Combined navigation method and device for unmanned aerial vehicle based on expanded auto-adaptive interval Kalman

Abstract

The invention discloses a combined navigation method and device for an unmanned aerial vehicle based on expanded auto-adaptive interval Kalman. The method comprises the following steps: measuring an inertial navigation parameter of the unmanned aerial vehicle by using a micro electro mechanical system inertial measurement unit and measuring a visual navigation parameter of the unmanned aerial vehicle by a visual navigation system; and by taking a digital signal processing module as a navigation computer, obtaining the optimum estimation of the navigation parameter of the system by taking an expanded auto-adaptive interval Kalman filter technology as a combined method fused with inertial and visual navigation parameters. The device disclosed by the invention is easy to realize, low in cost, relatively small in mass, relatively small in power consumption and suitable for being popularized in the civil field. A linear system model can be obtained by means of the expanded auto-adaptive interval Kalman filter technology. A time-varying parameter of the system is modeled as an auto-adaptive interval model, so that the capability of the navigation system operating the navigation positioning task during temperature variation, frequent gesture variation of a transporter and unsatisfactory lighting condition or in complex and severe conditions that a visual sensor is hard to fix the focus is improved.

Description

technical field [0001] The invention relates to an inertial / visual / ultrasonic integrated navigation method and device based on extended adaptive interval Kalman for unmanned aerial vehicles. Background technique [0002] With the development trend of miniaturization, intelligence, autonomy and low cost of the navigation system, and under the background that the importance of the autonomy of the navigation system is becoming more and more prominent, the inertial / visual integrated navigation technology, with its good complementary advantages and autonomous characteristics, has been It has become an important development direction in the field of navigation. In tasks such as fixed-point hovering, autonomous flight and landing of drones, target recognition and tracking of robots, and environmental detection, inertial / visual navigation technology has shown many advantages such as being able to interact with the external environment, good dynamic performance, and high real-time pe...

AI Technical Summary

Problems solved by technology

However, when the system parameters or noise statistical characteristics are time-varying or not known exactly, the extended Kalman filter cannot do anything about it.

Images