一种具有全局定位能力的移动机器人传感混合算法

基于激光雷达(LiDAR)的移动机器人定位,由 于其传感反馈维度低,难以实现精确 的场景识别。尤其是在几何特征重复的环境中,全局定位更难以实现。基于该问题,提出了 一种基于传感融合的移动机器人定位系统,该系统具有全局定位的能力。 基于LiDAR和视觉 传感器的传感融合,利用视觉传感器引入的丰富信息和LiDAR的鲁棒几何信息反馈,实现可 靠的全局定位。本系统构建用于环境表达的混合栅格地图,并且利用视觉全局描述符以加速 定位收敛,结合姿态优化算法以提高定位精度。此外,引入截断触发机制实现对跟踪定位的 监督。经实验验证,本文提出方法的表现与现有定位系统比较,具有更强的定位收敛能力。 同时,所提出的传感混合算法能够成功处理全局定位问题,提高系统鲁棒性。     

自主移动机器人即时定位与地图构建方法研究

自主移动机器人的即时定位与地图构建是机器人领域中的关键问题。文中对自主移动机器人即时定位与地图构建研究的两个主要问题进行了分析,研究了不确定信息的描述和处理方法,以及数据关联的方法。综述了现阶段即时定位与地图构建的实现方法,其中包括两种卡尔曼滤波算法和粒子滤波算法,以及基于视觉的SLAM算法。通过分析现有算法的特点,提出了自主移动机器人即时定位与地图构建未来研究的发展方向。     

基于PF/CUKF/EKF的移动机器人SLAM框架算法

 提出了一种基于混合滤波的移动机器人同时定位与地图创建(SLAM)算法框架,并利用统计理论对SLAM算法进行一致性评估,该算法框架将机器人SLAM中的联合后验概率分布分解为机器人路径部分及以机器人路径为条件的地图部分,使滤波器变成低维滤波,能够有效地提高计算效率.采用约束的无色卡尔曼滤波(CUKF)算法并融合新的观测数据使提议分布更加接近后验概率分布,并且能够精确估计移动机器人的位姿,进而通过扩展卡尔曼滤波(EKF)算法更新特征地图的位置.仿真实验表明该混合滤波技术为SLAM算法提供了一种有效可靠的途径,在一定条件下与其他SLAM算法比较会得到更高的精度要求.     

基于多个传感器的VLC室内定位算法研究

为了实现移动机器人的快速高精度定位,提出了一种基于多个传感器的室内定位模型,研究了其可见光通信技术(VLC)室内定位算法,并对该算法进行了实验验证。首先研究基于AOA定位算法,利用传感器的响应曲线,结合室内定位模型,通过拟合预测算法计算出信号到达角度实现定位;然后综合多个传感器的定位模型和AOA定位算法,分析得出一种室内定位的实现方式,通过实验验证了该定位模型和定位算法的实现可行性。结果表明:其定位精度达13.6cm,定位周期为0.1s,相较于传统的AOA定位算法,该算法定位精度高、成本低、可行性高且定位速度快。     

基于自适应感知复位算法的移动机器人定位

本文在传统粒子滤波的基础上提出了一种基于自适应感知复位定位算法(ASRL:Adaptive Sensor Resetting Localization)的移动机器人定位方法.该方法通过带有权值的样本集描绘机器人的可信度,根据有效的样本数计算需要生成的新样本数,然后从感知分布中采样代替原来的样本.ASRL算法已经在安装有编码器和彩色摄像机两种传感器的实际移动机器人上进行实验,结果表明该算法鲁棒性更好,收敛更快.     

面向家庭环境的移动机器人自主导航系统

移动机器人在家庭环境下使用传统A*算法规划经过门的路线时,存在因规划的路径靠近障碍物而导致定位失败的问题.针对该问题,设计一种面向家庭环境的自主导航系统,并提出基于栅格-拓扑混合地图的SHS(Segmented Hybrid Search,SHS)路径规划方法.首先,在已建立的栅格地图上选取拓扑点构建栅格-拓扑混合地图;其次,通过Dijkstra算法搜索先验安全航路点序列,将航路点视为局部目标节点;最后,采用A*算法实现分段路径搜索.实验结果表明,在较复杂的家庭环境中,所提的算法能快速规划通过门的安全无碰撞路径.     

