RADAR METHOD AND SYSTEM FOR DETERMINING THE ANGLE, LOCATION AND VECTORY VELOCITY OF A TARGET

Description

[0001] The present invention relates to a radar method and system for determining the angular position, the location and / or the, in particular vectorial, speed of a target.

[0002] Known radar methods, particularly for estimating vector velocity (cf. [1]-[3]), use distributed radar devices that individually and independently measure a Doppler velocity relative to an object (target). The Doppler velocity can be interpreted as the projection of the vector velocity onto a connecting vector between the radar and the target. If the positions of the radars and the target are known or can be determined, the vector velocity can be determined from the individual projections by solving a system of linear equations. However, the accuracy of this method is strongly dependent on the geometric distribution of the stations relative to the target ("dilution of precision").

[0003] Another possibility for radar measurement, especially for determining vector velocity, is the evaluation of the phase differences between two antennas (cf. [4]-

[11] ). This method enables higher accuracies and is less dependent on the problem geometry. However, the state of the art according to [4]-[7] only deals with the case where the target is at 0° azimuth (boresight) and in the array far field. Publications [8]-

[11] also deal with the case of azimuth ≠ 0°, but not the array near field. The state of the art methods according to [4]-

[11] , which are based on phase differences, use two antennas connected to a radar device.

[0004] Since comparatively large apertures are necessary for the accurate measurement of the tangential velocity, the case where the target is not at 0° azimuth and / or in the near field is of great interest. [1] D. Kellner, M. Barjenbruch, K. Dietmayer, J. Klappstein, and J. Dickmann, "Instantaneous lateral velocity estimation of a vehicle using doppler radar," in Information Fusion (FUSION), 2013 16th International Conference on, 2013, pp. 877-884. [2] H. Rohling, F. Folster, and H. Ritter, "Lateral velocity estimation for automotive radar applications," in 2007 IET Int. Conf. on Radar Systems, Edinburgh, UK, 2007, pp. 181-181. [3] W. Montlouis and P.-R. J. Cornely, "Direction of Arrival and Angular Velocities (DOAV) Estimation using Minimum Variance Beamforming," in Radar Conference, 2007 IEEE, 2007, pp. 641-646. [4] A. W. Doerry, "Patch diameter limitation due to high chirp rates in focused SAR images," IEEE transactions on aerospace and electronic systems, vol. 30, no. 4, pp. 1125-1129, 1994. [5] A. Doerry, "Tangential Velocity Measurement using Interferometric MTI Radar," 2002. [Online]. Available: http: / / prod.sandia.gov / techlib / access-control.cgi / 2002 / 023614.pdf. [Accessed Sep.26, 2017] [6] A. Doerry, B. Mileshosky, and D. Bickel, "Tangential velocity measurement using interferometric MTI radar," U.S. Patent 6 982 668 B1, Jan. 3, 2006. [7] J. A. Nanzer, "Millimeter-Wave Interferometric Angular Velocity Detection," IEEE Transactions on Microwave Theory and Techniques, Dec. 2010. [8] J. A. Nanzer and A. H. Zai, "Correction of frequency uncertainty in wide field of view interferometric angular velocity measurements," in Microwave Symposium Digest (MTT), 2012 IEEE MTT-S International, 2012, pp. 1-3. [9] J. A. Nanzer, K. Kammerman, and K. S. Zilevu, "A 29.5 GHz radar interferometer for measuring the angular velocity of moving objects," in Microwave Symposium Digest (IMS), 2013 IEEE MTT-S International, 2013, pp. 1-3.

[10] J. A. Nanzer, "Resolution of interferometric angular velocity measurements," in Antennas and Propagation (APSURSI), 2011 IEEE International Symposium on, 2011, pp. 3229-3232.

[11] J. A.Nanzer, “Micro-motion signatures in radar angular velocity measurements,” in Radar Conference (RadarConf), 2016 IEEE, 2016, pp. 1-4.

