• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    The invention relates to an automatic landing system of an aircraft on an airstrip, comprising an on-board imaging system (110, 510), image analysis means (120, 520) for detecting in an image a landing runway and determining in this image the characteristics of a segment [P, Ω] connecting a predicted point of impact P on said track and a vanishing point Ω of said image, measurement means (130, 530) for measuring a plurality of observables in an inertial reference linked to the track, from the characteristics of the segment [P, Ω], said plurality of observables comprising a first observable defined by the relative heading angle (Ψ) of the aircraft with respect to a median axis of the runway, a second observable defined by the ratio of a lateral deviation of the position of the aircraft relative to said point of impact and a third observable defined by the ratio of a longitudinal deviation on a vertical ratio of the p of the aircraft with respect to said point of impact, estimating means (140, 540) for estimating longitudinal (ΔX), lateral (ΔY) and vertical (ΔH) positional deviations of the position of the aircraft relative to said point of impact, expressed in the inertial coordinate system, from the measurements of the first, second and third observables, guide means (150, 550) for calculating aircraft guidance orders from the deviations of longitudinal, lateral and vertical position thus estimated, and said relative heading angle.
    公开号:FR3024127A1
    申请号:FR1457208
    申请日:2014-07-25
    公开日:2016-01-29
    发明作者:Josep Boada-Bauxell；Victor Gibert；Laurent Burlion；Abdelhamid Chriette；Franck Plestan
    申请人:Airbus Operations SAS；
    IPC主号:
    专利说明:

    [0001] TECHNICAL FIELD The present invention relates generally to the field of aeronautics and more particularly to that of automatic landing systems. STATE OF THE PRIOR ART The landing phase is a critical phase in the flight of an aircraft because it is during this period that most incidents and accidents occur. For the pilots' convenience, many automatic landing systems have been proposed. In particular, the instrument landing system (ILS) for high-capacity airliners with ground-based infrastructures at international airports is known. The Microwave Landing System (MLS) microwave landing system is also available in some European airports. These systems use radio or microwave signals to guide the aircraft in its approach phase both laterally, with respect to the axis of the runway and vertically with respect to the approach plane (gilde path). However, they have the disadvantage of not being available at all airports because of their high costs and high use constraints. In addition, the ground infrastructure has a probability of failure of the order of 10-3 / hour. Aircraft can not be certain of being able to make a landing in automatic mode. More recently appeared automatic landing systems using satellite positioning, called Ground Based Augmentation System (GLS) or SLS (Satellite-based Landing System). The current satellite positioning systems do not provide the precision required to make a landing, they must be increased by ground reference stations such as Wide Area Augmentation System (WAAS) or EGNOS (European Geostationary). Navigation Overlay Service). These systems are still little widespread and have a relatively low availability rate (of the order of 99.5%) which does not, again, ensure a permanent landing in automatic mode. Due to the spatial and temporal unavailability of such ground infrastructures, the interest has been in autonomous landing systems, using video images of the ground taken by an on-board camera. These autonomous systems generally use the video images taken by the on-board camera and track information to estimate the attitude and position of the aircraft relative to the runway. The guidance orders of the aircraft are then calculated from the position and attitude thus determined. However, this estimation is a complex operation, which generally requires having a digital model embarked on the airport topography or at least geometric and topographical information relating to the runway. In addition, when the database containing the digital model is unavailable or in case of emergency landing on any terrain, the aforementioned autonomous systems are inoperative. The application FR-A-2835314 filed by the present applicant proposes a landing aid system that does not require knowing the characteristics of the runway. However, this system only works when the aircraft is a few meters above the ground, according to a principle similar to a driving aid in the context of road traffic. In particular, it does not make it possible to guide the aircraft throughout the final approach, that is to say during the last ten kilometers or so. The object of the present invention is to propose an automatic landing system of an aircraft which is particularly robust, can operate in total or partial absence of information on the airstrip, and guide it autonomously while along the final approach. PRESENTATION OF THE INVENTION The present invention relates to an automatic landing system for an aircraft on an airstrip comprising: an onboard imaging system on board said aircraft and intended to take a series of successive images of the soil; image analysis means for detecting in an image a landing runway and determining in this image the characteristics of a segment [P, n] connecting a predicted point of impact P on said runway and a point of leakage of said image. The automatic landing system is advantageous in that it comprises: measurement means for measuring a plurality of observables in an inertial frame linked to the track, from the characteristics of the segment [P, S2], said plurality of observables comprising a first observable defined by the relative heading angle (vf) of the aircraft with respect to a median axis of the track, a second observable (-AY) defined by AH the ratio of a lateral deviation the position of the aircraft with respect to said point of impact and a third observable (-) defined by the ratio of a longitudinal deviation on a vertical ratio AH of the position of the aircraft with respect to said point of impact; estimating means for estimating longitudinal position (AX), lateral (AY) and vertical (AH) positional deviations of the position of the aircraft relative to said point of impact, expressed in the inertial reference frame, from first, second and third observable measures; guidance means for calculating aircraft guidance orders from the estimated longitudinal, lateral and vertical position deviations, and said relative heading angle. The particular choice of observables makes it possible to estimate in a particularly simple manner a state vector constituted by the longitudinal, lateral and vertical positional deviations of the aircraft from the point of impact, expressed in the inertial reference frame. These differences can be exploited directly by the guide means.
    [0002] Advantageously, the characteristics of the segment [P, n] in the image are the angle entre between the line (PS2) with the vertical of the image, the distance dy from the point P to a line of horizon Dh and the distance dF between the orthogonal projection, Ph, of the point P 3024127 4 on the horizon line and the vanishing point here. In a first embodiment, the imaging system comprises an on-board camera, and the relative heading angle yr is measured in the image by the measuring means by means of the relation: ((, d, where f is the focal length of the on-board camera, 0 is a roll angle and 0 is the attitude angle of the aircraft.
