Wind measurement using a multicopter

Abstract

A system (1) for determining a wind speed in the atmosphere, comprising a multicopter (3) and a computing unit (5), wherein the multicopter (3) has an attitude angle sensor unit designed to determine an orientation of the multicopter (3) relative to an earth-fixed coordinate system and to transmit it to the computing unit (5), wherein the computing unit (5) is designed to determine a vector of external forces based on the orientation and an acceleration determined on the multicopter (3), as well as the mass of the multicopter (3) known by specification or currently estimated, and an estimate of a thrust force of all propellers of the multicopter (3) as known variables of a dynamic model of the multicopter (3), and to determine a vector of the wind speed from the vector of external forces by means of a stored and pre-calibrated relationship based on the Rayleigh equation,wherein the dynamic model comprises a lift term with respect to a body-fixed vertical axis of the multicopter (3), wherein the computing unit (5) is designed to determine the respective value of the lift term from a stored and pre-calibrated dependence of the lift term on an external force acting perpendicular to the body-fixed vertical axis.

Description

[0001] The invention relates to a system for determining a wind speed in the atmosphere, as well as a method for determining a wind speed in the atmosphere.

[0002] The following information is generally available knowledge and does not refer to any specific state-of-the-art document: In the Earth's atmospheric boundary layer, i.e., up to 2 km above ground, cup anemometers are frequently used to measure wind. However, due to their design, these can only determine wind speed and direction, i.e., a two-dimensional wind vector—such anemometers cannot normally measure vertical wind speed. If a third component of the wind vector is desired, an ultrasonic anemometer can also be used, which can approximately determine the three-dimensional wind vector at a point in the atmosphere. For practical purposes, anemometers are usually mounted on masts, either on flat terrain or near the tops of masts, such as communication masts; weather stations on mountain peaks are also frequently equipped with anemometers.Although the stationary arrangement of such anemometers allows for point-based measurements, it does not yet provide information about the flow conditions at separate points in the atmospheric boundary layer. It is therefore desirable to be able to set the measurement points for determining the current wind vector at any point in the atmospheric boundary layer in order to better capture the flow conditions in this atmospheric layer.

[0003] In this context, US 2016 / 0364990 A1 relates to a navigation platform for the autonomous and non-autonomous navigation of an unmanned aircraft, comprising: an inertial measurement unit that provides inertial measurement data; a global satellite navigation system device that provides at least one of the position measurement data, velocity measurement data, or range-related measurements from receiver to satellite; at least one autopilot or manual control device that provides control data to the actuators of the unmanned aircraft; and a navigation system.

[0004] The article, entitled "Wind Estimation in the Lower Atmosphere Using Multirotor Aircraft," by Ross T. Palomaki, Nathan T. Rose, Michael van den Bossche, Thomas J. Sherman, and Stephan FJ De Wekker, DOI: https: / / doi.org / 10.1175 / JTECH-D-16-0177.1, Page(s): 1183-1191, further examines the effectiveness of estimating wind speed and direction using 1) a direct approach with an anemometer mounted on a hexacopter and 2) an indirect approach using attitude data from a quadcopter. Data is collected by the multirotor aircraft hovering 10 m above ground next to one or more anemometers.

[0005] Furthermore, the paper entitled "Wind field estimation through autonomous quadcopter avionics" by authors Xingyu Xiang; Zhonghai Wang; Zijian Mo; Genshe Chen; Khanh Pham; and Erik Blasch; from the 2016 IEEE / AIAA 35th Digital Avionics Systems Conference (DASC) addresses the calculation of a wind profile by applying algorithms that relate the aircraft attitude to the local wind speed and direction, while conserving the payload of external devices such as multi-hole ducts. Several existing wind field estimation algorithms are evaluated and compared with the proposed dynamic behavioral fusion of the Kalman filter, which also uses data from the wind sensors.

[0006] The publication "Thielicke, W., Hübert, W., Müller, U., Eggert, M., and Wilhelm, P.: Towards accurate and practical drone-based wind measurements with an ultrasonic anemometer, Atmospheric Measurement Techniques, 14, 1303-1318, https: / / doi.org / 10.5194 / amt-14-1303-2021, 2021" also describes how an anemometer can be attached to a multicopter to measure the local wind vector in the atmosphere by changing the location of the multicopter. However, this anemometer attached to the multicopter is strongly subject to the influence of the thrust of the multicopter itself. Furthermore, due to the additional drag and the additional mass of the attached anemometer, the multicopter is subject to a significantly reduced flight time due to increased power consumption.

