WO2010097329A1

WO2010097329A1 - Method for estimating the travelling speed of an aircraft

Info

Publication number: WO2010097329A1
Application number: PCT/EP2010/052065
Authority: WO
Inventors: Nicolas Guenard
Original assignee: Commissariat A L`Energie Atomique
Priority date: 2009-02-25
Filing date: 2010-02-18
Publication date: 2010-09-02
Also published as: FR2942536A1

Abstract

The invention relates to a method for determining a travelling speed of a flying vehicle, comprising a step of using an onboard computing means (10, 11) for: estimating a thrust (F1) applied to the vehicle by the thrust means (2,3) thereof; estimating an inertial force (F2), where m is the mass of the vehicle, (F3) is the acceleration of the vehicle and (F4) the acceleration of gravity, applied to the vehicle during the travel thereof; deriving an aerodynamic force (F5) exerted on the vehicle therefrom by Newton's formula (F6), and then estimating the modulus thereof; deriving the travelling speed of the vehicle therefrom using the formula (F7), where v is the desired speed and k a proportionality coefficient.

Description

Method for estimating the speed of movement of an aircraft

The present invention relates to the estimation of the speed of displacement of a flying vehicle by using on-board measuring means, and more particularly to the estimation of the speed of a drone by using accelerometers and an on-board microcontroller.

BACKGROUND OF THE INVENTION

In the field of flying vehicles, various means are known for estimating the flight speed of an aircraft.

The most conventional measurement is carried out by means of an anemometer or badin. However, such an instrument is not accurate for low speeds.

It is also possible to estimate the speed using a GPS receiver. However, it is currently envisaged to fly small drones into buildings, especially for inspection purposes.

In such a situation, the GPS signal does not arrive or hurt until the drone, and is therefore not exploitable. If such a drone is equipped with accelerometers, it is tempting to integrate the signal from these accelerometers to deduce the speed. However, it is well known that such a signal ends up drifting rather quickly. We can of course use more efficient accelerometers, at the expense of cost and mass.

It has also been envisaged to estimate the speed of the drone by exploiting images taken by an on-board camera. However, the image processing requires significant computing power and must therefore in practice be performed by a ground station. The transmission and processing times lead to a delay in the estimation of the speed. It is also possible to use laser or doppler rangefinders, which also pose the problem of their mass and their cost.

OBJECT OF THE INVENTION The object of the invention is to provide a simple and inexpensive means for estimating the speed of movement of an aircraft, not suffering from the abovementioned disadvantages.

BRIEF DESCRIPTION OF THE INVENTION In order to achieve this goal, a method is proposed for determining a traveling speed of a flying vehicle, comprising the step of implementing calculation means for: estimating the module aerodynamic force F _aero exercising on the flying vehicle; deduce the velocity of the displacement vehicle by the relation F _aero = kv ² , where v is the desired velocity, and k a coefficient of proportionality.

If there is no wind (as expected during an indoor flight), or if the wind is weak, we obtain an estimate of the speed of the vehicle flying from the ground.

According to a preferred embodiment, the estimation of the aerodynamic force module comprises the steps of: estimating a thrust force F _p experienced by the vehicle due to the action of its or its thrust means; estimate an inertial force F _m = m- (a-g), where m is the mass of the vehicle, a is the acceleration of the vehicle and g the acceleration of gravity; deduce the aerodynamic force by Newton's relation F _aero = F _m -F _p , then estimate its modulus.

These estimates can be made from very simple and inexpensive means. The orientation of drone compared to a fixed reference is known using accelerometers and gyrometers, which allows to correctly orient the forces.

The estimation of the inertial force can be carried out simply by measuring, thanks to the accelerometers, the acceleration experienced by the drone, and, knowing the mass of the drone entered into memory before the flight, it is easy to deduce the value of the force. inertial.

Thus, the accelerometers are used according to the invention as a force sensor and their signal is not integrated, which eliminates the problem of drift. In practice, if the acceleration undergone by the drone is very low, the inertial force is reduced to the weight of the drone F _m ≈-mg, which is simple to estimate, knowing the orientation of the drone.

Finally, the thrust force F _p can be estimated simply by measuring the power actually supplied to the engine (s) of the flying vehicle, by assuming a relationship between the power supplied and the thrust delivered. In practice, if the drone is equipped with electric motors, it will be sufficient to measure the current supplied to the engines to estimate the developed thrusts, or to deduce the thrust delivered by the simple measurement of the rotational speed of the engine, or the voltage applied to its terminals.

