Research On Material Recognition Based On Multi-channel Electrostatic Signal In Complex Environment

When palletizing robots perform tasks such as picking,handling and palletizing of workpieces,front-end sensors can identify the type of object and help the robot complete more stable and efficient actions.Material identification technology based on electrostatic detection have the advantages of unaffected by lighting and smoke conditions,simple structural design,wide identification range.On the basis of predecessors' method of material identification based on electrostatic signal,by studying the surface electrostatic charge distribution and release characteristics and the relationship between environmental humidity,this paper presents a method of obtaining the static charge distribution characteristics of the material surface by setting up a multi metal electrode to achieve the purpose of material identification.First,the change of induced charge and contact charge of metal electrode in the process of electrostatic signal detection and contact electrode with different positions of dielectric material is studied.A theoretical model of multi-channel electrostatic signal material recognition is established based on electrostatic detection in complex environment.According to the theory of the optimization model,the design of dielectric materials of electrostatic charge detection system,including detection circuit,conditioning circuit,detection electrode and back-end detection mechanism,special structure,were collected and pretreatment on the electrostatic signal typical dielectric materials in indoor environment.The characteristic vectors of charge distribution,peak attenuation and synthetic feature vectors are defined as the eigenvalue vectors representing the discharge characteristics of material charges,and used as the input parameters of classifiers,and the classification algorithm is designed to classify and identify multiple kinds of dielectric materials.Finally,simulate the recognition process of material in the indoor environment and complex environment,application of humidity regulator and humidity control and observation environment humidity,through the static charge detection system to collect the electrostatic signal every kind of medium material under different humidity conditions respectively for feature extraction and classification.Experiments show that the proposed material recognition method based on multi-channel electrostatic signals has higher recognition accuracy than single electrode detection and recognition method,and it still maintains a high recognition rate in complex environment.