[12] T. Wagner, R. Feger, and A. Stelzer, "Wide-band range-Doppler processing for FMCW systems," in Radar Conference (EuRAD), 2013 European, 2013, pp. 160-163. .

[0005] The object of the invention is to determine an angular position, a location and / or a, in particular vectorial, speed of a target with comparatively high accuracy in the simplest possible manner.

[0006] This object is achieved by the features of claims 1, 11 and 12.

[0007] The second transceiver unit should be considered non-synchronized with the first transceiver unit, in particular, if a temporal deviation (trigger offset time) between the first and second transceiver units is greater than 1 ps, preferably greater than 1 ns, possibly greater than 10 ns. However, the temporal deviation Δt n can, if necessary, be ≤ 10 µs, preferably ≤ 1 µs (particularly for a dynamic target). In particular, the temporal deviation can be greater for a stationary target. Coherent processing means, in particular, that the measurements from the transceiver units are further processed as if they were generated by a common local oscillator. An example of this would be two receiving antennas of a SIMO radar.

[0008] This allows a target to be easily detected, in particular, its angular position, location, and / or a preferably vector velocity of the target to be determined. In particular, the present invention enables a simple and accurate estimation (determination) of the vector velocity even at an azimuth ≠ 0° and / or in the near field. Furthermore, the use of coarsely synchronized distributed radar stations to generate large apertures is possible.

[0009] A temporal deviation (trigger offset time) Δt n is understood to mean, in particular, a deviation that arises due to the process or system, in particular due to wireless and / or wired transmissions used. The temporal deviation Δt n is therefore particularly inherent in the process or system. Preferably, the measurements (or transmission signals) of the transmit / receive units are not generated by a common local oscillator.

[0010] At least two measurement signals (reflected from a common target) are preferably interfered with each other, in particular by complex conjugate multiplication, especially in the time domain (= non-exclusive alternative a)). It should be noted that complex conjugate multiplication can correspond to division by a complex phasor with normalized amplitude.

[0011] Alternatively or additionally, compressed, in particular range-Doppler compressed, signals can be interfered with each other, preferably by convolution, in particular in the frequency domain (= non-exclusive alternative b)).

[0012] Holographic processing preferably occurs in the xy direction. Interference preferably occurs along a Doppler direction, e.g., by multiplication in the time domain (alternative a)) or by convolution in the frequency domain (alternative b)). Holography can be provided in both a) and b) as the basis for the interference.

[0013] In particular (alternatively or additionally) at least two signals can be holographically interfered, first spatially, for example in a Cartesian x,y coordinate system, and then in the Doppler plane (velocity plane) by multiplication in the time domain (alternative a), in particular by conjugate complex multiplication, or by convolution in the frequency domain (alternative b).

[0014] Preferably, a vectorial velocity of the target is determined from a resulting signal.

[0015] In principle, the method is particularly preferably configured such that a (2D or 3D) vector velocity of the target can be determined.

[0016] In an advantageous embodiment, the respective (first and / or second) transmitting-receiving unit determines its own (vector) speed(s) based on a known or determined (vector) speed of one or more (for example, static / static) target(s) ( ego-motion estimation ).

[0017] In a first alternative, the target can be located in the near field. Alternatively, the target can also be located in the far field. Near field is preferably understood to mean a distance to the target that is less than or equal to 10 times the distance between the two transmit / receive units (or, in the case of multiple transmit / receive units, the distance between the two transmit / receive units that are furthest apart), or less than or equal to 10 times the aperture size of a system of transmit / receive units. A far field is then understood to mean, in particular, a distance to the target that lies above the relative values ​​just mentioned. The target is particularly preferably located in the near field. It is precisely in the near field that precise measurements can be achieved in a simple manner (in contrast to the prior art).

[0018] In a specific embodiment, the transmit / receive units form a distributed aperture. The distance between multiple transmit / receive units can be at least 20 cm, more preferably at least 50 cm, more preferably at least 100 cm (in the case of multiple transmit / receive units, this can apply either to the distance to the nearest transmit / receive unit or, alternatively, to the distance between the two transmit / receive units that are the largest apart among all pairs of possible transmit / receive units).