This speed measuring method finds an advantageous application in the case of a vehicle for which the lift is totally provided by the engine (s). In this case, the aerodynamic force is reduced to a drag which can, under normal flight conditions, be estimated using a constant coefficient, independent of the attitude of the vehicle, so that the estimation of the module aerodynamic force is significantly simplified. BRIEF DESCRIPTION OF THE FIGURES The invention will be better understood in the light of the description of the single figure which is a perspective view of a drone equipped with speed estimation means according to the invention. DETAILED DESCRIPTION OF THE FIGURES

With reference to FIG. 1, the invention is here described in relation with a drone 1 of a particular type comprising four powertrains 2, here electric motors each equipped with a rotor 3 without cyclic variation of pitch, and carried each at the end of an arm to deliver a substantially vertical thrust when the drone is in a stationary position, according to a horizontal attitude. The powertrains form thrust means participating both in the lift and the propulsion of the drone.

To raise the altitude from such a stationary position, it is sufficient to simultaneously increase the power of the four powertrains. To cause a longitudinal movement, it is sufficient to slightly incline the drone in the desired direction by temporarily increasing the power of one or two powertrains relative to the power delivered to other powertrains. Once the drone in an inclined attitude, it advances alone under the effect of the non-bearing component of the cumulative thrust of powertrains.

The object of the invention is to estimate the speed by implementing very simple on-board calculation means.

For this purpose, the drone is equipped with a microcontroller 10 adapted to implement according to the invention a small program for calculating the speed stored in a read-only memory 11. In addition, the drone comprises an inertial unit 12, comprising essentially accelerometers, capable of determining an acceleration experienced by the drone during its displacement.

The inertial unit allows at any time to determine a direction of the drone in space, relative to a fixed reference. In particular, the inertial unit 12 comprises gyrometers which make it possible to know the orientation of a unit vector n which locates a drone orientation in space with respect to a fixed reference R, as well as the angle Φ which marks the angular position of the drone around an axis directed along the vector n, which completely determines the position of the drone in one space. In addition, the inertial unit comprises accelerometers which make it possible to measure the acceleration experienced by the drone during its movements. Let the acceleration of the drone measured by the accelerometers of the inertial unit. In practice, the control unit simultaneously measures accelerations along three determined axes, for example those of the fixed reference R.

Knowing the mass of the drone, which can be memorized before the flight of the drone, it is therefore easy to determine the inertial force experienced by the drone

where g is the acceleration of the gravity which is of course oriented according to the local vertical. In practice, the three components of the inertial force are determined in the reference frame R.

The drone also comprises powertrain power management means 13, which, in response to driving signals from a ground station, distributes power from power generating means. Each powertrain 2. Knowing that the thrust developed by each of the powertrain 2 is directly related to the power it receives, it is then very simple to determine the thrust modulus of each of the powertrains. For example, it can be assumed that the thrust of a powertrain is directly related to the rotational speed of the engine associated so that the simple measurement of the rotational speed of the engine allows to know the thrust developed by the group power train.

Knowing the orientation of the thrust for each powertrain (here, the thrusts are all parallel to the unit vector n, one thus deduces the thrusts, respectively F ₁ , F ₂ , F ₃ , F ₄ , of each of the powertrains.

By vector sum of the thrusts of all the powertrains, thus determines the total thrust force F _p applied to the drone, whose three components are also expressed in the reference R.

According to Newton's equation, the aerodynamic force undergone by the drone by its displacement can be deduced by the following relation:

F ¹ aero = F ^λ m -F ^λ p ( ^\ 2 ^) 'whose values of the three components, in the reference R, are obtained by subtraction of the respective components of the inertial force and the thrust force in this same mark.

It is known that the aerodynamic force undergone by the drone is proportional to the speed v of movement of the drone squared, and to a proportionality coefficient k which depends in particular on the size of the drone. Thus, the speed is deduced from the simple calculation: _{V =} J _2Ξ2. L ₍₃₎

Where F _aero is the modulus of the aerodynamic force F _aero , that is to say the value [F _a ² _eroX + F _has ² _eroY + F _has ² _eroZ f ² , where F _aero, x _> F _aero, γ _> F _aero, z are the components of the aerodynamic force in the fixed reference frame R. It is therefore sufficient to know the value of the coefficient of proportionality k, which can be entered into memory before the flight starts. In the simplest model, this one can be regarded as constant and can thus be also in memory. In the present case, the aerodynamic force is essentially a drag force, for which the coefficient k is only slightly dependent on the angle of incidence for attitudes and normal flight incidences. So we do not make a big mistake considering the coefficient k as constant.

