Research On The Technology Of Outdoor Integrated Navigation System For Hospital Wheelchair

Positioning navigation is an indispensable technology for intelligent wheelchairs for automatic navigation in hospital scenes,and whether they can work normally is related to the personal safety of users.In outdoor scenes facing hospitals,complex environments such as centralized tall buildings and trees often cause loss of satellite signals and paralyze positioning and navigation systems.In order to solve this problem,this paper uses multi-sensor information fusion technology and algorithms based on recurrent neural networks,in the case of long-term signal loss,the system position and navigation module still has positioning and navigation capabilities,while ensuring a certain positioning accuracy and anti-interference ability,the main work of this paper is as follows:(1)According to the characteristics of the noise signal output by MEMS-IMU,the distribution and cause of the noise are studied;this paper proposes IMU filtering method based on recurrent neural network of gated unit,and trains and construct GRURNF model;the experimental results show that relative to the original data,after the data processed by this method,the IMU measurement error is reduced by about 65%,and the volatility of the inertial navigation system(INS)solution result is significantly reduced.(2)In order to achieve the requirements of high-precision positioning and strong stability of the wheelchair navigation system in the hospital scene,this paper uses realtime kinematics(RTK)technology for satellite positioning and uses extended Kalman filter(EKF)to fuse RTK and inertial navigation system(INS)information,this paper conducts wheelchair driving experiments in a hospital environment and analyzes the results of trajectory coordinates,driving speed,and posture,the experimental results show that the RTK/INS fusion system has a faster response speed,and the system can still maintain a positioning accuracy of about 0.3m even when the satellite signal is lost for a short time(about 10s).(3)In order to further improve the navigation system’s ability to locate and navigate in the weak signal or signal loss environment,this paper proposes an auxiliary INS pseudo measurement algorithm based on LSTM.When the satellite signal reception is normal,the system start to train the model,when entering the signal loss environment,the system use the trained model to predict the position of the wheelchair to replace the failed RTK to ensure that the navigation system continues to work.In order to test the feasibility of the algorithm,this article verifies the prediction accuracy of the model by setting the satellite signal loss time to 10s,30s and 60s,the experimental results in the three cases predicted the average position error of the model to be 0.084m,0.381m and 0.614m,respectively.The anti-interference ability and stability of the navigation system have been significantly improved.In the future,in-depth research can be carried out in the fusion navigation of sensors such as Lidar and dynamic obstacle avoidance.