Real-time VDM-based autonomous navigation for UAV

If you are a student interested with UAV navigation, this project might interest you!

Keywords: Kalman Filter, Navigation, Gazebo, QGroundControl

Introduction

Success of a drone mission is incumbent on accurate determination of its real-time position, velocity, and attitude, also known as navigation states altogether. For small Unmanned Aerial Vehicles, this is conventionally determined by a fusion of Inertial Navigation System (INS) and Global Navigation Satellite System (GNSS). Very recently, a Vehicle Dynamic Model based navigation system – VDMNav (in post-processing) has demonstrated an improvement in 1) attitude determination in normal flying condition and 2) in position in case of GNSS outage in comparison to traditional INS-based navigation.

Task Description:

The project focuses on the development and integration of the following three software modules in VDMNav architecture on the embedded computer of a fixed-wing UAV:

  • On field pre-flight (WP-1): Calibration phase; wherein deterministic errors of sensors are envisaged to be estimated on the field and compensated while flying. This will improve the real time accuracy of navigation solution. Major tasks:
    1. Adaptation of an existing software to estimate automatically, IMU bias and scale factorusing a 6 pose procedure.
    2. Design of new software to estimate barometer/Pitot’s tube bias/scale factor
    3. Analyze the impact of the compensation on navigation performance
  • In-flight (WP-2): All the existing EKF sensor fusion architectures have a fixed number of states. VDMNav is unique, in a way, that it can function with different number of states. A challenge is to adapt the number of states depending on the phases of the flight dynamically and in real-time. Major tasks:
    • Implementation of
  1. a state-of-the-art, probably the “first” of its kind,real-time dynamic state adaption architecture.
  2. Test and analyze the performance and limitationsof the dynamic state adaptation mechanism
  • (if time) Fully autonomous control (WP-3): The existing VDMnav solution is not used to update the UAV autopilot’s solution. In-order to scrutinize the autonomy of VDMNav, it is envisaged to test the fusion of VDMNav solution with PX4 during GNSS outage. It will be for the first time that such a system shall be put to use.
    1. Closed loop control of PX4 during a GNSS outage by using the VDMnav solution
    2. Field tests and demonstration