[0007] It is therefore an object of the invention to improve, and in particular to simplify, wind measurement at a given location in the atmospheric boundary layer.

[0008] The invention is based on the features of the independent claims. Advantageous developments and refinements are the subject of the dependent claims.

[0009] A first aspect of the invention relates to a system for determining a wind speed in the atmosphere, comprising a multicopter and a computing unit, wherein the multicopter has an attitude angle sensor unit designed to determine an orientation of the multicopter relative to an earth-fixed coordinate system and transmit it to the computing unit, wherein the computing unit is designed to determine a vector of external forces based on the orientation and an acceleration determined on the multicopter as well as the mass of the multicopter known by default or currently estimated as well as an estimate of a thrust force of all propellers of the multicopter as known variables of a dynamic model of the multicopter, and to determine a vector of the wind speed from the vector of external forces by means of a stored and pre-calibrated relationship based on the Rayleigh equation,wherein the dynamic model comprises a lift term with respect to a body-fixed vertical axis of the multicopter, wherein the computing unit is designed to determine the respective value of the lift term from a stored and pre-calibrated dependence of the lift term on an external force acting perpendicular to the body-fixed vertical axis.

[0010] A multicopter is an unmanned aerial vehicle, typically in the 0.25-25 kg range, with at least two propellers aligned so that the multicopter can take off and land vertically using the thrust of these propellers. The most common multicopter configurations are those with four, six, or eight propellers, but configurations with three propellers, as well as additional wings to generate lift for forward flight, are also known in the state of the art.

[0011] The acceleration of the multicopter is preferably determined by an acceleration sensor unit of the multicopter. This acceleration sensor unit preferably has at least translational acceleration sensors to directly measure translational accelerations, typically in three mutually perpendicular spatial directions, the body-fixed x, y, and z directions. The measurements can, of course, be filtered to filter out unwanted signals in certain frequency bands, such as sensor noise. Alternatively, a determined speed of the multicopter can be derived over time to determine the current acceleration. Other methods for determining acceleration known in the prior art are applicable. The determined acceleration also includes gravity, preferably expressed as a gravity vector, so that an acceleration non-zero is measured even in stationary hovering flight.This measured acceleration extends along the vertical axis and only along this axis when all attitude angles describing the orientation of the multicopter with respect to the horizontal are zero, i.e. the multicopter is hovering in a horizontal plane without any geometrical orientation with respect to an earth-fixed coordinate system, which occurs, for example, in a symmetrically configured multicopter and in the absence of horizontal wind.

[0012] The mass of the multicopter is either known a priori and can thus be stored as a constant value, or it is determined on a current basis using a mass estimator. Such mass estimators are well known in the art and typically rely on the relationship between a thrust increment and an additional acceleration of the multicopter in order to analytically determine the mass of the multicopter.

[0013] The dynamic model of the multicopter is based on equations of motion that follow the principle of Newton's second law of coupling an inertial force, expressed as the product of mass and acceleration, with external forces such as gravity. Gyroscopic terms are generally also considered in the equations of motion to capture dynamic effects at non-zero rotation speeds. The equations of motion can be written as follows for the three spatial directions of a body-fixed coordinate system of the multicopter: m(x¨+qz˙−ry˙)=−mg(sin(θ))+Fx m(y¨+pz˙−rx˙)=−mg(cos(θ)sin(ϕ))+Fy m(z¨+py˙−qx˙)=mg(cos(θ)cos(ϕ))+Fz−T where ϕ and θ indicate the angles relative to the horizontal and are part of the Euler angles ϕ, θ, ψ, and T is the thrust of the propellers along the body-fixed vertical axis of the multicopter.