However, in a more sophisticated model, the coefficient k may depend on the angle of incidence of the drone. This dependence can be taken into account by programming a simple relation, or by means of a table of values linking the coefficient k to the angle of incidence. It will then be necessary to equip the drone with an incidence sensor, or to deduce the incidence from the data measured by the inertial unit. Thus, the speed can be deduced very simply, with very small calculations, from an acceleration measurement and an indication of the thrust. This estimate of speed does not use any specific speed sensor. Preferably, the speed thus estimated is not used directly, but is filtered. Preferably, a registration and prediction filter is implemented, as explained below. If v is the measured speed according to the invention, a filtered speed v _wire is determined in the following manner. In a first step, the filtered speed is recalibrated with respect to its value at the preceding step taking into account the estimated speed: v _wire (0 = v _wire (i - 1)) + K (v _is (i - 1) - v _wire (i - 1)) (4)

Where K is a weight gain of the registration. In a second step, the filtered speed is predicted in the next step as follows. We start by estimating an aerodynamic force at the next step by using the value of the filtered speed recaled at step i:

F _aero {i) = kv _fl ² _l (i) (5)

Then, knowing the value of the thrust force F _p \ i), measured as explained above, we deduce an estimated acceleration:

Alternatively, it will also be possible to use the acceleration measured by the accelerometers of the inertial unit. The filtered speed at the next step will then be predicted by adding to the current value of the filtered speed an increase calculated from the acceleration thus estimated, in this case: v _wire {i + l) = v _wire {i) + Has â {i). (7) Where Δt is the sampling step. We then wait for the measurement of the speed step i + 1 to reset the speed thus estimated according to the relationship (4).

Tests have made it possible to verify that such a filtering gives a reliable speed signal without drifting. The invention is not limited to what has just been described, but encompasses any variant within the scope defined by the claims. In particular, although the invention has been described in relation to a flying vehicle whose lift effort is ensured by the powertrains, the invention can also be applied to a flying vehicle whose support force is provided partially or completely by one or more bearing surfaces. In this case, the aerodynamic force estimated according to the invention is the sum of a drag force and a lift force. The latter depends essentially on the incidence of the device, and it will be appropriate to measure this impact using a specific sensor. In addition, this determination of the speed according to the invention can be coupled with a vision device provided with a camera whose images are used to periodically readjust the speed thus determined.

Claims

1. Method for determining a traveling speed of a flying vehicle, comprising the step of implementing on-board calculation means (10, 11) for: estimating a thrust force F _p experienced by the vehicle due to the action of his or her thrust means (2,3); estimating an inertial force F _m = m- (ag), where m is the mass of the vehicle, a is the acceleration of the vehicle and g the acceleration of the gravity experienced by the vehicle during its displacement; deduce a aerodynamic force F _aero exerting on the vehicle by the relation of Newton F _aero = F _m -F _p , then to estimate its module; deduce the velocity of the displacement vehicle by the relation F _aero = kv ² , where v is the desired velocity, and k a coefficient of proportionality.

2. Method according to one of the preceding claims, further comprising the step of filtering the speed thus estimated to determine a filtered speed v _fll (i).

3. Method according to claim 3, wherein the filtering step comprises the operation of determining the value of the filtered speed at the following step v _β1 (i + 1) adding to the speed filtered at the current step an increase which depends on an acceleration (i) experienced by the vehicle at the current pace.

4. Method according to claim 4, wherein the acceleration is estimated according to the relation a (i) = - γng + FAi) + F {i) \, where FAi) is an estimate of the m thrust force at the current pitch , and F _aero (i) is an estimate of the aerodynamic force experienced by the flying vehicle using the value of the speed filtered at the current step:

The method of claim 1, wherein the acceleration is measured by means of accelerometers This measurement is used to estimate the inertial force.

6. The method of claim 5, wherein the accelerometers are part of an inertial unit.