    [0003] In this same embodiment, the second observable can be measured in the image by the measuring means by means of the relation: AY (cos Osin sin 0 - sin Ocos) - (sin çasin sin 0+ cos Ocos) tan In the same way, the third observable can be measured in the image by the measuring means by means of the relation: i - ((sin vicos 2 0) AY + cos 0 sin 0 The longitudinal, lateral and vertical positional deviations of the position of the aircraft relative to said point of impact are advantageously estimated by the estimation means by means of a dynamic model. using as state equation ù (t) = v (t), where u (t) = (AX, AY, AH) T is a vector consisting of said longitudinal, lateral and vertical positional deviations at time t, v (t) is the velocity vector of the aircraft in an inertial frame, and as observation equation r (t) = g (u (t)) + n (t), where r (t) is the vector of the third and second observed g, and g is the function that at any vector 30 u (t) = (AX, AY, AH) T matches the vector g (u (t)) = (AX DY T n (t) is a BAH AH) measurement noise vector. The guidance commands of the aircraft comprise, for example, a setpoint load factor and a target roll rate, the guide means calculating the setpoint load factor by the relation: Nz = K AZ + Kv, V, where AZ -AH, V = (A4 is the vertical velocity, and K,, Kv, are predetermined constants, and the set roll rate by the relation: Oc = Ky AY + Ky + K00 where KY, K and K0 are predetermined constants.
    [0004] The landing system may further comprise a flight control computer adapted to calculate commands to be applied to the actuators of the control surfaces of the aircraft to satisfy said guide commands. In a second embodiment, the image analysis means further determine the characteristics of at least one segment ([A, S2]) among the plurality of segments each connecting a vertex of the track to the vanishing point. here, the measuring means measure a plurality of additional observables in an inertial frame linked to the track from the characteristics of said at least one segment, said plurality of additional observables comprising an additional first observable defined by the relative heading angle (Y A) of the aircraft relative to a straight line passing through the top of the track and the leak point, an additional second observable (AYA) defined by the ratio of a lateral deviation of AZA the position of the aircraft relative to said upper and a third additional observable 3024127 6 ('A) defined by the ratio of a longitudinal deviation on a vertical ratio of the AZA position of the aircraft relative to said vertex, the estimation means estimating then longitudinal (AXA), lateral (AYA) and vertical (AZA) positional positional deviations from the position of the aircraft relative to said top of the track, expressed in the inertial frame, from the measurements of the first, second and third additional observables, the automatic landing system further comprising consolidation means for merging the estimates of the longitudinal, lateral and vertical position deviations with the estimates of said additional longitudinal, lateral and vertical position deviations with respect to said vertex of the runway and provide consolidated longitudinal, lateral and vertical position differences between the aircraft and the estimated point of impact. In the second embodiment, the guiding means calculates the guiding orders of the aircraft from the estimates of longitudinal, lateral and vertical position deviations after they have been merged with the additional longitudinal, lateral and vertical positional deviations. .
    [0005] The estimating means may then also provide from the estimates of the additional longitudinal, lateral and vertical positional deviations an estimate of the length and / or slope of the landing runway. Finally, the present invention further relates to a method of automatically landing an aircraft on a landing runway, comprising: - taking a series of successive images of the ground; an image analysis for detecting in an image the presence of an airstrip and determining in this image the characteristics of a segment [P, S2] connecting a predicted point of impact P on said track and a point leakage of said image; A measurement of a plurality of observables in an inertial frame linked to the track from the characteristics of the segment [P, n], said plurality of observables comprising a first observable defined by the relative heading angle (vf ) of the aircraft with respect to a median axis of the runway, a second observable (-AY) defined by the ratio of a lateral deviation AH 3024127 7 of the position of the aircraft with respect to said point of impact and a third observable AX (-) defined by the ratio of a longitudinal deviation on a vertical ratio of the position of AH the aircraft with respect to said point of impact; an estimate of the longitudinal (AX), lateral (AY) and vertical (AH) positional deviations of the position of the aircraft with respect to said point of impact, expressed in the inertial reference frame, from the measurements of the first, second and third observables; a calculation of the aircraft's guidance orders from the estimated longitudinal, lateral and vertical position deviations, and of the relative heading angle.
    [0006] BRIEF DESCRIPTION OF THE DRAWINGS Other features and advantages of the invention will appear on reading a preferred embodiment of the invention in relation to the following figures: FIG. 1 schematically shows an automatic landing system according to a first embodiment of the invention; FIG. 2 shows an image of the landing strip taken by the image capture device in connection with the first embodiment; Fig. 3 represents the relative position of the aircraft with respect to the runway in the inertial reference frame; Fig. 4 illustrates, in side view, the formation of an image of the landing strip 20 in the image plane of the on-board camera; Fig. 5 schematically shows an automatic landing system according to a second embodiment of the invention; Fig. 6 shows an image of the landing strip taken by the image capture device in connection with the second embodiment.
    [0007] DETAILED DESCRIPTION OF PARTICULAR EMBODIMENTS We will consider in the following an automatic landing system using the images taken by an onboard image capture system on board the aircraft. This image capture system includes one or more cameras. By automatic landing system 3024127 8, we mean in the following as well a system allowing the aircraft to land on the airstrip without the intervention of the pilot (Automatic Landing System) that a system assisting the pilot (Aided Landing System) during landing. In particular, the pilot can at any time correct the commands generated by the automatic landing system according to visual indications. The automatic landing system according to the invention makes it possible to guide or assist the pilot throughout the final approach. It can operate as soon as an airstrip has been identified (conventional landing on an airport runway), for example by means of image processing or designated by the pilot, especially in the event of landing. emergency on an improvised track. Fig. 1 schematically represents an automatic landing system according to a first embodiment of the invention. The automatic landing system, 100, comprises an image capture system, 110, for example an onboard camera, adapted to take successive images of the ground, mounted in the vertical plane of symmetry of the aircraft. This onboard camera can advantageously be a camera of an EVS (Enhanced Vision System) system fitted to most recently built aircraft. This system generally consists of a plurality of sensors (IR / visible cameras, millimetric radar, etc.) capable of providing the pilot with ground images in conditions of reduced visibility. Without loss of generality, we will also assume in the following that the axis of the camera is coincident with the longitudinal axis of the aircraft. In general, the axis of the camera is located in the plane of vertical symmetry of the aircraft and may have, with the aforementioned longitudinal axis, a known angular separation 9am, so that it is always possible to reduce to the previous hypothesis of zero angular difference. The intrinsic characteristics of the camera (especially its focal length) are assumed to be known. Image analysis means, 120, receive the successive images taken by the camera and process them to detect the presence of an airstrip. This detection can be carried out by means of the image processing method described in international application WO2014 / 075657. In the first step, the image processing method uses a horizontal differentiation filter to determine the edges of the track and in a second step, a line-by-line correlation of the image thus filtered with the symmetry of this image. (relative to its median axis).