[0014] If the multicopter is in stationary hover, the gyroscopic terms can be neglected. This, and the external forces solved for, together with the introduced lift term F L for the third of the equations (which refers to the body-fixed vertical axis of the multicopter): Fx=mg(sin(θ))+mx¨ Fy=mg(cos(θ)sin(ϕ))+my¨ Fz=−mg(cos(θ)cos(ϕ))+mz¨+T+FL(Fx)

[0015] The buoyancy term F L is influenced by an external force F acting perpendicular to the body's vertical axis x dependent, that is, F L =F L (F x ) . This function is given by the stored and pre-calibrated dependence of the buoyancy term on the external force acting perpendicular to the body's vertical axis, in particular as a polynomial function of a given order and with stored coefficients, with F x as input variable and FL as the output variable. The force F x is typically the horizontal force acting on the multicopter in the body-fixed x-direction, whereby in this case the body-fixed x-direction is rotated in the main wind direction of the horizontal plane (relative to the earth). For this purpose, the multicopter rotates around the body-fixed vertical axis so that this body-fixed x-coordinate axis points in the horizontal main wind direction. Alternatively, the body-fixed coordinate system could be rotated accordingly relative to the multicopter, or if the body-fixed coordinate system is fixed to the multicopter, the wind component in the body-fixed x-direction and in the y-direction can be calculated as a force acting perpendicular to the body-fixed vertical axis using: Fx:=(Fx2+Fy2).

[0016] The Rayleigh equation gives, based on the dynamic pressure, calculated by ρ / 2·v 2, of a fluid flowing around a body (where ρ is the density of the fluid and v is the relative velocity of the fluid with respect to the body) together with a reference area A multiplied by a coefficient c, the resulting force F of the fluid on the body is: ρ / 2·v 2·c·A=F . The Rayleigh equation is therefore typically also referred to as the "drag equation" because it specifies the relationship between speed and force of the fluid flowing against a body, based on properties of the body that determine this resistance, such as the drag area A and geometric properties of the body collected in the coefficient c . The stored and pre-calibrated relationship based on the Rayleigh equation inverts the above equation in that a speed to be determined is determined from a known force and reflects the fact that deviations from the ideal Rayleigh equation can be observed in reality, and the actual relationship can deviate from this ideal Rayleigh equation.This is due in particular to the fact that the multicopter has complex and branched geometric structures, is an aerodynamically complex entity, and also deviates somewhat dynamically from a rigid body, i.e., exhibits certain eigenmodes. The computing unit therefore fundamentally takes into account the effect that occurs naturally and is described by the Rayleigh equation in order to infer the wind speed from the determined external forces. At the same time, however, the stored calibration takes into account a corresponding deviation from this ideal equation, so that very accurate estimates of the wind speed can be expected based on the calculated external force, since the pre-stored and pre-calibrated relationship based on the Rayleigh equation reflects actual deviations that can be expected for precisely this type of multicopter.The pre-stored and pre-calibrated relationship based on the Rayleigh equation can itself be based on the form of the rearranged Rayleigh equation and include additional parameters such as a calibrated coefficient and a calibrated exponent for the respective determined external force to account for deviations from the ideal relationship. This is shown below:

[0017] If the ideal Rayleigh equation is solved as described above in such a way that a speed to be determined is obtained from a known force, the equation results: v=(2F / (ρ⋅c⋅A))

[0018] If a calibration with a coefficient and a calibrated exponent for the determined external force is also taken into account in this equation, the equation is: v=k·F b , where k=m1⋅(2 / (ρ⋅c⋅A)) and b=m2·1 / 2, with m1 and m2 being the factors that calibrate the ideal terms. For practical reasons, k and b are not stored explicitly, but rather m1 and m2, in order to avoid unnecessary conversions when determining the wind speed. Using only the coefficient k and the exponent b advantageously allows the wind speed to be determined based on the respective external force F, even in a computing unit mounted on the multicopter and powered by its battery, with very little computational effort and thus energy consumption. The relationships described apply along one axis of a Euclidean coordinate system in particular. If the wind speed is determined along the three mutually perpendicular basis vectors of a Euclidean coordinate system, the relationship mentioned above is repeated for all three coordinate axes to obtain a three-dimensional wind vector.In vector notation this results in: . (uvw)=(kxkykz)⋅⋅(Fxb,xFyb,yFzb,z) where the vector (u, v, w) T which specifies the three velocity components for the three mutually perpendicular coordinate axes of a body-fixed coordinate system of the multicopter (the analogue to the above case, in which only one quantity v was designated as the velocity for a single direction considered), the vector (k x , k y , k z ) T the respective coefficients for the three mutually perpendicular coordinate axes, b x , b y , b z the exponents valid for the coordinate directions, and the vector (Fxb,x,Fyb,y,Fzb,z)T which specifies the respective external forces for the three mutually perpendicular coordinate axes in the body-fixed coordinate system of the multicopter. The operator' . ' indicates that the vector (kx , k y , k z ) T and the vector (Fxb,x,Fyb,y,Fzb,z)T the right side of the above equation are multiplied element by element, so that the column vector (u, v , w) T the left side of the equation.

