移动机器人的同步自定位与地图创建研究进展

自主移动机器人在未知环境下作业时,首先要解决的基本问题就是其自身的定位问题,而定位问题与环境地图的创建又是相辅相成的.本文从相关理论和关键技术等方面,系统地总结了同步自定位和地图创建的研究现状,着重介绍了基于概率论的方法,分析了目前存在的难题,并指出了未来研究的发展方向.     

未知环境下移动机器人同步地图创建与定位研究进展

移动机器人同步地图创建与定位(SLAM)是移动机器人的核心研究课题.本文对SLAM的最新研究进展和关键技术进行了综述:并从地图创建模型、计算复杂度和算法鲁棒性等方面对现有方法进行了对比分析.最后总结分析了SLAM研究存在的难题,探讨了今后的发展方向.     

基于建筑特征及二维地图的复杂城市场景中移动机器人视觉定位算法

针对城市环境中全球定位系统(GPS)信号易受到高层建筑遮挡而无法提供准确位置信息的问题,提出了一种基于建筑物竖直侧平面特征及建筑物二维轮廓地图的移动机器人定位方法。该方法利用车载视觉,首先对两视图间的竖直直线特征进行匹配;然后基于匹配的竖直线特征对建筑物的竖直侧平面进行重建;最后,利用建筑物竖直侧平面特征及建筑物二维俯视轮廓地图,设计了一种基于随机采样一致性(RANSAC)的移动机器人视觉定位算法,从而解决了在建筑物方向任意的复杂城市环境中的机器人定位问题。实验结果表明,算法的平均定位误差约为3.6m,可以有效地提高移动机器人在复杂城市环境中自主定位的精度及鲁棒性。     

Appearance-based concurrent map building and localization

In appearance-based robot localization the environment map does not represent geometrical features but consists of an appearance map, which is a collection of robot poses and corresponding sensor observations. In this paper, we describe a concurrent map-building and localization (CML) system based on a multi-hypotheses tracker that is able to build and refine autonomously the appearance map required for localization as the robot moves in the environment. The results included in this paper validate our approach.     

受限移动机器人构建地图的最优探测法

对复杂未知环境构建地图是移动机器人面临的一大问题.通常忽略未知环境的几何特征,将其抽象成未知无向连通图,机器人只沿着图的边进行搜索,并将走过每条边的成本看成是1.机器人构建地图的成本用走过的总边数来表示.对于一个完全未知的环境,从一点S出发,限制移动机器人最远能走r(如燃料问题及安全线或通信线等)步(边数)的范围内,基于深度受限剪枝生成子树的方法,结合广度优先搜索和受限的深度优先搜索染色策略,给出了对未知环境构建完整地图的有效算法,该算法的成本为|E|+O|V|,这是目前最优结果.     

基于声纳的移动机器人环境建图的设计与实现*

针对未知环境中声纳传感器定位与地图创建时传感数据不确定性高、可靠性低的问题,提出了一种新的室内环境建图方法。该方法建立容忍函数以判断噪声和镜面反射,同时借鉴了ArcTransversal Median Algorithm的思想和栅格概率估计并采用贝叶斯法则进行两次数据融合以减小声纳传感器信息的不确定性。在MORCS2机器人平台上实时创建地图实验表明,这种方法能快速实现从局部地图到全局地图的更新且有较好的精确性与鲁棒性。     

基于聚类匹配的移动机器人地图实时创建算法

提出了一种基于模式识别聚类思想的数据点集匹配算法.该匹配算法具有传统迭代匹配算法和非迭代匹配算法的优点,匹配速度快,精度高.结合上述匹配算法,给出了一种基于激光测距仪的移动机器人环境地图实时创建方法.该方法使用从环境数据中提取出的特征点来完成两组激光数据点集的匹配,进而完成环境地图的创建.利用本实验室自主研发的救援机器人平台对该算法进行了验证,实验结果表明,该算法能够完成室内环境下移动机器人实时准确有效的环境地图创建.     

Simultaneous localization and odometry self calibration for mobile robot

This paper presents both the theory and the experimental results of a method allowing simultaneous robot localization and odometry error estimation (both systematic and non-systematic) during the navigation. The estimation of the systematic components is carried out through an augmented Kalman filter, which estimates a state containing the robot configuration and the parameters characterizing the systematic component of the odometry error. It uses encoder readings as inputs and the readings from a laser range finder as observations. In this first filter, the non-systematic error is defined as constant and it is overestimated. Then, the estimation of the real non-systematic component is carried out through another Kalman filter, where the observations are obtained by two subsequent robot configurations provided by the previous augmented Kalman filter. There, the systematic parameters in the model are regularly updated with the values estimated by the first filter. The approach is theoretically developed for both the synchronous and the differential drive. A first validation is performed through very accurate simulations where both the drive systems are considered. Then, a series of experiments are carried out in an indoor environment by using a mobile platform with a differential drive.     

