基于无线传感器网络的移动机器人导航研究

发布时间：2018-08-25 17:36

【摘要】：救援机器人能够代替人员深入火场、矿井等危险的现场进行救援,受到了国内外研究机构的广泛关注。然而,传统救援机器人受到遥控距离短、感知范围有限、定位精度不准确等限制,对其实际应用造成了很大的障碍。本文将移动机器人与无线传感器网络(WSN)相结合,研究基于无线传感器网络的移动机器人导航系统,具有积极的科学价值和现实意义。针对传统遥控和无线控制作用距离有限的问题,本文设计并实现了无线传感器网络多跳通信链路,通过WSN自组网的功能,形成远大于传统方式的机器人控制链路,能够很好地满足复杂环境灾难救援的需求。同时,由无线传感器网络节点携带温度、湿度、气体浓度等传感器,实时感知周围环境的信息。机器人的机动性和强计算能力以及无线传感器网络的通信距离、分布式计算、环境感知等能力相结合,可以显著提高救援任务的执行效率。为增强移动机器人的运动控制能力,本文引入ROS机器人操作系统设计和构建了一种移动机器人平台,硬件采用了iRobot差分驱动底盘和UTM 30LX激光雷达。通过建立数学模型并进行分析,仅依托里程计的航迹推算方法具有较大的误差,而且随时间单调递增。因此,本文使用环境特征点对其进行矫正,通过扩展卡尔曼滤波的不断迭代,使得系统趋于稳定,误差协方差不随时间变化。在此基础上,针对室内环境,提出了基于Hector SLAM的同步定位与构图算法,使用激光雷达的数据匹配实现定位和局部地图到全局地图的连接。实验结果表明,所构建的移动机器人平台具有很好的鲁棒性,能够准确地构建室内环境的二维平面图,对于存在干扰的地面也能达到较高的精度。结合无线传感器网络全局动态感知的特点,本文提出了一种室内环境的移动机器人导航方法。将具备超声波传感器的WSN节点部署于天花板,通过预设的网格形式进行均匀放置。通过WSN节点感知其正下方的环境信息,在数据离散化后得到占有栅格地图,获得了动态环境的实时模型。在此基础上,本文引入D*Lite算法和PRM算法实施动态环境下的移动机器人路径规划。仿真实验表明,D*Lite算法能够在目标区域出现动态障碍时进行重规划,并显著减小对于包含动态障碍物的环境的路径规划时间。
[Abstract]:The rescue robot can replace the personnel to go deep into the fire, mine and other dangerous sites to rescue, has been widely concerned by domestic and foreign research institutions. However, the traditional rescue robot is limited by the short distance of remote control, limited range of perception and inaccurate positioning accuracy, which has caused great obstacles to its practical application. This paper combines mobile robot with wireless sensor network (WSN) to study the navigation system of mobile robot based on wireless sensor network, which has positive scientific value and practical significance. Aiming at the problem of limited distance between traditional remote control and wireless control, this paper designs and implements the multi-hop communication link of wireless sensor network. Through the function of WSN ad hoc network, the robot control link which is far larger than the traditional mode is formed. Can well meet the needs of complex environmental disaster relief. At the same time, the wireless sensor network nodes carry sensors such as temperature, humidity, gas concentration and so on, and perceive the information of the surrounding environment in real time. The combination of mobility, strong computing ability, communication distance of wireless sensor network, distributed computing, environment perception and so on can significantly improve the efficiency of the rescue mission. In order to enhance the motion control ability of the mobile robot, a mobile robot platform is designed and constructed by introducing the ROS robot operating system. The iRobot differential drive chassis and the UTM 30LX lidar are used in the hardware. By establishing and analyzing the mathematical model, it is found that the track calculation method based only on the mileage meter has a large error and increases monotonously with time. Therefore, the environment feature point is used to correct it, and the system is stabilized by extended Kalman filter iteration, and the error covariance does not change with time. On this basis, a synchronous localization and mapping algorithm based on Hector SLAM is proposed for indoor environment. The location and the connection of local map to global map are realized by using the data matching of lidar. The experimental results show that the mobile robot platform has good robustness, can accurately construct the two-dimensional plan of indoor environment, and can achieve high accuracy for the ground with interference. Based on the characteristics of global dynamic sensing in wireless sensor networks, a mobile robot navigation method in indoor environment is proposed in this paper. The WSN nodes with ultrasonic sensors are deployed to the ceiling and placed evenly through a preset mesh. The WSN node perceives the environmental information directly below it, and after the data is discretized, the occupied grid map is obtained, and the real-time model of the dynamic environment is obtained. On this basis, D*Lite algorithm and PRM algorithm are introduced to implement path planning of mobile robot in dynamic environment. Simulation results show that the DLite algorithm can replan the target area and reduce the path planning time for the environment with dynamic obstacles.
【学位授予单位】：东北大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP242;TP212.9;TN929.5

【相似文献】