[0019] Alternatively, the stored and pre-calibrated relationship based on the Rayleigh equation can be presented in a tabular data format (a so-called "look-up table") without explicitly including the aforementioned mathematical form of the Rayleigh equation. However, the effect described by the Rayleigh equation is also implicitly contained in the tabular data in this case, as are the calibrated deviations from this ideal relationship. Both alternatives, as well as other data formats that exhibit the described relationships, may be encompassed by the term "stored and pre-calibrated relationship based on the Rayleigh equation."

[0020] The stored and pre-calibrated dependence of the lift term on an external force acting perpendicular to the body-fixed vertical axis refers in particular to the dependence on the external force in the direction of the main flow in a horizontal plane, i.e. along the direction of the horizontal wind direction.

[0021] Preferably, a three-dimensional translational acceleration vector is determined, the two relevant of the three Euler angles (the rotation around the vertical axis, also called yaw angle, is not important here) are determined to describe the orientation of the multicopter, and a three-dimensional wind speed is determined on the basis of this data. According to the procedure described above and below, however, a vector of wind speeds, preferably a three-dimensional vector, is first obtained in the body-fixed coordinate system of the multicopter. This vector of the determined wind speeds is preferably transformed from the body-fixed coordinate system of the multicopter into an earth-fixed coordinate system by a coordinate system transformation, wherein two axes of the earth-fixed coordinate system preferably lie in a horizontal plane, i.e. in a local tangential plane of the earth's reference ellipsoid.

[0022] Wind measurements are preferably carried out in the Earth's atmospheric boundary layer. In principle, the wind measurement method according to the invention can also be used in space missions exploring other planets or other celestial bodies, for example, in the study of the atmosphere of Mars, Venus, etc. A multicopter is already being used in the atmosphere of Mars to explore the planet.

[0023] Preferably, one or more relatively small multicopters are used, especially with a mass of less than 5 kg, particularly preferably less than 2 kg. The smaller the mass of the multicopter, the lower its impact on the surrounding atmosphere, since any aircraft that uses aerodynamic lift, rather than aerostatic lift like an airship, to overcome gravity accelerates the surrounding air downwards through momentum conservation to generate its own aerodynamic lift, which compensates for gravity.

[0024] An advantageous effect of the invention is that the proposed system and method for wind measurement using a multicopter allows measurement at (approximately) a selected point in the atmospheric boundary layer. This utilizes the ability of a multicopter to hover at a predetermined location in the atmosphere for a certain period of time or to follow any predetermined spatial trajectory. Thus, unlike a fast-flying fixed-wing aircraft, limited sensor bandwidth is required to simply average measurements. Instead, the flexibility and ability to perform maneuvers at any speed, including hovering, allow the respective flight condition during wind measurement to be adapted, in particular to the properties of the sensors used, for optimal measurement results.

[0025] A further advantageous effect of the invention is that no additional sensors are required on the at least one multicopter to perform the up to three-dimensional wind measurement. The wind force is measured solely on the basis of the sensor data already used in the multicopter's avionics, such as the attitude angle (to indicate the orientation of the multicopter relative to an earth-fixed coordinate system). In particular, the calibrated data regarding the aforementioned relationships and dependencies allow the immediate use of the system and method according to the invention with the multicopter matching the calibrated data, without the need for further modifications to the multicopter.

[0026] According to an advantageous embodiment, each of the propellers of the multicopter is connected to an electric motor for driving the respective propeller and each of the electric motors is connected to a battery for supplying electrical power to the electric motors, wherein the estimation of the respective thrust force of the propellers of the multicopter is based on pulse position modulation of a respective electric motor and a battery voltage of the battery.