    [0008] The image processing can be performed at the frame rate or even at a lower rate if necessary. If necessary, this treatment may further include Kalman filtering for tracking the landing runway, in a manner known per se. In all cases, the processing makes it possible to identify the edges of the landing runway, ie the straight lines (AB) and (CD) which delimit the track, 210, on either side of its median axis ( A) as shown in FIG. 2. The lines (AB) and (CD) intersect at a vanishing point, here, belonging to the horizon line Dh. In other words, the point here is the intersection of the straight line Dh and the median axis 0.
    [0009] In FIG. 2, the predicted point of impact of the aircraft on the landing strip (also referred to as the touch point) is also represented by P. This point can be determined by the image analysis means from the rectangular marks on the track defining the beginning of the touch zone, or can be manually designated on the image by the pilot (by means of an interface touch, for example).
    [0010] The image analysis means further determine the characteristics of the segment [P, n], that is to say a set of parameters making it possible to define this segment in the image. A preferred example of features of this segment is: the angle entre between the vertical axis and the line (SSP), the distance dY from the point P to the horizon line Dh, the distance dF between the orthogonal projection , Ph, of the point P on the horizon line and the vanishing point here. Those skilled in the art will understand that other features may alternatively be selected without departing from the scope of the present invention. For example, we can choose instead of Ç the angle between the line (S2P) and the horizon line Dh. Similarly, it will be possible to choose instead of the distances dy and dF, the distance S2P between the vanishing point and the point of predicted impact. The coordinates of the points P and ç in the image may also be chosen as characteristics of the segment. Optionally, the image analysis means can also determine the roll angle, 0, of the aircraft as well as its attitude, 0. Indeed, the roll angle θ can be determined as the angle between the horizon line Dh and the horizontal axis of the image 10 (see Fig. 2). Similarly, the attitude 0 of the aircraft (assuming Sam = 0) is defined by 0 = arctan where 00h is the distance from the center O of the image to the horizon right Dh and where f is the focal length of the camera (see Fig. 4). Whatever the variant envisaged, the characteristics of the segment [P, s2] are then supplied to a measurement module in the inertial reference frame, 130. Inertial reference mark will be called a local land reference linked to the landing runway, as schematically represented in FIG. 3. The inertial reference R, is defined by an origin 0, and axes, 0, X ,, 0 ,,,, Y and 0 Z, related to the track. Preferably, the axis O, X, is chosen to coincide with the longitudinal axis (or central axis) of the track, the axis OY is the transverse axis of the track and the axis 0, Z, is a vertical axis, pointing downwards, the R mark being chosen orthogonal direct. The peaks of the runway have been designated A, B, C, D, and the estimated impact point on the runway by P,. It will be understood that the points A, B, C, D, P of the image represent, in the image plane of the camera, the points A, _ ,,,,, B, _ ,,,,, C, _, ,,,, D, P of the track.
    [0011] It is also shown in FIG. 3 a reference Rc linked to the camera, defined by its origin O (optical center of the camera in a pinhole model), and its axes OcXc, OcYc 3024127 11 and 0, Z,. The axis 0, X, is the focal axis of the camera and points in the direction of observation. The axes 0, Y, and 0, Z, are respectively parallel to the horizontal axis and to the vertical axis of the image, the R mark, being chosen orthogonal direct. The reference R, is considered identical to the reference linked to the aircraft.
    [0012] Fig. 4 shows in side view the formation of an image of the landing strip in the image plane II of the camera. We have adopted a pinhole model for the camera and represented its optical center in O. It was also assumed that the aircraft for reasons of simplification the longitudinal axis was in the plane of symmetry of the runway (plane 10 of the figure). The vanishing point, n is defined by the intersection of the horizontal passing through O with the image plane II. Also shown in the figure is the inertial reference R ,,,,. Returning to FIG. 1, the measurement module in the inertial frame, 130, measures a plurality of observables in this frame from the characteristics of the segment [P, s2], namely: The first observable in the inertial frame is the angle of cap yr of the aircraft relative to the median axis of the runway. AX The second observable in the inertial reference frame is the ratio AH - where AX is the longitudinal deviation between the position of the aircraft and the position of the estimated point of impact and AH = -AZ where AZ is the vertical deviation between the position of the aircraft and the position of the estimated impact point. The third observable in the inertial reference frame is the -AY ratio where AY is the lateral AH difference between the position of the aircraft and the position of the estimated point of impact and where AH 25 is the vertical distance, previously defined. These three observables are expressed according to the characteristics of the segment [P, s2]. Indeed, if we write R the rotation matrix passing the inertial reference R, 3024127 12 to the reference of the camera: R = R014014, /, (1) where the matrix Ro is the roll matrix, Ro is the pitch matrix and R. is the heading matrix. We note 0 the roll angle and G the attitude of the aircraft. Yr is the relative heading angle between the aircraft and the runway. More precisely, this heading angle is that formed in the Onyny horizontal plane between the longitudinal axis of the track 0, X, and the longitudinal axis of the aircraft.