[0019] The target is preferably located using a holographic principle.

[0020] In specific embodiments, the method is an FMCW radar method (where FMCW stands for frequency modulated continuous wave).

[0021] Preferably, the method works according to the range Doppler principle.

[0022] In specific embodiments, the at least two transmit / receive units are monostatic. The distance between the transmit / receive units is preferably at least 5 times, more preferably at least 10 times, the respective distance between the transmit and receive antennas of one and the same transmit / receive unit (hereinafter also referred to as Rx antenna = receive antenna or Tx antenna = transmit antenna).

[0023] In particular, after holographic interference, a Fourier transformation is performed along a slow time scale. "Slow time" is defined in more detail below. A peak search can be performed to determine an ellipse / hyperbolic velocity, particularly in several or all pixels (of the target). Furthermore, a determination of ellipse / hyperbolic parameters can be performed, particularly in several or all pixels (of the target). Alternatively or additionally, a transformation of an ellipse / hyperbolic velocity into a (Cartesian) vector velocity can be performed.

[0024] A signal frequency can be halved before interference occurs.

[0025] A Doppler velocity can be determined, in particular to at least partially compensate for a Doppler shift.

[0026] An optimal filter approach can be used, in particular to at least partially compensate for a Doppler shift and / or when the radar bandwidth is comparatively high and / or the target speed is comparatively high.

[0027] Range Doppler compression can be performed using a fractional Fourier transform (FRFT), especially when the radar bandwidth is comparatively high and / or the target speed is comparatively high.

[0028] One or more of the (range-Doppler) compressed signals can be additionally compressed in the azimuth direction, preferably by means of Fourier transformation and / or digital beamforming algorithms, in particular for side lobe suppression.

[0029] In embodiments, methods are proposed wherein at least three transmit-receive units are used, wherein preferably a 3D velocity determination is carried out, in particular by forming an intersection of two pairs, each consisting of an ellipsoid and a hyperboloid. Alternatively or additionally, an optimal filter approach can be used, wherein a hypothesis(es) for several or all [ x , y , z , v x , v y , v z ] combinations are formed in a search area and compared with measured data.

[0030] According to embodiments, methods are proposed, wherein in the first transmitting-receiving unit, a first signal is generated and transmitted, in particular broadcast, via a path, in the second transmitting-receiving unit, a further first signal is generated and transmitted, in particular broadcast, via the path, a first comparison signal is formed from the first signal of the first transmitting-receiving unit and from such a first signal received by the second transmitting-receiving unit via the path, and a further comparison signal is formed from the first signal of the second transmitting-receiving unit and from such a first signal received by the first transmitting-receiving unit via the path, wherein the further comparison signal is preferably transmitted, in particular communicated, from the second transmitting-receiving unit to the first transmitting-receiving unit and / or wherein preferably a comparison comparison signal is formed from the first comparison signal and the further comparison signal and / or wherein in a first step deviations of the comparison signals which are caused by systematic deviations in the transmitting / receiving units are compensated and in a second step at least one complex value from a first of the two comparison signals or from a signal which was derived from this first comparison signal is used to adapt at least one complex value of the second of the two comparison signals or a value of a signal which was derived from this second comparison signal and thus to form an adapted signal, wherein the adaptation takes place in such a way that the vectorial sum or the difference of the complex values ​​is formed or the sum or the difference of the phases of the complex values ​​is formed by a mathematical operation.Further developments of these procedures (hereinafter also referred to as Procedure I and Procedure II) can be obtained from . DE 10 2014 104 273 A1 and / or WO 2017 / 118621 A1 be taken.

[0031] A corresponding radar system is also described in DE 10 2014 104 273 A1. The method according to DE 10 2014 104 273 A1 will be referred to below as "Method I." The configuration of the radar system according to DE 10 2014 104 273 A1 will be referred to as "Configuration I." Another method and another configuration for a radar system to increase coherence can be found in WO 2017 / 118621 A1. The method and radar system described in these two applications will be referred to below as "Method II" and "Configuration II." The present method is preferably applied to a direct path and a cross path according to Method I and / or II.