[0027] Pulse position modulation, abbreviated PPM, is an electrical variable for controlling an electric motor. Using this and the determined battery voltage of the multicopter, the current energy consumption for the respective electric motor can be determined and thus, using known relationships, the respective thrust at the respective propeller of the multicopter can be estimated. According to this embodiment, the thrust of each individual propeller is advantageously determined to determine the total thrust of all propellers, even if no speed sensor is available on a respective propeller, nor is a torque sensor or even a (hypothetical) thrust sensor itself. The estimation of the respective thrust of the multicopter's propellers is preferably carried out in the computing unit, but can also be carried out in a separate estimation unit, particularly in one of the multicopter.

[0028] According to a further advantageous embodiment, the computing unit is arranged in a ground station, wherein the multicopter has a communication unit in order to transmit at least one value of the orientation and at least one value of the acceleration determined on the multicopter and at least one value of the estimate of the thrust of the propellers of the multicopter or data for this estimate to the ground station.

[0029] The multicopter's sensor data can be transmitted online, i.e., virtually in real time, from the multicopter to the ground station, allowing wind speed to be determined directly at the ground station. Alternatively, the data set of sensor signals is stored in a storage unit on the multicopter, for example, on an SD card, for offline use after the multicopter has landed at the ground station.

[0030] According to a further advantageous embodiment, the computing unit is designed to command a flight control unit of the multicopter for determining the wind speed of the atmosphere in such a way that the multicopter is kept in a stationary hover while obtaining the acceleration measurements and the attitude angle for determining the orientation for determining the wind speed.

[0031] According to a further advantageous embodiment, the stored and pre-calibrated dependence of the lift term on the external force acting perpendicular to the body-fixed vertical axis comprises different calibrations for a flow direction of the multicopter along the body-fixed vertical axis of the multicopter from bottom to top and for a flow direction from top to bottom.

[0032] According to this embodiment, the above formula (uvw)=(kxkykz)⋅⋅(Fxb,xFyb,yFzb,z) to calculate a three-dimensional wind speed vector from the determined external forces, its third component is modified so that: w={kz,u⋅Fzb,u, if Fz≥0kz,d⋅Fzb,d, if Fz<0} where k z,u is a factor for the upward flow (the body-fixed vertical axis is defined positively upwards), and k z,d is the corresponding factor for the downward flow. The exponents b u for an upward flow and b d for the air flow through the multicopter downwards are also calibrated separately.

[0033] According to a further advantageous embodiment, the stored and pre-calibrated relationship based on the Rayleigh equation indicates the relationship between a determined external force along an axis and a wind speed in the same axis by multiplying the determined external force by a pre-calibrated factor and raising the external force of this axis to the power of a pre-calibrated exponent.

[0034] This embodiment corresponds to the case of a modified Rayleigh equation v=k·F shown above b , where k=m1⋅(2 / (ρ⋅c⋅A)) and b=m2·1 / 2, with m1 and m2 being the factors calibrating the ideal terms of the Rayleigh equation.

[0035] According to a further advantageous embodiment, the computing unit is designed to determine a vector of wind speeds with three components in three mutually perpendicular spatial directions on the basis of the orientation and a spatial translational acceleration determined on the multicopter.

[0036] Measuring the three-dimensional wind vector advantageously enables the optimal recording of turbulent flows in the atmospheric boundary layer, which arise in particular from convection caused by the heat flow from the Earth's surface, shear winds, and heat exchange with overlying layers of the atmosphere. If a large number of multicopters are advantageously used at a large number of different measuring points, a complex flow pattern can be generated in a certain local area of the atmospheric boundary layer. This can be used, for example, to improve modeling of the atmosphere in this area, for use in weather models, to improve weather forecasts, and to plan flight operations and other activities. In addition to the wind measurement, it is advantageous to also perform a temperature measurement and / or an air pressure measurement and / or an air humidity measurement on each of the multicopters.

[0037] According to a further advantageous embodiment, the stored and pre-calibrated dependence of the buoyancy term on the external force acting perpendicular to the body-fixed vertical axis is expressed as a polynomial function, wherein the value of the external force acting perpendicular to the body-fixed vertical axis is the input value of the polynomial function and the value of the buoyancy term is the function value of the polynomial function.

[0038] According to a further advantageous embodiment, the system comprises a plurality of multicopters, for each of which a separate wind measurement is carried out, wherein a determination unit is designed to combine all wind measurements of the multicopters into a flow image of an atmospheric region.