移动机器人的同时定位与地图创建的算法研究

移动机器人的同时定位与地图创建是当前机器人领域研究的基本问题与热点,也是实现智能机器人实现自主导航和控制决策的关键。本文从基于贝叶斯滤波器模型和基于图优化平滑模型两个方面对移动机器人同时定位与地图创建(SLAM)的研究进行了综述,介绍了这两种模型的基本框架以及关键技术,阐述了模型的各种不同实现形式,分析了这些算法的性能,并探讨了这些算法的优缺点以及相关难点的解决思路。最后对基于滤波器模型和基于平滑模型的同时定位与地图创建的未来发展进行展望。     

多模型ASM及其在人脸面部特征点定位中的应用

为了提高ASM在非均匀光照条件下的人脸面部特征点定位的精确度,提出了一种融合Log-Gabor小波特征的多模型ASM方法.该方法的主要特点有:在精确定位目标图像虹膜位置的基础上对全局形状模型进行较准确的初始化;特征点局部纹理特征采用灰度和Log-Gabor小波特征共同描述,减少光照和噪音对算法的影响;建立包括全局ASM和基于人脸面部显著特征区域的局部ASM的多模型ASM,交替使用这两种ASM模型在边缘约束策略基础上对特征点的定位结果进行约束.实验表明,多模型ASM算法对人脸面部特征点定位的准确率比传统ASM算法有明显提高.     

基于无线传感器网络的移动机器人定位导航系统

设计了一种基于无线传感器网络(WSN)的移动机器人轨迹跟踪定位系统,通过测算移动机器人和信标节点之间的无线电接收信号强度(RSSI)进而估计出它们之间的距离,采用自适应的扩展卡尔曼滤波算法对RSSI数据处理得到机器人的当前位置、速度、加速度等状态信息.为了提高系统的实时性和计算效率,动态选用网络中一部份信标节点进行轨迹跟踪定位计算,降低了移动机器人CPU的负担.采用CC2430芯片作为节点的通信和数据处理单元,现场试验结果表明该方法具有轨迹跟踪定位精度高、实现简单、成本低廉等特点.     

基于小生境遗传算法的移动机器人路径优化

针对标准遗传算法存在收敛性慢和局部最优解的缺陷,结合移动机器人行走特点,提出一种基于预选择机制小生境技术的改进遗传算法中移动机器人路径规划方法.该方法兼顾对局部最优解和全局最优解的搜索,维持群体的多样性,避免了早期收敛现象的发生;同时也增强了自然群体进化的并行性,加快了搜索进程.计算机仿真结果表明,该算法在收敛速度和输出全局最优解概率方面相对于标准遗传算法有了显著提高.     

一种移动机器人的运动控制方法研究

针对反应式移动机器人运动控制中存在的缺乏自主运动规划等问题,介绍了一种新型移动机器人的运动控制方法。设计了一种混合式控制结构实现移动机器人的运动控制。采用基于行为的运动控制策略,将反应式行为和规划行为融合在一起。实验结果表明该策略能够很好地完成复杂环境下移动机器人的运动控制。     

PSD – probabilistic algorithm for mobile robot 6D localization without natural and artificial landmarks based on 2.5D map and a new type of laser scanner in GPS-denied scenarios

This paper presents an approach to mobile robot 6D localization based on a 3D laser scanner in GPS-denied scenarios. Commonly, 6D localization using laser scanners is performed with the use of extraction and association of the features or by comparison of the whole scans (very often off-line) using the ICP algorithm or its modifications. However, in some unstructured non-urbanized rough terrain environments, feature extraction does not seem to be reliable enough. For such kind of environment, we present a new method to mobile robot localization in GPS-denied applications, called PSD (Point-to-Surfel Distance). Unlike state of the art localization methods using laser scanners, we consider every single laser scanner measurement as an observation and use Point-to-Surfel Distance for correction of position and orientation of the robot. Mobile robot localization is based on a specific representation of the terrain in the 2.5D surfel map (terrain height and inclination). The simulation tests compared our method using extended Kalman filter (EKF) and single laser scanner measurements with an up-to-date method using particle filter (PF) and comparing the scan lines with the reference map and with another method using Gaussian mixture maps. The tests confirmed that the proposed method provides satisfying results for GPS-denied scenarios in rough terrain without extractable landmarks and our method is thirty times faster than the PF method (serial implementation). KITTI benchmark tests and real terrain experiments confirmed its usefulness and advantages as well.     

A behavior-based mobile robot with a visual landmark-recognition system

In this paper, based on behavior-based artificial intelligence we have built a fully autonomous mobile robot. Several modules are developed for the mobile robot to implement different levels of competences and behaviors, where each module itself generates behaviors. New modules can be easily added to the robot system to improve in the competence without changing any existing modules. A vision-based landmark recognition system for robot navigation is developed as the highest layer in the subsumption architecture. A genetic-algorithm-based search method for pattern recognition of digital images is proposed and implemented to recognize artificial landmarks by searching all the predefined patterns. The vision layer is capable of generating the desired behaviors corresponding to various landmarks. A combination of eight ultrasonic sensors is designed to implement obstacle-avoidance behaviors through a set of fuzzy rules. The effectiveness of this behavior-based mobile robot is demonstrated by experimental studies.     