Object-spatial layout-route based hybrid map and global localization for mobile robots

This paper presents new object-spatial layout-route based hybrid map representation and global localization approaches using a stereo camera. By representing objects as high-level features in a map, a robot can deal more effectively with different contexts such as dynamic environments, human-robot interaction, and semantic information. However, the use of objects alone for map representation has inherent problems. For example, it is difficult to represent empty spaces for robot navigation, and objects are limited to readily recognizable things. One way to overcome these problems is to develop a hybrid map that includes objects and the spatial layout of a local space. The map developed in this research has a hybrid structure that combines a global topological map and a local hybrid map. The topological map represents the spatial relationships between local spaces. The local hybrid map combines the spatial layout of the local space with the objects found in that space. Based on the proposed map, we suggest a novel coarse-to-fine global localization method that uses object recognition, point cloud fitting and probabilistic scan matching. This approach can accurately estimate robot pose with respect to the correct local space. Recommended by Editor Jae-Bok Song. This research was performed for the Intelligent Robotics Development Program, one of the 21st Century Frontier R&D Programs funded by the Ministry of Knowledge Economy of Korea. Soonyong Park received the B.S. and M.S. degrees from the Department of Mechanical Engineering, Kyunghee University, Seoul, Korea, in 2001 and 2003, respectively. He is currently working toward the Ph.D. degree in the Department of Electrical and Electronic Engineering, Yonsei University, Seoul, Korea. Since 2001, he has been a student researcher in the Center for Cognitive Robotics Research, Korea Institute of Science and Technology (KIST), Seoul, Korea. His research interests include mobile robot navigation and computer vision. Mignon Park received the B.S. and M.S. degrees in Electronics from Yonsei University, Seoul, Korea, in 1973 and 1977, respectively. He received the Ph.D. degree in University of Tokyo, Japan, 1982. He was a researcher with the Institute of Biomedical Engineering, University of Tokyo, Japan, from 1972 to 1982, as well as at the Massachusetts Institute of Technology, Cambridge, and the University of California Berkeley, in 1982. He was a visiting researcher in Robotics Division, Mechanical Engineering Laboratory, Ministry of International Trade and Industry, Tsukuba, Japan, from 1986 to 1987. He has been a Professor in the Department of Electrical and Electronic Engineering in Yonsei University, since 1982. His research interests include fuzzy control and application, robotics, and fuzzy biomedical system. Sung-Kee Park is a principal research scientist for Korea Institute of Science and Technology (KIST). He received the B.S. and M.S. degrees in Mechanical Design and Production Engineering from Seoul National University, Seoul, Korea, in 1987 and 1989, respectively. He received the Ph.D. degree (2000) from Korea Advanced Institue of Science and Technology (KAIST), Korea, in the area of computer vision. Since then, he has been working for the center for cognitive robotics research at KIST. During his period at KIST, he held a visiting position at the Robotics Institute of Carnegie Mellon University in 2005, where he did research on object recognition. His recent work has been on cognitive visual processing, object recognition, visual navigation, and human-robot interaction.     

基于声纳传感器的移动机器人地图创建方法研究*

提出了一种改进的基于声纳传感器信息进行栅格地图创建的方法。将Bayes法则用于移动机器人地图创建,对多个声纳传感器信息进行融合,解决信息间的冲突问题,并根据声纳模型将测量数据集成到局部地图中,改变栅格被障碍物占有的概率。经过坐标变换后,利用Bayes法则更新全局地图中的栅格信息,实现从局部地图到全局地图的更新。实验验证了该算法的可行性与有效性。     

基于全景视觉的移动机器人地图创建与定位

提出一种高效的基于全景视觉的室内移动机器人地图构建和定位方法. 该方法充分利用全景视觉系统视野广阔、获取环境信息完整的特点, 根据全景图像生成环境描述子; 利用上述环境描述子描述环境, 创建拓扑地图, 将地图表示为环境描述子的集合. 在此基础上, 提出一种基于贝叶斯理论的定位方法, 根据当前全景摄像头的观测值, 利用已生成的地图完成状态跟踪, 全局定位和“绑架”定位. 最后通过实验验证了该方法的有效性, 并给出了计算成本分析.     

基于粒子群算法的单目移动机器人地图创建

提出了一种基于粒子群优化算法的移动机器人地图创建方法,该方法模型简单,算法复杂度低,收敛速度快.首先对实验环境进行了描述,通过建立模型推导出了路标的坐标公式,提出了问题的关键.详细介绍了粒子群优化的工作原理,论述了该算法在地图创建中的具体实现.通过与算术平均法实验结果比较,证实了该方法的有效性.     