[0039] Advantageously, wind measurements are performed simultaneously with several multicopters at different locations, so that the system comprises a plurality of the single multicopter described above. In this case, the system preferably comprises an additional detection unit that combines all the detected wind measurements based on the sensor data from each of the multicopters to obtain not only the wind vector at a single point, but also a complex flow pattern across a local area of the atmospheric boundary layer. For example, 5, 10, but also up to 50 or 100 or more multicopters can be used.

[0040] A further aspect of the invention relates to a method for determining a wind speed in the atmosphere, comprising the steps: - Determining the orientation of a multicopter relative to an earth-fixed coordinate system by means of an attitude angle sensor unit of the multicopter and transmitting the orientation to a computing unit, - Determination of a vector of external forces by the computing unit based on the orientation and an acceleration determined on the multicopter as well as the mass of the multicopter known by default or currently estimated as well as an estimate of the thrust of all propellers of the multicopter as known quantities of a dynamic model of the multicopter, wherein the dynamic model includes a lift term in relation to a body-fixed vertical axis of the multicopter, wherein the computing unit determines the respective value of the lift term from a stored and pre-calibrated dependence of the lift term on an external force acting perpendicular to the body-fixed vertical axis, and - Determination of a vector of the wind speed by the computing unit from the vector of external forces using a stored and pre-calibrated relationship based on the Rayleigh equation.

[0041] Advantages and preferred developments of the proposed method result from an analogous and analogous transfer of the statements made above in connection with the proposed system.

[0042] Further advantages, features, and details will become apparent from the following description, which – where appropriate with reference to the drawings – describes at least one embodiment in detail. Identical, similar, and / or functionally equivalent parts are provided with the same reference numerals.

[0043] They show: Fig. 1: A system for determining a wind speed in the atmosphere according to an embodiment of the invention. Fig. 2: A corresponding procedure to the system according to Fig. 1 within the same embodiment.

[0044] The representations in the figures are schematic and not to scale.

[0045] Fig. 1 shows a system 1 for determining a wind speed in the atmosphere. The method of Fig. Figure 2 shows the necessary steps for determining a wind speed in the atmosphere and is carried out on system 1 as shown in the Fig. 1 shown, which is why the following description describes the process steps according to Fig. 2, but refers to the system components of System 1 from the Fig. 1. To understand the following description of the figures, both the drawings of the Fig. 1 as well as the Fig.2 can be used equally. The system 1 has a multicopter 3 with four equally aligned propellers and a computing unit 5. In a first step, an orientation S1 of a multicopter 3 relative to an earth-fixed coordinate system is determined using an attitude angle sensor unit of the multicopter 3. The orientation thus determined is transmitted via the attitude angles: pitch angle θ and roll angle ϕ to a computing unit 5 of the system 1, which is arranged in the multicopter 3 itself. In the subsequent second step, a vector of external forces F is determined S2. x , F y , F xby the computing unit 5 based on the orientation θ, ϕ and an acceleration ẍ, ÿ, z̈ determined on the multicopter 3 as well as the mass of the multicopter 3 known by default or currently estimated as well as an estimate of the thrust of all propellers of the multicopter 3 as known quantities of a dynamic model of the multicopter 3. This is done by calculating the following relationships for the three predefined body-fixed coordinate axes with the designations x, y, z of the multicopter 3, where the axis in the direction z describes a body-fixed vertical axis around which a yaw movement can take place: Fx=mg(sin(θ))+mx¨ Fy=mg(cos(θ)sin(ϕ))+my¨ Fz=−mg(cos(θ)cos(ϕ))+mz¨+T+FL(Fx)

[0046] To determine the atmospheric wind speed, the computing unit 5 commands a flight control unit of the multicopter 3 such that the multicopter 3 is maintained in a stationary hover while obtaining the acceleration measurements and the attitude angles to determine the orientation for the wind speed determination. Therefore, no gyroscopic terms are taken into account.