Autonomous Mobile Robot Navigation Using Passive RFID in Indoor Environment

This paper proposes an efficient method for localization and pose estimation for mobile robot navigation using passive radio-frequency identification (RFID). We assume that the robot is able to identify IC tags and measure the robot's pose based on the relation between the previous and current location according to the IC tags. However, there arises the problem of uncertainty of location due to the nature of the antenna and IC tags. In other words, an error is always present which is relative to the sensing area of the antenna. Many researches have used external sensors in order to reduce the location errors, with few researches presented involving purely RFID driven systems. Our proposed algorithm that uses only passive RFID is able to estimate the robot's location and orientation more precisely by using trigonometric functions and the IC tags' Cartesian coordinates in a regular gridlike pattern. The experimental results show that the proposed method effectively estimates both the location and the pose of a mobile robot during navigation.      

Fuzzy neural network-based real-time self-reaction of mobile robot in unknown environments

In this paper, a hybrid intelligent method including fuzzy inference and neural network is presented for real-time self-reaction of a mobile robot in unknown environments. A neural network with fuzzy inference (fuzzy neural network, FNN) presented can effectively improve the learning speed of the neural network. The method can be used to control a mobile robot based on the present motion situations of the robot in real-time; these situations include the distances in different directions between the obstacles and the robot provided by ultrasonic sensors, the target orientation sensed by a simple optical range-finder and the movement direction of the robot. Simulation results showed that the above method can quickly map the fuzzy relationship between the inputs and the output of the control system of the mobile robot.     

基于移动式全景成像的增强现实关键技术研究

将全景成像和单自由度移动机器人的关键技术应用到增强现实领域中,解决了当图像采集装置与头盔显示器分离并且图像采集装置在场景中移动情况下的增强现实系统的技术难点。建立完整的基于全景成像的增强现实系统,将单自由度移动机器人引入增强现实领域,并提出图像共享的概念。实验表明,全景成像方法对于亮度差异较大图像的拼接效果良好,适宜于室外复杂环境的图像拼接,全景图像的局部增强显示效果良好。     

ALO: An ultrasound system for localization and orientation based on angles

This paper presents a low cost system based on ultrasound transducers to obtain the localization and orientation information of a mobile node, such as a robot, in a 2D indoor space. The system applies a new differential time of arrival (DTOA) technique with reduced computational cost, which is called ALO (angle localization and orientation). Instead of directly calculating its position, the system calculates the direction of arrival of the received ultrasonic signal and, through it, its position and orientation. A prototype of a robot has been built in order to show the validity of the method through experimental results.     

Design and implementation of an adaptive sliding-mode dynamic controller for wheeled mobile robots

This paper designs an adaptive sliding-mode dynamic controller for wheeled mobile robots to implement the trajectory-tracking mission. First, a kinematic controller is introduced for the wheeled mobile robot. Secondly, the adaptive sliding-mode dynamic controller is proposed to make the real velocity of the wheeled mobile robot reach the desired velocity command, although the wheeled mobile robot is even with system uncertainties and disturbances. The convergence of the complete equations of motion of the wheeled mobile robot is proved by the Lyapunov stability theory. Computer simulation results illustrate the effectiveness of the proposed control schemes. Finally, the real-time experiments of the adaptive sliding-mode dynamic controller on the test ground demonstrate the feasibility of practical wheeled mobile robot maneuvers.     

Optimal control at energy performance index of the mobile robots following dynamically created trajectories

In practice, the problem of motion control of the wheeled mobile robots is often neglected. Wheeled mobile robots are strongly nonlinear systems and restricted by non-holonomic constraints. Motion control of such systems is not trivial task and usage of non-optimal control signals can lead to deterioration of the overall robot system's performance. In case of autonomous application of the mobile robots all parts of its control system should work perfectly. The paper presents the theory and application of the optimal control method at the energy performance index towards motion control of the two-wheeled mobile robot during the realisation of complex, dynamically created trajectories. With the use of the proposed control method the two-wheeled mobile robot can realise effectively the desired trajectory, which is generated ad-hoc by the navigation system of the robot. Thus the proposed method can be used for motion control of autonomous or semi-autonomous wheeled mobile robots. The presented results of both computer simulations and experiments indicate that the proposed method works effectively from the point of view of the motion control of two-wheeled mobile robot. Movement of the mobile robot appeared reliable and predictable during all the tests.