METHOD FOR CONTROLLING A MULTI-ROTOR ROTARY-WING DRONE, WITH CROSS WIND AND ACCELEROMETER BIAS ESTIMATION AND COMPENSATION

• US 20150057844A1
• Filed: 03/27/2013
• Published: 02/26/2015
• Est. Priority Date: 03/30/2012
• Status: Active Grant

• Current Assignee: Parrot Drones
• Original Assignee: Parrot SA
• Inventors: Callou, Francois, Foinet, Gilles
• Granted Patent: US 9,488,978 B2
• Time in Patent Office: Days
• US Class Current: 701/3
• CPC Class Codes: B64U 10/10 Rotorcrafts
• CPC Class Codes: B64U 10/14 with four distinct rotor axes, e.g. quadcopters
• CPC Class Codes: B64U 2201/00 UAVs characterised by their flight controls
• CPC Class Codes: B64U 2201/10 autonomous, i.e. by navigating independently from ground or air stations, e.g. by using inertial navigation systems [INS]
• CPC Class Codes: B64U 2201/20 Remote controls
• CPC Class Codes: G05D 1/0011 associated with a remote control arrangement
• CPC Class Codes: G05D 1/0204 to counteract a sudden perturbation, e.g. cross-wind, gust

Abstract

The attitude and speed of the drone are controlled by angular commands applied to a control loop (120) for controlling the engines of the drone according to the pitch and roll axes. A dynamic model of the drone, including, in particular, a Kalman predictive filter, represents the horizontal speed components of the drone on the basis of the drone mass and drag coefficients, the Euler angles of the drone relative to an absolute terrestrial reference, and the rotation of same about a vertical axis. The acceleration of the drone along the three axes and the relative speed of same in relation to the ground are measured and applied to the model as to estimate (128) the horizontal speed components of the cross wind. This estimation can be used to generate corrective commands (126) that are combined with the angular commands applied to the control loop of the drone in terms of pitch and roll.

Claims

1. A method for piloting a rotary-wing drone (10) with multiple rotors (12) driven by respective motors (110) able to be controlled to pilot the drone in attitude and speed,

said method comprising:

the generation of angular set-points (θ

, φ

) and the application of these set-points to a control loop (120) of the drone motors, these set-points being adapted to control the attitude of the drone about pitch (22) and roll (24) axes,

the establishment of at least one dynamic model of the drone, describing horizontal speed components of the drone as a function of the drag coefficients and the mass of the drone, of the Euler angles characterizing the attitude of the drone with respect to an absolute terrestrial coordinate system, as well as of the rotational speed of the drone about a vertical yaw axis;

the measurement of the aerodynamic drag force of the drone, derived from a measurement of acceleration of the drone;

the measurement of the relative speed of the drone with respect to the ground; and

the application to said dynamic model of the drone, by a Kalman predictive filter, of said measurements of aerodynamic drag force and of speed relative to the ground, so as to produce an estimation of the horizontal speed components of the cross wind,

the method being characterized in that;

the Kalman predictive filter is a six-state filter, such states comprising;

two components of the speed of displacement of the drone relative to the ground, expressed in a coordinate system linked to the drone,

two components of the speed of wind relative to the ground, expressed in an absolute terrestrial coordinate system linked, and

two horizontal components of drone accelerometer bias.

2. The method of claim 1, wherein the establishment of at least one dynamic model of the drone comprises the establishment of two distinct models, with:

i) a dynamic model of the drone in flight, and ii) a dynamic model of the drone on the ground, these models being used selectively as a function of the state of the drone.

3. The method of claim 1, wherein the dynamic model of the drone in flight uses a measurement of acceleration of the drone as a measurement proportional to the speed of air in the coordinate system of the drone.

4. The method of claim 2, wherein the dynamic model of the drone in flight is of the type:

5. The method of claim 2, wherein the dynamic model of the drone on the ground is of the type:

6. The method of claim 1, further comprising a step of compensation for the effect of cross wind on the positioning and the displacements of the drone, by

generating corrective set-points, function of the estimated components of horizontal speed of the cross wind, and

combining these corrective set-points to the angular set-points of pitch and roll applied to the control loop of the drone motors.

7. The method of claim 6, wherein the corrective set-points are of the type:

8. The method of claim 6, wherein the compensation step further comprises the definition for the drone of an open-loop reference pitch.

9. The method of claim 6, wherein, when the drone is in user-controlled flight, the corrective set-points (θ

<sub>wind—

comp</sub>, φ

<sub>wind—

comp</sub>) are combined to the piloting set-points (θ

_pilote, φ

_pilote) applied by the user.

10. The method of claim 6, wherein, when the drone is in auto-piloted standstill flight, the corrective set-points (θ

<sub>wind—

comp</sub>, φ

<sub>wind—

comp</sub>) are combined to fixed-point stabilization set-points (θ

_p, φ

_p) produced in response to a measurement of the horizontal speed of the drone relative to the ground (V_D/SX, V_D/SY).