[0047] Thrust T is calculated as follows: The propellers are speed-controlled, meaning the thrust of each propeller depends on its current speed at a constant angle of attack of the propeller blades. The voltage at each motor can be determined from the currently known PPM (pulse-position-modulation) value of the motor control of each electric motor driving each propeller. The PPM signal correlates directly with the voltage applied to the motor and is theoretically directly proportional to the speed of the motor and thus of the propeller. The voltage U i on every engine is: Ui=Ubsi−sminsmax−smin with U b the battery voltage, s i the servo PPM command, and s min s max the minimum and maximum values. Assuming a linear relationship, the speed Ω of each of the motors can be determined: Ωi =l·U i , with I a factor, so that the respective thrust of a propeller can be calculated as: T i =C t ρAΩ i R 2 with C t a thrust coefficient, ρ the air density, A the propeller circle area and R the propeller radius. In particular, using a propeller test rig, the parameters l and C t be determined experimentally and stored, so that the computing unit 5 uses the input variables mentioned to calculate the current propeller thrust T i can be calculated. The propeller thrust T i is calculated individually for each propeller as explained above, so that the sum of the individual thrust forces T i the i propeller the total force acting on the multicopter 3 is known by thrust T.

[0048] This dynamic model also includes the buoyancy term F L (F x) in relation to the vertical axis of the multicopter 3, whereby the respective value of the lift term is calculated by the computing unit 5 from a stored and pre-calibrated dependence of the lift term on an external force F acting perpendicular to the vertical axis of the multicopter 3 x This function is stored in the form of a polynomial, which, if F is known, x an associated F L (F x ) can be calculated.

[0049] Finally, in the next step, the computing unit 5 determines a wind speed vector S3 from the vector of external forces using a stored and pre-calibrated relationship based on the Rayleigh equation. The stored and pre-calibrated relationship based on the Rayleigh equation specifies the relationship between a determined external force along an axis and a wind speed in the same axis by multiplying the determined external force by a pre-calibrated factor and raising the external force of this axis to a power by a pre-calibrated exponent: (uvw)=(kykykz)⋅(FxbFybFzb) where the vector (u, v, w) T which specifies the three velocity components for the three mutually perpendicular coordinate axes of a body-fixed coordinate system of the multicopter, the vector (k x , k y , k z ) Tthe respective coefficients for the three mutually perpendicular coordinate axes, b x , b y , b z the exponents valid for the coordinate directions, and the vector (Fxb,x,Fyb,y,Fzb,z)T which specifies the respective external forces for the three mutually perpendicular coordinate axes in the body-fixed coordinate system of the multicopter. The operator ' . ' indicates that the vector (k x , k y , k z ) T and the vector (Fxb,x,Fyb,y,Fzb,z)T the right side of the above equation are multiplied element by element, so that the column vector (u,v,w) Tthe left side of the equation. However, different relationships are used for the downward and upward flow. This directional differentiation of the flow is necessary because the ambient air flow, due to wind, has different effects on the force transmitted to the multicopter 3: w={ku⋅Fzb,u, if Fz≥0kd⋅Fzb,d, if Fz<0}

[0050] Since the wind speed thus obtained is in a vector (u,v,w) TWith respect to the body-fixed coordinate system of the multicopter 3, a coordinate system transformation was finally performed by the computing unit 5 to transform the vector into an earth-fixed coordinate system, in particular to be combined with further measurements from other multicopters for a complex flow pattern of the atmosphere in the area under investigation. If the wind measurement was not taken while the multicopter was hovering, the speed vector of the multicopter relative to an earth-fixed coordinate system must also be subtracted from the determined wind speed. The actual speed vector of the multicopter can be determined, for example, using the inertial sensors and / or satellite-based signals.

[0051] Although the invention has been illustrated and explained in detail by means of preferred embodiments, the invention is not limited by the disclosed examples, and other variations may be derived therefrom by those skilled in the art without departing from the scope of the invention. It is therefore clear that a multitude of possible variations exist. It is also clear that the embodiments mentioned by way of example are truly only examples and should not be construed as limiting the scope, possible applications, or configuration of the invention in any way.Rather, the preceding description and the description of the figures enable the person skilled in the art to implement the exemplary embodiments in concrete terms, whereby the person skilled in the art, with knowledge of the disclosed inventive concept, can make various changes, for example with regard to the function or the arrangement of individual elements mentioned in an exemplary embodiment, without departing from the scope of protection defined by the claims and their legal equivalents, such as further explanations in the description. List of reference symbols 1 system 3 multicopters 5 Computing unit S1 Determine S2 Determine S3 Determine