    [0013] The rotation matrix R can then be expressed in the following form: cos 0 0 0 0 cos 0 cos 0 cos 0 0 -sine 0 1 0 sine 0 cos 0 cos sin 0 -sin cos 0 0 0 1 R = (2) is again: cos cos 0 sin sine -sin 0 -cos Osin + sin Ocos sin 0 cos Ocos + sin Osin sin 0 sin Ocos 0 sin Osin + cos Ocos sin 0 cos Osin sin 0 - sin Ocos cos Oos 0 (3) The positional deviations AX, AY, AZ between the point of impact P and the aircraft, expressed in the inertial frame and the deviations AX, P, AyP, AZ: expressed in the reference of the The coordinates of the estimated impact point P in the image plane 25 of the camera are given by: R = = R An (4) AY are related by the vector relationship: AX P c AYl 'c AZ c AZ 3024127 13 xP = Yp zP f (5) Similarly, the positional deviations, AX, n, Ar.n, AZ, Û, between the vanishing point and the aircraft, expressed in the reference frame of the camera, are given by the vectorial relationship: AX Δ AYû AZn = R 1 (6) 0 0 given that the vanishing point is located at infinity on the axis 0, X ,.
    [0014] The coordinates of the vanishing point in the image plane of the camera are given by: f AYn f. men Azn f. The characteristics, dy, dF of the segment [P, n] in the image can be expressed by means of the coordinates of the predicted impact point and the vanishing point in the image plane: xn = Yn zn (7) YE2 - Yp tan; = dy = sin 0. (yp - yQ,) + cos 0. (zp - z / 2) (8-2) 5 d = Yn (8-3) F COS 0J By replacing in relations (8-1) , (8-2), (8-3), the expressions of the coordinates xp, yp, zp and xn, yn, z12 provided by the relations (4), (5) and (6), (7), are obtained finally: 10 = = arctan ((cos Ocos 0) AY + (cos Osin yisin 0 - sin Ocosyt) AH (9-1) (sin Ocos 0) AY + (sin 0 sin yi sin 0- cos Ocos yi) AH) -z12 + zP 14 (8-1) dy = (f AH (9-2) cos cos2 0) AX + (sin cos20) AY + (cos Osin 0) AH 15 dF = f (tan y / - + tan 0 tan 9 (9-3) cos 0 i AX AY The observables in the inertial reference, yr, - AH AH and - are therefore deduced from the characteristics of the segment [P, n] in the image, by means of the relations: (( d (10-1) 2 0 yi = arctan cos O. F-tan Otan 0 f)) AY (cos Osin sin 0-sin Ocos) - (sin sinin sin 0+ cos Ocos) tan; = AH tan; sin Ocos 0 -cos Ocos 0 (10-2) 3024127 15 AX = f (f 2 e) AY + 0 0 (10-3) - dY sin cos cos AH i AH cos yf cos2 0 In the calculation of the observables, the angles of roll and 0.0 pitch can be obtained by means of onboard sensors on the aircraft or be determined from the image as previously explained. In contrast to the relative heading angle, the differences AX, AY, AH can not be directly estimated from the characteristics of the segment If we denote by u (t) the vector defined by u (t) = (AX Where T expresses here the dependence of the vector as a function of time, v (t) = (vi v2 v3) T the velocity vector of the aircraft in the inertial frame as provided, for example, by the inertial units of the aircraft, we have the following relations: ti (t) = A (t) u (t) + v (t) r (t) = g (u (t)) + n (t) ( 11-2) where A (t) is a rotation matrix which is generally dependent on time, g is the function which at any vector u (t) = (AX, AY, AH) T matches the vector, r (t ) the vector of the third and second observables in the (AX Ay -.T g (u (t)) = AH AH inertial reference defined previously and n (t) is a random noise vector, assumed to be Gaussian. noise vector in relation (11-2) simply means that the measure of the observables is tainted by a daughter-in-law It is recognized in the relations (11-1) and (11-2) the classical formalism of a dynamic model including an evolution equation (11-1), where v (t) is the generating process, and an observation equation (11-2). The state vector is the vector u (t) and the vector of observables measured is the vector r (t).
    [0015] The judicious choice of the state vector (positional discrepancies between the estimated impact point and the aircraft expressed in the inertial frame) and that of the observables (measured directly from the characteristics of the segment [P, S2] ) leads to a particularly simple estimation of the aforementioned positional deviations.
    [0016] Indeed, the fact of using a state vector directly in the inertial coordinate system and not in the reference of the aircraft (animated with a rotational movement relative to the inertial reference) makes it possible to obtain an equation of evolution of particularly simple form, since in this case A (t) = O. Therefore, the dynamic model can be described by the relations ti (t) = v (t) (12-1) r (t) = g (u (t)) + n (t) (12-2) Returning to FIG. 1, the second and third observables measured by the measurement module 130, that is to say the components of the vector are supplied at each instant to an estimation module 140. This estimation module also receives the components of the speed vector of the aircraft, v (t), provided for example by the inertial unit of the aircraft. The module 140 performs at each instant t an estimate of the vector u (t) of the positional differences between the aircraft and the estimated point of impact. This estimate noted û (t) is obtained so as to minimize the measurement noise n (t). For example, the estimator described in the article by D. Karagiannis et al. entitled "A new solution to the problem of range identification in perspective vision systems" published in IEEE Trans. on Automatic Control, Vol. 50, No. 12, Dec. 2005, pp. 2074-2077, namely: (1 (13-1) A1-1) t + Att + t = the (t) + ft a (i) - di with 3024127 17 F '(t) = 2 [(- (AH -) t AH - (qt) v3 (t) + 2v1 (t) (AH) t + 2v2 (t) (AH / t1 (13-2) where A, is a predetermined constant governing the convergence of 5 the algorithm, and: V / 1 (0 - V3 (t) (-àX) 2AH t (1 f 1 afl afl [Kt) = -v3 (t) --TA, i) t - at a , / AY mi) t H2 (t) - V3 (t) A, H) ti (13-3), afl (a ar ar arfl afl`T AX AY 10 or - = - with Y1 = and r2 = AH AH 1 2 / In the above equations, the time index t (or t + At) indicates the moment at which the value is considered, note that equation (13-1) gives the update from the estimate of the inverse of the vertical difference, where At represents the periodicity of this update AY The estimate û (t) is then deduced from the vector r (t) - = (AX -AH '- AH) and the estimate of the vertical deviation, AH, other estimators may alternatively be used without departing from the scope of the present invention. Use the estimator described in Section 20 of O. Dahl et al. entitled "Linear design of a nonlinear observer for perspective systems", published in Proc. of the 2005 IEEE Conference on Robotics and Automation, Barcelona, Spain, 2005, pp. 429-435. The estimate û (t), ie the estimated deviations AX, AY, AH, as well as the measurement are then supplied to a guidance calculator 150. This calculator calculates the guiding orders so that the aircraft comes move in the approach plane (generally inclined 3% from the ground) and align along the longitudinal axis of the runway. In general, the set of guide commands defines a guide law that the aircraft must follow during the final approach. Guidance orders are expressed by the demand factor (loacl factor demand) and the roll rate demand. It is recalled that in aerodynamics the load factor is the ratio of the apparent weight to the actual weight and that the roll rate is the derivative of the roll angle. The set load factor is given by: Nz, = K, AZ + Kv, V, (14-1) where AZ-AH, V - (AZ) * is the vertical velocity, and K,, Kv, are predetermined constants. The set roll rate is given by: 0, = KyAY + Kyy ± K00 (14-2) where KY, K and K0 are predetermined constants and where the relative course angle relative to the track, yi, is measured from (10-1). The roll angle θ can be obtained from sensor measurements on board the aircraft or measured from the angle formed between the horizon line and the horizontal of the image, as explained above.
