基于物联网的分拣机器人故障检测系统设计

Design of a Sorting Robot Fault Detection System Based on the IoT

目前研究的分拣机器人故障检测系统检测准确性较低,导致检测结果误差较大、实时性较差;为此,基于物联网设计一种新的分拣机器人故障检测系统;选用滑轮式机器人载体设定分拣机器人,硬件部分采用Zigbee压力传感器采集机器人故障信息,利用XBEE模块负责数据传输,协调分拣中控机接收各个传感器采集的信息,通过STMP3550芯片实现控制器设计;通过信息标定、信息采集、特征提取、故障识别实现软件工作流程,应用非极大值最大类间方差法来筛选出最优的高低阈值解,得到连续但含有假边缘的故障信息图像边缘;将提取到的图像特征向量映射到类型空间之中,确定故障原因,完成故障识别;实验结果表明,所设计分拣机器人故障检测系统在6次检验中都准确地检测出故障原因,故障检测耗时平均值为3.27 min,能够有效提高检测准确性,加强检测结果的实时性。

The current research on the sorting robot fault detection system has low detection accuracy,resulting in large errors in detection results and poor real-time performance.To this end,a new sorting robot fault detection system is designed based on the Internet of Things.A pulley-type robot carrier is selected to set the sorting robot.The hardware part adopts Zigbee pressure sensor to collect robot fault information,uses the XBEE module to be responsible for data transmission,coordinates the sorting central control machine to receive the information collected by each sensor,and realizes the controller design through the STMP3550 chip.The software workflow is realized through information calibration,information collection,feature extraction,and fault identification.The non-maximum maximum between-class variance method is used to screen out the optimal high and low threshold solutions,and the continuous but false edges of the fault information image edges are obtained.Map the extracted image feature vector to the type space,determine the cause of the fault,and complete the fault identification.The experimental results show that the designed sorting robot fault detection system accurately detects the cause of the fault in the six inspections,and the average time of fault detection is 3.27 min,which can effectively improve the detection accuracy and strengthen the real-time performance of the detection results.