Claims

[1] System (1) for determining a wind speed in the atmosphere, comprising a multicopter (3) and a computing unit (5), wherein the multicopter (3) has an attitude angle sensor unit designed to determine an orientation of the multicopter (3) relative to an earth-fixed coordinate system and to transmit it to the computing unit (5), wherein the computing unit (5) is designed to determine a vector of external forces on the basis of the orientation and an acceleration determined on the multicopter (3) and the mass of the multicopter (3) known by default or currently estimated as well as an estimate of a thrust of all propellers of the multicopter (3) as known variables of a dynamic model of the multicopter (3), and to determine a vector of the wind speed from the vector of external forces by means of a stored and pre-calibrated relationship based on the Rayleigh equation,wherein the dynamic model comprises a lift term with respect to a body-fixed vertical axis of the multicopter (3), wherein the computing unit (5) is designed to determine the respective value of the lift term from a stored and pre-calibrated dependence of the lift term on an external force acting perpendicular to the body-fixed vertical axis. [2] System (1) according to claim 1, wherein each of the propellers of the multicopter (3) is connected to an electric motor for driving the respective propeller and each of the electric motors is connected to a battery for supplying electrical power to the electric motors, wherein the estimation of the respective thrust force of the propellers of the multicopter (3) is based on pulse position modulation of a respective electric motor and a battery voltage of the battery. [3] System (1) according to one of the preceding claims, wherein the computing unit (5) is arranged in a ground station, wherein the multicopter (3) has a communication unit for transmitting at least one value of the orientation and at least one value of the acceleration determined on the multicopter (3) and at least one value of the estimate of the thrust of the propellers of the multicopter (3) or data for this estimate to the ground station. [4] System (1) according to one of the preceding claims, wherein the computing unit (5) is designed to command a flight control unit of the multicopter (3) for determining the wind speed in the atmosphere such that the multicopter (3) is kept in a stationary hover while obtaining the acceleration measurements and the attitude angles for determining the orientation for determining the wind speed. [5] System (1) according to one of the preceding claims, wherein the stored and pre-calibrated dependence of the lift term on the external force acting perpendicular to the body-fixed vertical axis comprises different calibrations for a flow direction of the multicopter (3) along the body-fixed vertical axis of the multicopter (3) from bottom to top and for a flow direction from top to bottom. [6] System (1) according to one of the preceding claims, wherein the stored and pre-calibrated relationship based on the Rayleigh equation indicates the relationship between a determined external force along an axis and a wind speed in the same axis by multiplying the determined external force by a pre-calibrated factor and raising the external force of this axis to a power by a pre-calibrated exponent. [7] System (1) according to one of the preceding claims, wherein the computing unit (5) is designed to determine a vector of wind speeds with three components in three mutually perpendicular spatial directions on the basis of the orientation and a spatial translational acceleration determined on the multicopter (3). [8] System (1) according to one of the preceding claims, wherein the stored and pre-calibrated dependence of the buoyancy term on the external force acting perpendicular to the body-fixed vertical axis is expressed as a polynomial function, wherein the value of the external force acting perpendicular to the body-fixed vertical axis is the input value of the polynomial function and the value of the buoyancy term is the function value of the polynomial function. [9] System (1) according to one of the preceding claims, wherein the system (1) comprises a plurality of multicopters (3) for each of which a separate wind measurement is carried out, wherein a determination unit is designed to combine all wind measurements of the multicopters (3) to form a flow image of an atmospheric region. [10] A method for determining a wind speed in the atmosphere, comprising the steps of: - determining (S1) an orientation of a multicopter (3) relative to an earth-fixed coordinate system by means of an attitude angle sensor unit of the multicopter (3) and transmitting the orientation to a computing unit (5), - Determining (S2) a vector of external forces by the computing unit (5) on the basis of the orientation and an acceleration determined on the multicopter (3) and the mass of the multicopter (3) known by default or currently estimated, as well as an estimate of a thrust force of all propellers of the multicopter (3) as known variables of a dynamic model of the multicopter (3), wherein the dynamic model comprises a lift term in relation to a body-fixed vertical axis of the multicopter (3), wherein the computing unit (5) determines the respective value of the lift term from a stored and pre-calibrated dependence of the lift term on an external force acting perpendicular to the body-fixed vertical axis, and - Determining (S3) a vector of the wind speed by the computing unit (5) from the vector of external forces by means of a stored and pre-calibrated relationship based on the Rayleigh equation.

Images