    [0017] The guidance orders are then transmitted to a flight control computer 160, which then determines the commands to be applied to the control surface actuators to satisfy the guidance commands. The autopilot system can notably use a C * law for the longitudinal control and a Y * law for the transversal control, in a manner known per se. Thus, the law C * controls the elevators by means of: DQ = Kiq + K2nz + K, f (nz-Nzjdt + 1 (41 1z, (15-1) 3024127 19 where DO is the control of the steering actuator , Nz, is the previously calculated target load factor, nz is the instantaneous load factor, q is the pitch speed and K4 are predetermined constants, Similarly, the Y * law controls the fins and fins, respectively by means of: DP = iç2p + iç3r + IÇ4ç 0+ iç5fi + iç60, (15-2) DR = k2i1 + 1 (22r + 1 (23V + k4 J + 1 (250, (15-3) 10 where DP and DFi are respectively the control of ailerons and drift, 'p is the roll angle, p is the roll speed (where p = r tan (0) with r yaw rate and G attitude angle), q is the speed of pitch, la is the slip angle and the coefficients ku are predetermined constants A detailed description of the above-mentioned control laws can be found in Farineau's article entitled "The The simplest form of the automatic landing system according to the invention does not include the control computer, which is described in Proceedings of the AIAA GNC Conference, 2004. In its simplest form, the automatic landing system according to the invention does not include the control computer. flight 160 (shown in broken lines), it may be simply the system CDVE (Electrical Flight Controls) existing on the aircraft. It is therefore easy to retrofit the aircraft without having to modify the CDVE system. Fig. 5 schematically shows an automatic landing system, 500, according to a second embodiment of the invention.
    [0018] This embodiment differs from the first in that, in addition to making an estimate of the positional differences between the aircraft and the estimated point of impact Prw, it is also possible to estimate all or part of the positional differences between the aircraft. and the various "corners" of the track (i.e. the vertices of the rectangle representing the track) and, more precisely: the positional deviations AXA, AYA, AZA between the aircraft and the proximal left vertex AT, ; positional deviations AXB, AYB, AZB between the aircraft and the distal left vertex, Bny; The positional deviations AXc, AYc, AZc between the aircraft and the right proximal vertex, C,; the positional deviations AXD, AYD, AZD between the aircraft and the right distal vertex, D 10. The automatic landing system according to the second embodiment of the invention comprises, as in the first embodiment, a system image capture system 510, identical to the system 110 previously described, image analysis means 520, a measurement module 530 adapted to measure observables in the inertial reference frame, an estimation module 540 for estimating , in addition to the positional deviations, AX, AY, AH, of the first embodiment, all or part of the positional deviations between the different peaks of the runway and the aircraft. All of the estimated differences in position are then provided to a consolidation module, 545 which merges the estimated position differences to obtain consolidated AX, AY, AH position differences between the estimated impact point and the aircraft. .
    [0019] Furthermore, the estimation module 540 can furthermore determine from the various estimated positional deviations, the length and / or the slope of the track, in order to adapt, for example, the braking power of the aircraft when it is in contact with the ground (runway length estimation) or to modify the flare (flora) of the trajectory (slope estimation).
    [0020] The image analysis means, 520, first perform a detection of the landing runway, as before. In addition to the characteristics of the segment [P, n], the image analysis means can also determine those of the segments [A, s2], [B, s2], [c, n], [D, s2]. It will be recalled that the characteristics of a segment are a set of parameters making it possible to define this segment in the image. Fig. 6 shows an image taken by the image system 510 and an example of the characteristics of the segment [A, n], namely: the angle ΔA between the vertical axis and the straight line (SCA), the distance dyA from point A to the horizon line Dh, the distance dFA between the orthogonal projection, Ah, of the point A on the horizon line and the vanishing point here.
