Research And Implementation Of Aircraft Navigation Technology Based On Multi-sensor Fusion

The aircraft has become one of the hot topics recently in many universities,research institutes,private enterprises,and which has many characteristics because of its special structure,such as vertical take-off and landing,hovering and flexible operation.It was equipped with high mapping equipment,thermal imager,high-definition camera,pesticide spraying device,widely used in city surveying and Mapping,UHV circuit inspection,Agricultural Plant Protection,Aerial Photography,etc..With the increasing complexity of aircraft application environment,The higher requirements for the capability of the aircraft are put forward.Navigation technology is the key technology for autonomous flight.At present,a relatively mature,reliable technology scheme of navigation is integrated navigation system,which is composed of MEMS IMU,magnetometer,GPS,ultrasonic sensor and barometer.However,as has been limited by manufacturing cost and the volume of circuit board,most of the onboard sensors of navigation system use low accuracy sensors,and the accuracy of navigation system depends on GPS sensor usually.Therefore,in the low-altitude flying environment with trees or buildings,the GPS sensor's error will increase rapidly because of weak signal,and then which will lead the navigation system failure.Tthis paper use a self-made quadrotor to study the navigation technology based on multi-sensor fusion in low-altitude flight environment.The main content of this paper is as follows:First of all,analyzing the structure and data fusion methods of multi-sensor data fusion,discussing each fusion method's practical advantages and disadvantages;Combined with the actual situation in this paper,multi-sensor fusion algorithm based on Kalman Filter is selected;Secondly,analyzing the characteristic of GPS signal in low-altitude flight environment is not stable,determined to add the optical flow sensor to the traditional integrated navigation system,which is composed of IMU,magnetometer,barometer,ultrasonic sensor and GPS sensor;proposed the multi sensor fusion framework by analyzing the role and measurement model of every sensor;Then,according to the multi-sensor fusion framework and the sensor measurement model,proposing the algorithm model of multi-sensor data fusion.From the perspective of software programming,drawing the flowchart of sensors data processing and fusion algorithm,and realize algorithm by programming.Finally,using quadrotor to verify the multi-sensor fusion algorithm described in this paper.The effectiveness of the algorithm is verified by the following experiments: the location experiment of GPS free;the robustness experiment of the algorithm under the condition of sensor failure;the autonomous navigation flight test in low-altitude environment with trees and buildings group.The final test results show that,the proposed algorithm based on multi-sensor fusion in this paper can accurately calculate the current position when the quadrotor in low altitude flight environment,and complete the navigation mission.The research contents in this paper have certain reference value for the research of aircraft navigation technology in low altitude environment.The multi-sensor fusion navigation algorithm is applied to an aircraft product.