Environmental map building for a mobile robot using infrared range-finder sensors

This paper presents a methodology for building a high-accuracy environmental map using a mobile robot. The design approach uses low-cost infrared range-finder sensors incorporating with neural networks. To enhance the map quality, the errors occurring from the sensors are corrected. The non-linearity error of the sensors is compensated using a backpropagation neural network and the random error of readings including the uncertainty of the environment is taken into a sensor model as a probabilistic approach. The map is represented by an occupancy grid framework and updated by the Bayesian estimation mechanism. The effectiveness of the proposed method is verified through a series of experiments.     

Online topological map building and qualitative localization in large-scale environment

A novel topological map representation as well as an online map construction approach is presented in this paper. By virtue of the topological map whose nodes are represented with the free beams of the laser range finder together with the visual scale-invariant features, the mobile robot can autonomously navigate in unknown, large-scale and indoor environments. Different from the traditional navigation methods that rely on precise global localization, the robot locates itself qualitatively by location recognition rather than calculating its global pose in the world reference frame. By combining the reactive navigational method, Beam Curvature Method (BCM), with the topological map, a smooth, real-time navigation without precise localization can be realized.     

基于WiFi指纹序列匹配的机器人同步定位与地图构建

同步定位与地图构建（SLAM）是当前机器人定位导航的研究热点,可靠的闭环检测是图优化SLAM的关键。而在大范围的复杂环境下,通过视觉或激光雷达进行闭环检测的可靠性低且计算开销大。针对这一问题,提出了一种基于WiFi指纹序列匹配的图优化SLAM算法。所提算法采用指纹序列进行闭环检测,由于指纹序列中包含多个指纹数据,信息量比单个指纹点对的数据丰富,因此将传统的基于指纹点对的匹配扩展到指纹序列的匹配可以大幅减小闭环误判的几率,从而确保了闭环检测的准确性,满足了SLAM在大范围复杂环境下的算法高精度要求。采用两组实验数据（机器人从不同的起点开始）对所提算法进行验证的结果表明：与高斯相似度的方法相比,所提算法的精度在第一组数据上提高了22.94%;在第二组数据上提高了39.18%。实验结果充分验证了所提算法在提高定位精度、确保闭环检测可靠性方面的优越性。     

Continuous mobile robot localization by using structured light and a geometric map

A mobile robot needs to know its position and orientation with accuracy in order to decide the control actions that permit it to finish the entrusted tasks successfully. To obtain this information, dead-reckoning-based systems have been used, and more recently inertial navigation systems. However, these systems have some errors that grow bigger as time goes by, therefore a moment comes when the information provided is useless. Because of this, there should be a periodic process that updates the robot position and orientation of the vehicle. The process to determine the robot position and orientation by using information originated from the external sensors is defined as the mobile robot relocalization. It is obvious that the greater the frequency of this process, the better the knowledge of its position the robot will have, and therefore its movements will be better directed to the point it must reach. The algorithm to achieve this can be classified in two large groups: relocalization through an a priori map of the environment and relocalization through the detection of landmarks present in that environment. The algorithm presented in the paper belongs to the first case. The sensor used is a combination of a laser diode and a CCD camera. The sensorial information is modelled as straight lines that will be matched with an a priori map of the environment. With this, the position of the mobile robot is estimated. The matching process is accomplished within an extended Kalman filter. The algorithm is able to work in real time, and it actualizes the position of the robot in a continuous way.     

Real-time map labelling for mobile applications

It is essential to label roads, landmarks, and other important features on maps for mobile applications to help users to understand their location and the environment. This paper aims to examine real-time map labelling methods suitable for the small screen on mobile devices. A slider method with a continuous search space was proposed to sequentially label both line and point features. The method starts with defining a range box for possible locations of the label. Then a search is performed, and the range box is reduced, if there are any cartographic objects that overlap the range box. Finally, the label is placed, at the best possible position in the reduced range box according to normal cartographic preferences, where it does not obscure any cartographic object. We implemented this method in a Java environment using the open source library JTS Topology Suite. A case study showed sound cartographic results of the labelling and acceptable computational efficiency.     

Simplified map representation and map learning system for autonomous navigation of mobile robots

Autonomous mobile robots need environmental maps to navigate to specific destinations, but there are difficulties in generating and acquiring efficient maps for them. Map learning systems and map representation for autonomous robot navigation are highly interrelated and need a total system design that combines these two factors. This study considers a combined simple map representation and map learning system. The proposed map representation includes geometrical relationships between important places and grid maps for these places, but not a total grid map of the environment. In particular, the study focuses on the ability to recognize places based on image features. Successful experiments on autonomous navigation with the proposed map representation using an actual mobile robot are described.     

Point cloud data map creation from factory design drawing for LiDAR localization of an autonomous mobile robot

Artificial Life and Robotics - When using an autonomous mobile robot, an environmental map should be created in advance. In this study, we propose a method for creating a point cloud data (PCD) map...