    [0021] It will be understood that similar characteristics can be determined for the other vertices B, C, D of the track. AX AY The measurement module 530 measures not only the observables yr, AH AH as the module 130, but also the corresponding observables relating to the different peaks of the track, for example the observables A, AXA, AY A relating to the point A. Where AZA AZA 15 the angle Y1A is the relative heading of the aircraft from the right (AS2). The estimation module 540 again estimates not only the positional deviations AX, AY, AH as in the first embodiment, but also the positional deviations relating to the different vertices of the track, for example the positional deviations AXA, AYA, AZA. The estimation is obtained according to the same principle as that of relations (12-1), (12-2) and (13). Unlike the first embodiment, the second embodiment comprises a consolidation module 545 which merges the estimates of the differences AX, AY, AH with those of the deviations obtained for the different vertices of the track, for example the positional deviations. AXA, AYA, AZA relating to the point A, the deviations AXB, AYB, AZB relating to the point B, etc. The estimated differences in the different peaks may be consolidated with those for the estimated point of impact in order to obtain estimates of differences ............, _ consolidated. For example, the estimate AY of the lateral deviation of the position of the aircraft from the point of impact can be consolidated with the estimates of the lateral deviations AYA and Mc of the position of the aircraft with respect to the vertices A , and C: 5 Ay, Ay + AYA + Ayc 2 (16) _. similarly, the estimate AX of the longitudinal deviation of the position of the aircraft with respect to the point of impact can be consolidated with the estimates of longitudinal deviations AXA and AzYc: AX = AX + AXA 2 ± AXc + δ (17) where δ is the normalized distance from the estimated point of impact to threshold of the track defined by the segment [A ,,,, C,]. Other forms of consolidation may be envisaged by those skilled in the art without departing from the scope of the present invention. In any case, the consolidated estimates AX, AY, AH have greater robustness and better tracking stability than the initial estimates AX, AY, AH. In particular, this consolidation makes it possible to improve the observability of the initial estimates AX, AY, AH when the aircraft is aligned in the axis of the runway and stable in the approach plane.
    权利要求:
    Claims (12)
    [0001]
    REVENDICATIONS1. An automatic landing system for an aircraft on an airstrip characterized in that it comprises: an imaging system (110, 510) embarked on board said aircraft and intended to take a series of images successive soil; image analysis means (120, 520) for detecting an airstrip in an image and determining in this image the characteristics of a segment [P, S2] connecting a predicted point of impact P on said track and a vanishing point there of said image; measurement means (130, 530) for measuring a plurality of observables in an inertial frame linked to the track, from the characteristics of the segment [P, S2], said plurality of observables comprising a first observable defined by the relative heading angle (yi) of the aircraft with respect to a median axis of the runway, a second observable AH defined by the ratio of a lateral deviation on a vertical deviation of the position of the aircraft relative to said point impact and a third observable (-) defined by the ratio of an AH longitudinal deviation on said vertical deviation of the position of the aircraft with respect to said point of impact; estimating means (140, 540) for estimating longitudinal (AX), lateral (AY) and vertical (AH) positional deviations from the position of the aircraft relative to said point of impact, expressed in the reference inertial, from the measures of the first, second and third observables; guidance means (150, 550) for calculating aircraft guidance orders from the estimated longitudinal, lateral and vertical position deviations, and said relative heading angle.
    [0002]
    An automatic landing system according to claim 1, characterized in that the characteristics of the segment [P, n] in the image are the angle; between the straight line 3024127 24 (Pçà) with the vertical of the image, the distance dy from the point P to a horizon line Dh and the distance dF between the orthogonal projection, Ph, of the point P on the horizon line and the vanishing point here. 5
    [0003]
    Automatic landing system according to claim 2, characterized in that the imaging system comprises an on-board camera and that the relative heading angle yr is measured in the image by the measuring means (130). , 530) by means of the relation: (id d, 10 vu = arctan cos = - tan Otan 0 f) / where f is the focal length of the onboard camera, 0 is a roll angle and G is the angle of the aircraft. 15
    [0004]
    4. automatic landing system according to claim 3, characterized in that the second observable is measured in the image by the measuring means (130, 530) by means of the relation: AY (cos Osin sin 0 - sin Ocos ) - (sin çasin sin 0+ cos Ocos) tan = AH tan sin Ocos 0 - cos Ocos 0
    [0005]
    5. Automatic landing system according to claim 4, characterized in that the third observable is measured in the image by the measuring means (130, 530) by means of the relation: i - ((sin vicos2 0) AY ± cos 0 sin 0 AX dy AH i = AH cosy f cos2 6 20 25 3024127
    [0006]
    6. Automatic landing system according to claim 5, characterized in that the longitudinal, lateral and vertical positional deviations of the position of the aircraft with respect to said point of impact are estimated by the estimation means (140, 540) by means of a dynamic model using as equation of state ù (t) = v (t), where u (t) = (3./Y, A, Y, AH) t is a vector consisting of longitudinal, lateral and vertical positional deviations at time t, v (t) is the speed vector of the aircraft in an inertial reference, and as observation equation r (t) = g (u (t)) + n (t), where r (t) is the vector of the third and second observables, and g is the function that at any vector u (t) = (AX, AY, AH) T matches the vector g (u (t) ) = (AX Al 7 T n (t) is a BAH AH) 10 measurement noise vector.
    [0007]
    7. automatic landing system according to claim 1, characterized in that the guidance orders of the aircraft comprise a setpoint load factor and a target roll rate, the guide means (150, 550) calculating the setpoint factor 15 by the relation: Nz, = K AZ + Kv, 17, - where AZ -AH, V - (3,4 is the vertical velocity, and K,, Kv, are predetermined constants, and the set roll rate by the relation: Oc = Ky AY + Ky + K00 where KY, K and K0 are predetermined constants.
    [0008]
    8. Automatic landing system according to claim 1, characterized in that it further comprises a flight control computer (160, 560) adapted to calculate 3024127 26 commands to be applied to the actuators of the control surfaces of the aircraft for meet the said guidance orders.