[0032] To the extent that calculations, evaluations, or other process steps are performed (e.g., in the (respective) transceiver unit), this also includes a physically independent evaluation device connected to one or more transceiver units. A control device, e.g., for triggering the start of a measurement, can also be designed as a physically independent control device (possibly in a common assembly, in particular a housing, with the evaluation device) connected to one or more transceiver units. For example, the respective transceiver unit can be designed as an arrangement comprising, in particular, one or more antennas with a few signal-generating or signal-processing components, while other components, such as signal comparison units or a control and / or evaluation device, can be connected to such an arrangement as structurally independent components.If components are used, these can (as far as technically feasible) be designed as so-called hardware consisting of processing components and / or be implemented as signals or data processing steps executed entirely or partially in a processor.

[0033] In general, the control and / or evaluation device can be a component of one or more transceiver units, or can be connected to one or more such transceiver units. If necessary, a physically independent control and / or evaluation device can be provided, which is connected to the respective transceiver unit or the other components of the respective transceiver unit. Alternatively, the control and / or evaluation device can be integrated into the first and / or second (generally additional) transceiver units, for example, in a common housing and / or as a structural unit.

[0034] Each transceiver unit can have one or more transmitting or receiving antennas.

[0035] The above-mentioned object is further achieved by a mobile device, in particular a vehicle, preferably a passenger car and / or a truck, comprising the system according to claim 12.

[0036] In the following description, further principles, aspects, and embodiments of the invention are described with reference to the accompanying figures. Herein: Fig. 1 two receiving units at a distance from each other; Fig. 2 two receiving units in a Fig. 1 different configuration; Fig. 3 Representation of the movement of a target in the near field; Fig. 4 Resulting 3D spectra for estimating a speed; Fig. 5 A measurement situation with two passenger cars; Fig. 6 A measurement situation with four passenger cars and a pedestrian.

[0037] In the following description, the same reference numbers are used for identical and equivalent parts. Coherent processing of distributed radar stations for positioning:

[0038] The phase of the transmission signal ϕ Tx or receive signal ϕ Rx of an FMCW radar operating on the Range-Doppler principle can generally be described as φ Tx t f t i = 2 π f 0 t f + μt f 2 2 + φ 0 . and φ Rx t f t i = 2 π f 0 t f − τ + μ t f − τ 2 2 + φ 0 + φ t .

[0039] With: tf = 0 ...T = t - ti : "quick time" or "fast time"; T is a sweep duration, t is absolute time. ti = iT : "slow time", i = 0 ... M is a sweep number, M is the number of ramps in a burst. τ = d rt t f t i c : "Time of flight" from the Tx antenna to the target and back to the Rx antenna; d rt ( tf , ti ) is the corresponding distance (see (4)). f 0 : carrier frequency; μ = B T : sweep slope; B:Bandwidth; c: propagation speed. ϕ 0 : Unknown initial phase of the sweep. ϕ t : reflection phase of the target.

[0040] The phase of the (baseband) signal resulting from mixing the Rx signal (received signal) with the Tx signal (transmitted signal) is then φ t f t i = φ Tx t f t i − φ Rx t f t i ≈ d rt 2 t f t i ≪ d rt t f t i t 2 π c f 0 + μt f d rt t f t i + φ t , with the round-trip distance d rt t f t i ≈ d ≫ v r MT d rt , 0 + 2 v r t f + t i .

[0041] This refers to d rt,0 is the round-trip distance to the target at the beginning of a burst and v r is the radial velocity of the target relative to the radar. Inserting (4) into (3) and neglecting terms that t f 2 , t i 2 and t f ti contain, leads to φ t f t i ≈ 2 π c f 0 , d rt , 0 + μd rt , 0 + 2 f o v r t f + 2 f 0 v r t i + φ t .

[0042] The resulting signal phase in (5) depends only on the distance to the target, the relative radial velocity of the target, and the reflection phase of the target. The unknown initial phase ϕ 0 disappears during the mixing process in (3).

