【作者】 周兵；

【导师】 卫琳娜；

【作者基本信息】 安徽工业大学 ， 计算机技术， 2022， 硕士

【摘要】 在无线传感器网络中,因无线节点能量耗尽或节点部署不均匀等因素的影响,容易造成检测区域出现覆盖洞,降低网络的连通性与覆盖率,服务质量难以得到有效保障。因此,实施覆盖洞修复需要利用移动节点(移动机构搭载普通无线节点)按照一定的轨迹移动。然而,这种移动机构存在一定的缺陷,如车轮卡涩或车轮打滑等因素以及移动节点受到风力和定位精度等影响,从而造成移动机构定位不准。针对此问题,本研究通过加入PID算法以及卡尔曼滤波的方式实现移动节点。移动节点由三部分组成:移动机构、定位模块与普通无线节点。首先,这三个部分通过串口进行通信,以便实时的获取数据,考虑到低功耗的MCU串口资源较少,所以通过IO进行模拟,解决串口资源不足的问题;其次,移动机构的轮速加入PID算法,解决发送一定数量PWM,车轮也会运行相应的圈数,不会因为车轮卡涩而造成脉冲丢失的情况产生;最后,移动节点的定位受到定位模块精度影响,会出现观测误差,为解决观测误差,引入卡尔曼滤波解决定位精度问题。移动节点实现后,为了确保移动节点能有效的进行覆盖洞修补,需要对移动节点进行路径规划,指引移动节点移动到目标位置。然而,现有路径规划A*算法需要使用全局地图并且要求障碍位置已知。为解决未知环境下的路径规划问题,本研究在障碍物未知情况下,通过估价函数确定移动节点附近某一邻居节点作为临时过渡点,使用A*算法进行二次路径规划,绕开障碍物后再进行三次路径规划,实现移动节点至目标位置的快速路径规划。同时,移动节点会将检测到的障碍物信息供给附近的一跳邻居节点,为后续经过此处的移动节点供信息。在障碍物已知环境下,对传统A*算法中的估计代价值进行改进,使用向量积的方式确定移动过程中的临时过渡点,降低A*算法的计算量,高移动节点的运行效率。仿真实验和实际实验结果显示,本研究实现了移动节点并对其进行了有效的以实现覆盖洞修补为目的路径规划。更多还原

【Abstract】 In wireless sensor networks,due to the influence of factors such as wireless node energy depletion or uneven node deployment,it is easy to cause coverage holes in the detection area,reduce network connectivity and coverage,and it is difficult to effectively guarantee the quality of service.Therefore,the implementation of coverage hole repair requires the use of mobile nodes(mobile mechanisms carrying ordinary wireless nodes)to move along a certain trajectory.However,this mobile mechanism has some defects,such as wheel jamming or wheel slipping and other factors,as well as the impact of wind and positioning accuracy on the mobile node,resulting in inaccurate positioning of the mobile mechanism.To solve this problem,this research implements mobile node by adding PID algorithm and Kalman filter.The mobile node consists of three parts: mobile mechanism,positioning module and common wireless node.First of all,these three parts communicate through the serial port to obtain data in real time.Considering the low power consumption MCU serial port resources are less,we simulate through IO to solve the problem of insufficient serial port resources;Secondly,PID algorithm is added to the wheel speed of the mobile mechanism to solve the problem that when a certain amount of PWM is sent,the wheel will also run for a corresponding number of cycles,and the pulse will not be lost because the wheel is jammed;Finally,the positioning of mobile nodes is affected by the accuracy of the positioning module,and there will be observation errors.In order to solve the observation error,Kalman filter is introduced to solve the positioning accuracy problem.After the mobile node is implemented,in order to ensure that the mobile node can effectively repair the coverage hole,it is necessary to plan the path of the mobile node to guide the mobile node to move to the target location.However,the existing path planning A* algorithm requires the use of a global map and requires that the obstacle location be known.In order to solve the path planning problem in an unknown environment,this research determines a neighbor node near the mobile node as a temporary transition point through the evaluation function when the obstacle is unknown,uses A* algorithm to carry out secondary path planning,bypasses the obstacle,and then carries out third path planning,so as to achieve fast path planning from the mobile node to the target location.At the same time,the mobile node will provide the detected obstacle information to the nearby one hop neighbor node to provide information for the subsequent mobile nodes passing through here.In the environment where the obstacles are known,the estimated generation value in the traditional A* algorithm is improved,and the temporary transition points in the moving process are determined by using the vector product to reduce the calculation amount of A* algorithm and improve the operation efficiency of mobile nodes.The simulation and actual experiment results show that this research has realized the mobile node and made effective path planning for the purpose of covering hole repair.更多还原