    [0009]
    The automatic landing system of claim 1, characterized in that the image analysis means (520) further determines the characteristics of at least one segment ([A, n]) among the plurality of segments each connecting one vertex of the track to the vanishing point here, the measuring means (530) measure a plurality of additional observables in an inertial frame linked to the track from the characteristics of said at least one segment, said plurality of additional observables comprising an additional first observable defined by the relative heading angle (A) of the aircraft relative to a line passing through the top of the track and the vanishing point, an additional second observable (AYA) defined by the ratio of a lateral deviation of the position of AZA from the aircraft relative to said elevation and a third additional observable (`" 1A) defined AZA by the ratio of a longitudinal deviation on a vertical ratio of the position of the aeron ef with respect to said vertex, the estimating means (540) estimate additional longitudinal (AXA), lateral (AYA) and vertical (AZA) positional deviations of the position of the aircraft with respect to said top of the runway, expressed in the inertial frame, from the measurements of the first, second and third additional observables, the automatic landing system further comprising consolidation means (545) for merging the estimates of longitudinal, lateral and vertical position deviations with estimates of said additional longitudinal, lateral and vertical positional deviations from said top of the runway and providing consolidated longitudinal, lateral and vertical positional deviations between the aircraft and the estimated point of impact. 25
    [0010]
    10. automatic landing system according to claim 9, characterized in that the guiding means calculate the guidance orders of the aircraft from the estimates of longitudinal, lateral and vertical position deviations after they have been merged with the additional longitudinal, lateral and vertical positional deviations. 3024127 27
    [0011]
    11. automatic landing system according to claim 9, characterized in that the estimating means also provide from the estimations of the longitudinal, lateral and vertical positional additional deviations, an estimate of the length and / or the slope of the airstrip. 5
    [0012]
    12. A method of automatically landing an aircraft on an airstrip, characterized in that it comprises: - taking a series of successive images of the ground; an image analysis for detecting in an image the presence of a landing runway and determining in this image the characteristics of a segment [P, n] connecting a predicted point of impact P on said runway and a vanishing point here of said image; a measurement of a plurality of observables in an inertial reference linked to the track from the characteristics of the segment [P, s2], said plurality of observables comprising a first observable defined by the relative heading angle (yi ) of the aircraft with respect to a median axis of the runway, a second observable (-AY) defined by the ratio of a lateral deviation AH of the position of the aircraft relative to said point of impact and a third observable AX AH) defined by the ratio of a longitudinal deviation on a vertical ratio of the position of the aircraft with respect to said point of impact; An estimate of the longitudinal (AX), lateral (AY) and vertical (AH) positional deviations of the position of the aircraft relative to the said point of impact, expressed in the inertial reference, from the measurements of the first second and third observables; a calculation of the aircraft's guidance orders from the estimated longitudinal, lateral and vertical position deviations, and of the relative heading angle. 25
    类似技术:
    公开号 | 公开日 | 专利标题
    FR3024127A1|2016-01-29|AUTONOMOUS AUTOMATIC LANDING METHOD AND SYSTEM
    FR3009117A1|2015-01-30|AUTONOMOUS AUTOMATIC LANDING METHOD AND SYSTEM
    EP2718730B1|2015-11-18|Device for measuring speed and position of a vehicle moving along a guidance track, method and computer program product corresponding thereto
    US10384679B2|2019-08-20|Travel control method and travel control apparatus
    EP2372308B1|2016-05-11|Image processing system and vehicle control system
    FR3000813A1|2014-07-11|Rotary wing drone i.e. quadricopter, has image analysis unit implementing Kalman filter estimator having representation of dynamic model of drone, with input of horizontal speed, position, linear acceleration, rotation and altitude signals
    FR2911713A1|2008-07-25|DEVICE AND METHOD FOR MEASURING DYNAMIC PARAMETERS OF AN AIRCRAFT EXTENDING ON A AIRPORT AREA
    CN107728175A|2018-02-23|The automatic driving vehicle navigation and positioning accuracy antidote merged based on GNSS and VO
    FR2983619A1|2013-06-07|METHOD, DEVICE AND SYSTEM FOR GUARANTEEING TEMPORAL SPACING BETWEEN AN AIRCRAFT AND AT LEAST ONE TARGET TRAFFIC
    FR2987911A1|2013-09-13|METHOD OF CORRECTING A LATERAL TRACK IN APPROACH IN RELATION TO ENERGY TO BE RESORBED
    EP2597544B1|2016-05-11|Steering method for correcting the trajectory of an aircraft
    EP3451314A1|2019-03-06|Method and device for avoidance of an object by detection of its proximity to an aircraft
    US20200250854A1|2020-08-06|Information processing apparatus, information processing method, program, and moving body
    EP3388914A1|2018-10-17|Target tracking method performed by a drone, related computer program, electronic system and drone
    EP3657213B1|2022-03-09|Learning method of a neural network on-board an aircraft for landing assistance of said aircraft and server for implementing such a method
    WO2014146884A1|2014-09-25|Method for observing an area by means of a drone
    KR20210082422A|2021-07-05|Apparatus for determining position of vehicle and method thereof
    FR3033903A1|2016-09-23|NAVIGATION ASSISTANCE SYSTEM FOR AN AIRCRAFT WITH HIGH HEAD DISPLAY SCREEN AND CAMERA.