[0043] If one uses N distributed radar stations (transmitting-receiving units) whose clocks are not frequency synchronized but are of the same type, where the measurement start is carried out via wire or wirelessly simultaneously with an accuracy of Δ tn is triggered and whose ramps are set with a frequency offset Δ fn are offset from each other, the following signal model results for the baseband signal phase in radar number n: φ n t f t i ≈ 2 π c f 0 + Δ f n d rt , 0 , n + μd rt , 0 , n + 2 f 0 + Δ f n v r , n t f + Δ t n + 2 f o + Δ f n v r , n t i + Δ t n + φ t .

[0044] The individual phase terms have the following meaning: φ ϕ , n = 2 π c f 0 1 + Δ f n f 0 d rt , 0 , n + φ t ≈ Δ f n ≪ f 0 2 π c f 0 d rt , 0 , n + φ t : Phase offset term that depends on the path difference between the stations and can be used for angle estimation. f d = ∂ φ n t f t i ∂ t f ≈ 2 π c μd rt , 0 , n : Frequency proportional to the distance. Resulting from compression (e.g., using FFT) in the "fast time" direction. f v r = ∂ φ n t f t i ∂ t i ≈ 2 f 0 1 + Δ f n f 0 v r , n ≈ Δ f n ≪ f 0 2 f 0 v r , n : Frequency proportional to the radial velocity. Resulting from compression (e.g., using FFT) in the "slow time" direction. ϕt : Phase term that is approximately the same at each station n, provided the reflectance of the target is identical in each direction illuminated by the radars.

[0045] Since the trigger offset time Δ tn ≈ 15 ns (possibly 1 to 100 ns, especially 5 to 30 ns) for wired and Δ tn ≈ 1 µs (possibly 0.1 to 10 µs, especially 0.5 to 2 µs) for wireless triggering, in general | v |Δ tn ≈ 0, ie the target position remains during the time Δ tn approximately constant.

[0046] ϕ φ,n It therefore only depends on the distance between the respective radar and the target and can be processed coherently at all stations. The radars form a distributed aperture.

[0047] For the signal sn in station n after a range compression (compression along the fast time t f ) applies s n d t 0 = W d d − d rt , 0 , n ⋅ exp j 2 π c f 0 d rt , 0 , n + φ t , where W dthe Fourier transform of the window function used in range. Based on this signal model, a 2D location of the target can be performed according to the holography principle using the following optimal filter: h n d n , hyp x hyp y hyp = exp − j 2 π c f 0 d n , hyp x hyp y hyp , with the hypothesis for the round-trip distance d n , hyp x hyp y hyp = p Tx , n − p t , hyp + p t , hyp − p Rx , n with p t , hyp = x hyp y hyp .

[0048] p Tx, n or p Rx, n is the known 2D position of the Tx antenna or Rx antenna of radar n. Each radar (transceiver unit) can have one or more Tx or Rx antennas. If more than one antenna is present, a hypothesis must be created for each Tx-Rx combination using the same principle. p t,hyp is a hypothesis to be tested for the 2D position of the target. The method can be directly adapted to 3D if the coordinates in (9) are replaced by 3D coordinates.

[0049] The probability that a target is at the position p t,hyp is then calculated as follows: I x hyp y hyp = ∑ n = 1 N h n d n , hyp x hyp y hyp ⋅ s n d t 0 .

[0050] If the target is in the far field of the aperture, (10) reduces to the beamforming approach and the 2D search over x and y can be replaced by a 1D search for the distance followed by a search for the angle.

[0051] If the target is located in the near field of the aperture, a 2D search according to (10) is preferably performed. Since the target may be located in different range bins in the individual radars ("range migration"), interpolation in the range direction may be necessary, which can be implemented as a complex-valued linear interpolation.

