Research On Fault Tolerant Multi-sensor Integrated Navigation Algorithm Based On Federated Kalman Filter

With the progress of current science and technology,navigation technology has also undergone rapid development.It is increasingly closely related to social life,integrating into all aspects of clothing,food,housing,and transportation.Although traditional navigation systems can achieve individual positioning,they all have certain shortcomings.Global Positioning System(GPS)is widely used in the field of outdoor positioning.However,with the development of cities,the complexity of road traffic conditions and severe building occlusion have led to a sharp decline in positioning accuracy,making it difficult to meet the positioning needs of modern cities.In order to make up for the shortcomings of GPS positioning alone,a method of integrated positioning with other navigation systems is adopted.It aims to give full play to the advantages of each system and complement each other’s advantages to improve positioning accuracy.Strap-down Inertial Navigation System(SINS)is an autonomous navigation system that is less affected by external factors and can compensate for the inherent shortcomings of GPS,but errors tend to accumulate over time.Arrival of angle(AOA)positioning based on 5G base stations further provides guarantee for fusion.5G base station positioning generally uses an AOA based positioning method.It utilizes the estimated arrival angles of transmitted signals from multiple base station array antennas to determine the location of the target object.Among them,the use of 5G large-scale multiple input multiple output(m MIMO)technology and Ultra Dense Network(UDN)technology also ensures the positioning accuracy of AOA.Due to the increase of sensors,the probability of integrated navigation system failure will also increase,resulting in abnormal positioning results.Therefore,it is necessary to add fault tolerant design to integrated navigation systems to detect and handle faults.The main research contents are as follows:1.Aiming at the problem of low accuracy of single GPS positioning in complex outdoor environments,a combined positioning scheme based on GPS/5G base station angle of arrival positioning/strapdown inertial navigation system(GPS/AOA/SINS)was proposed to increase the continuity and high accuracy of outdoor positioning.The system well integrates the advantages of GPS,AOA,and SINS.It uses a feedback type Federal Kalman Filter(FKF)for fusion.Finally,the integrated navigation and positioning system is simulated.Simulation experiments show that the integrated navigation system can effectively improve the navigation and positioning accuracy.2.Aiming at the problem that traditional fault processing methods are simple and cannot effectively suppress the divergence of positioning results,resulting in a decline in positioning accuracy.Based on the GPS/AOA/SINS integrated navigation and positioning scheme,an fault processing algorithm combined adaptive filtering with time update values of fault sub filters is proposed This algorithm improves the FKF in integrated navigation systems.In order to reduce the pollution of other sub filters caused by feedback,a fault detection and processing module is added between the FKF sub filter and the main filter.After fault detection,the fault processing process adaptively adjusts the filter gain matrix of the fault sub filter to change the degree of trust in the predicted and measured values,in order to achieve a reasonable use of useful information in the fault information.In order to simultaneously meet the requirements of small positioning error and fast convergence speed at abnormal locations,after the output of the main filter,the time update value of the fault sub filter was further fused.The experimental results show that the improved algorithm can process system faults in real time after fault detected,reduce positioning anomalies caused by faults,and improve the overall positioning accuracy and reliability of the integrated navigation system.3.Aiming at the problem that the fault detection algorithm in the fault detection module of the improved GPS/AOA/SINS integrated navigation system cannot effectively detect different types of faults,in order to further improve the fault tolerance of the integrated navigation system,improvements are made to the fault detection algorithm.The classical residual chi-square detection method is effective in detecting hard varying faults,but it cannot effectively identify slow varying faults,and there are often cases of missed and erroneous detection.The sliding window averaging method based on innovation sequences has strong detection sensitivity for both hard and slow varying faults,but adding windows can cause hysteresis effects.Combining the advantages of both,a combined fault detection method is designed.It combines the residual chi-square detection method with the sliding window averaging method based on the innovation sequence,and designs a combined decision logic to reduce the probability of false detection and missed detection,enhancing the ability to identify slowly changing faults.Simulation experiments show that the combined fault detection method is more accurate in identifying the time of fault occurrence and has higher sensitivity in fault detection.Its application in integrated navigation systems also improves positioning accuracy.