantics—this means that traces, not firing sequences, are given a probability. This model builds upon our previous work on probabilistic event structures. We use the notion of a branching cell for event structures, and show that the latter provides an adequate conception of local state for nets. We prove a Law of Large Numbers (LLN) for Markov nets, which constitutes the main contribution of the paper. This LLN allows for the characterization, in a quantitative way, of the asymptotic behavior of Markov nets.     

A three-dimensional probabilistic fuzzy control system for network queue management

A novel probabilistic fuzzy control system is proposed to treat the congestion avoidance problem in transmission control protocol （TCP） networks. Studies on traffic measurement of TCP networks have shown that the packet traffic exhibits long range dependent properties called self-similarity, which degrades the network performance greatly. The probabilistic fuzzy control （PFC） system is used to handle the complex stochastic features of self-similar traffic and the modeling uncertainties in the network system. A three-dimensional （3-D） membership function （MF） is embedded in the PFC to express and describe the stochastic feature of network traffic. The 3-D MF has extended the traditional fuzzy planar mapping and further provides a spatial mapping among ＂fuzziness-randomness-state＂. The additional stochastic expression of 3-D MF provides the PFC an additional freedom to handle the stochastic features of self-similar traffic. Simulation experiments show that the proposed control method achieves superior performance compared to traditional control schemes in a stochastic environment.     

Tank monitoring: a pAMN case study

The introduction of probabilistic behaviour into the B-method is a recent development. In addition to allowing probabilistic behaviour to be modelled, the relationship between expected values of the machine state can be expressed and verified. This paper explores the application of probabilistic B to a simple case study: tracking the volume of liquid held in a tank by measuring the flow of liquid into it. The flow can change as time progresses, and sensors are used to measure the flow with some degree of accuracy and reliability, modelled as non-deterministic and probabilistic behaviour respectively. At the specification level, the analysis is concerned with the expectation clause in the probabilistic B machine and its consistency with machine operations. At the refinement level, refinement and equivalence laws on probabilistic GSL are used to establish that a particular design of sensors delivers the required level of reliability.     

概率逻辑模型与学习研究进展

近年来，概率逻辑学习研究取得了很大进展，已经提出各种不同的形式化方法和学习方法，包括概率关系模（PRMs）、贝叶斯逻辑程序（BLPs）、逻辑贝叶斯网络（LBNs）和随机逻辑程序（SLPs）等。文章重点介绍了贝叶斯网络与一阶逻辑的结合，并以PRMs、BLPs和LBNs为例，描述了基于贝叶斯网络的概率逻辑模型（PLMs）的知识表示方法，给出了此类PLMs一般使用的参数估计方法和结构学习方法，并给出了建议的研究方向。