Wi-Fi Signal Heat Map Automatic Construction Method For Smart Factories

Following the advancement of flexible manufacturing,automated guided vehicles(AGV)in smart factories have been widely used.AGV usually uses Wi-Fi communication technology to achieve interaction with the background control center.The signal coverage of plant Wi-Fi access points(AP)directly affects the reliability of AGV communication.Aiming at the problem of low efficiency in manual evaluation of signal coverage performance,this article uses the low-cost inertial navigation and the Wi-Fi module on the AGV to realize the trajectory positioning of the AGV,and further realize the automatic construction of the Wi-Fi signal heat map in the factory.The main content of this article includes:1.Propose a method of IMU inertial navigation correction based on AGV motion constraints.The inertial navigation is used to dynamically detect the motion behaviors of the AGV,such as static,uniform speed,acceleration and deceleration,and combined with the error-state Kalman filter(ESKF),by performing corresponding posture and speed correction processing for different motion intervals,the drift error of the inertial navigation is suppressed,and the correction of the AGV positioning error is realized.Both the simulation experiment and the actual measurement of the three scenes show that the method reduces the root mean square error(RMSE)of each experiment scene by more than 99%.2.Designed a fusion positioning framework of inertial navigation and Wi-Fi.Using trajectory posture points and corresponding Wi-Fi signals,and a small amount of calibration point information,establish inertial navigation constraints,calibration point matching constraints,and curve matching constraints between global posture points,and optimize the graph of trajectory posture points to realize the improvement of positioning accuracy.Both the simulation experiment and the actual measurement of the three scenes show that the method further reduces the position root mean square error of each experiment scene by more than 86%,and the positioning accuracy reaches within 2m.3.Using Gaussian process regression(GPR)to realize the construction of Wi-Fi signal heat map in the factory.Combining the coordinate points of the positioning track and the corresponding Wi-Fi signals as a training sample set,the signal distribution is estimated through Gaussian process regression,and then a heat map of the wireless coverage of the factory area is constructed,which provides an intuitive means for the evaluation of the wireless coverage performance of the factory area and provides a reference for the reasonable erection and troubleshooting of APs.