    WO2018041978A1|2018-03-08|Device for determining a speed limit, on-board system comprising such a device, and method for determining a speed limit
    US20210072041A1|2021-03-11|Sensor localization from external source data
    FR3073655B1|2019-10-18|METHOD FOR DETERMINING A VISUAL SPEED VECTOR OF A MOBILE MACHINE, COMPUTER PROGRAM PRODUCT, AND DETERMINING SYSTEM THEREOF
    US20220058969A1|2022-02-24|Systems and methods for determining an angle and a shortest distance between longitudinal axes of a travel way line and a vehicle
    US20220058966A1|2022-02-24|Systems and methods using image processing to determine at least one kinematic state of a vehicle
    KR102336071B1|2021-12-07|Apparatus for determining position of vehicle and method thereof
    EP3578926A1|2019-12-11|Method for securing the operation of a system for synthetic view of an aircraft, associated computer program product and system
    同族专利:
    公开号 | 公开日
    FR3024127B1|2016-08-26|
    CN105302146A|2016-02-03|
    CN105302146B|2020-12-04|
    US9939818B2|2018-04-10|
    US20160026189A1|2016-01-28|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    GB2302318A|1995-06-14|1997-01-15|Bodenseewerk Geraetetech|Aircraft landing procedure|
    US20050075761A1|2000-05-29|2005-04-07|Francois Kubica|Method for operating an aircraft|
    FR2835314A1|2002-01-25|2003-08-01|Airbus France|METHOD FOR GUIDING AN AIRCRAFT IN THE LANDING PHASE AND CORRESPONDING DEVICE|US11022984B2|2016-08-06|2021-06-01|SZ DJI Technology Co., Ltd.|Automatic terrain evaluation of landing surfaces, and associated systems and methods|
    FR3110985A1|2020-05-29|2021-12-03|Airbus|HUMAN-MACHINE INTERFACE OF AN AIRCRAFT IN THE TAKE-OFF OR LANDING PHASE|
    FR3113166A1|2020-07-30|2022-02-04|Airbus|METHOD AND SYSTEM FOR AUTOMATIC TAKE-OFF OF AN AIRCRAFT|FR2725691B1|1994-10-18|1997-01-03|Sextant Avionique|PILOT AID DEVICE, DURING LANDING, OF AN AIRCRAFT|
    FR2752051B1|1996-08-02|1998-10-30|Sextant Avionique|DEVICE FOR ASSISTING THE GUIDANCE OF A VEHICLE ON A TRAJECTORY|
    US6629023B1|2002-04-30|2003-09-30|Sikorsky Aircraft Corporation|Method for performing an automated category a approach maneuver|
    FR2852097B1|2003-03-07|2005-05-06|METHOD AND DEVICE FOR CONSTRUCTING AN ENVIRONMENTAL SYNTHESIS IMAGE OF AN AIRCRAFT AND PRESENTING IT ON A SCREEN OF SAID AIRCRAFT|
    JP4253239B2|2003-10-07|2009-04-08|富士重工業株式会社|Navigation system using image recognition|
    FR2874371B1|2004-08-19|2007-12-21|Airbus France Sas|DISPLAY SYSTEM FOR AN AIRCRAFT|
    CN101109640A|2006-07-19|2008-01-23|北京航空航天大学|Unmanned aircraft landing navigation system based on vision|
    US7693617B2|2006-09-19|2010-04-06|The Boeing Company|Aircraft precision approach control|
    CN100567898C|2007-01-16|2009-12-09|北京航空航天大学|Pilotless plane landing navigation method and device thereof|
    KR101008259B1|2008-09-03|2011-01-13|한국항공우주연구원|Automatic landing system and control method using circular image data for aircraft|
    US8773289B2|2010-03-24|2014-07-08|The Boeing Company|Runway condition monitoring|
    DE102012111010A1|2012-11-15|2014-06-12|Eads Deutschland Gmbh|Method and device for image-based runway localization|
    CN103226356A|2013-02-27|2013-07-31|广东工业大学|Image-processing-based unmanned plane accurate position landing method|
    FR3009117B1|2013-07-24|2016-11-25|Airbus Operations Sas|AUTONOMOUS AUTOMATIC LANDING METHOD AND SYSTEM|
    CN103853156B|2014-02-07|2016-06-01|中山大学|A kind of small-sized four-rotor aircraft control system based on machine set sensor and method|
    CN103809598B|2014-03-12|2016-08-10|北京航空航天大学|A kind of rotor wing unmanned aerial vehicles based on three layers of isosceles triangle polychrome landing ramp autonomous landing system|FR3016222B1|2014-01-03|2016-02-05|Airbus Operations Sas|METHOD AND APPARATUS FOR VERTICAL GUIDANCE OF AN AIRCRAFT DURING AN APPROACH TO A LANDING TRACK ALONG A LATERAL APPROACH TRACK|
    FR3044809B1|2015-12-03|2017-11-24|Airbus Operations |METHOD AND SYSTEM FOR AIDING THE LANDING OF AN AIRCRAFT.|
    CN105810023B|2016-05-16|2018-08-21|福建福光股份有限公司|Airport undercarriage control automatic monitoring method|
    CN108195384B|2016-12-08|2021-12-17|中国航空无线电电子研究所|Low-altitude aircraft-oriented route point indication method based on real scene|
    CN108225273B|2016-12-14|2020-06-30|中国航空工业集团公司西安航空计算技术研究所|Real-time runway detection method based on sensor priori knowledge|
    IL249870A|2016-12-29|2022-02-01|Israel Aerospace Ind Ltd|Image sensor based autonomous landing|
    US10793286B1|2018-08-23|2020-10-06|Rockwell Collins, Inc.|Vision based autonomous landing using flight path vector|
    CN109458977A|2018-10-21|2019-03-12|合肥优控科技有限公司|Robot orientation method, system and computer readable storage medium|
    CN112198891B|2020-04-22|2021-12-07|北京理工大学|Multi-gyroplane autonomous recovery method|
    法律状态:
    2015-06-26| PLFP| Fee payment|Year of fee payment: 2 |
    2016-01-29| PLSC| Publication of the preliminary search report|Effective date: 20160129 |
    2016-07-21| PLFP| Fee payment|Year of fee payment: 3 |
    2017-07-24| PLFP| Fee payment|Year of fee payment: 4 |
    2018-07-25| PLFP| Fee payment|Year of fee payment: 5 |
    2020-07-21| PLFP| Fee payment|Year of fee payment: 7 |
    2021-07-27| PLFP| Fee payment|Year of fee payment: 8 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1457208A|FR3024127B1|2014-07-25|2014-07-25|AUTONOMOUS AUTOMATIC LANDING METHOD AND SYSTEM|FR1457208A| FR3024127B1|2014-07-25|2014-07-25|AUTONOMOUS AUTOMATIC LANDING METHOD AND SYSTEM|
    US14/808,646| US9939818B2|2014-07-25|2015-07-24|Method and system for automatic autonomous landing of an aircraft|
    CN201510447103.7A| CN105302146B|2014-07-25|2015-07-27|Method and system for automatic autonomous landing of an aircraft|
    [返回顶部]