[0052] This method allows for the coherent processing of non-synchronized distributed radar stations, enabling the realization of large apertures. This enables target localization with high accuracy. Coherent processing of distributed radar stations for determining the vectorial target velocity in the far field

[0053] Fig. 1shows two transmit-receive units 1,2 at a distance of bl from each other, which measure approximately simultaneously according to the FMCW Range Doppler principle to the target, which moves with a vectorial velocity v.

[0054] If a target is located in the far field of the antenna array that includes the transmit-receive units 1,2 (and possibly further transmit-receive units), the geometry can be configured as shown in Fig. 1 The target is originally located at [ d 0 , θ 0 ] in the radar polar coordinate system and moves with the vector velocity v The change in distance between the radars and the target (radial component) can be described as d t = d 0 + v r t with the radial velocity v r . The angle change (tangential component) is θ t = θ 0 + ωt = θ 0 + v t d o t , where ω the angular velocity and vt denotes the tangential velocity. The corresponding signal model in the transmit-receive units 1, 2, which operate according to the FMCW range-Doppler principle, after range compression (e.g. using the Fourier transform), is s 1 d t i = δ d d − 2 d 0 ⋅ exp j 2 π λ 2 d 0 + 2 v r t i , s 2 d t i = W d d − 2 d 0 ⋅ exp j 2 π λ 2 d 0 + 2 v r t i + 2 b I sin θ o + ω a t i , where W d the Fourier transform of the window function used in range. ti denotes the start time of the FMCW ramps ("slow time"). For the measured distances to the target in both radars applies d 1 ≈ d 2 = d 0 due to the far-field approximation d 0 » b l . Interference between both signals results in s d t i = s 1 ∗ d t i ⋅ s 2 d t i = δ d d − 2 d 0 ⋅ exp j 2 π λ b I sin θ 0 + ω a t i , where |·|* denotes the complex conjugation. With the approximation ω a ti ≈ 0 and by Taylor series expansion (14) can be approximated as s d t i ≈ W d d − 2 d 0 ⋅ exp j 2 π λ b l sin θ 0 + b l cos θ 0 ω a t i .

[0055] A subsequent Fourier transformation along ti results s d , 2 π f a = F s d t i , t i = W d d − 2 d 0 ⋅ W ω a 2 π f a − 2 π λ b l cos θ 0 ω a ⋅ exp j 2 π λ b l sin θ 0 . fa denotes the resulting signal frequency in the image area (azimuth direction). The angular velocity ω a / tangential speed v t then result in v t = ω a d 0 = λd 0 b l cos θ 0 f a , max , where f a,max is the position of the maximum in the 2D spectrum s ( d, 2π f a ) along the f a direction (azimuth direction in the image area).

[0056] The interference analogue (14) can also be applied to range-Doppler compressed data. The multiplication in (14) must then be replaced by a convolution.

[0057] By estimating v r using the standard range Doppler method and the orthogonal tangential velocity v t from (17) the complete velocity vector can be determined. This requires knowledge of the angle to the target θ0. This can be determined, for example, using well-known angle estimation techniques (beamforming). Resolution and uniqueness range

[0058] The resolution of the tangential velocity estimation for a rectangular window can be calculated by inserting f a = 1 NT in (17) to Δv t = λd 0 b l cos θ 0 NT . (For Hann windowing, the resolution is ≈ 2 times lower.)

[0059] By using f a = f sampling 2 = 1 2 T in (17) we get v t , max = ± λd 0 b l cos θ 0 2 T . for the uniqueness range.

[0060] Both resolution and unambiguity range therefore depend on the distance and angle to the target. Increasing the aperture b l leads to an improvement in resolution and a reduction in the uniqueness range. If additional receiving units are present between the existing two, this leads to an increase in the uniqueness range. General solution for coherent processing of distributed radar stations to determine the vectorial target speed (also in the near field)

[0061] At least two roughly synchronized radar units (transceiver units) are required. The distance between the units is b l (see Fig. 2; Fig. 2 shows two transmitting-receiving units at a distance of b l to each other, which measure approximately simultaneously to the target according to the FMCW Range-Doppler principle). Both radar units are quasi monostatic (respective distance between Rx and Tx << b l ). For the distance to the target d 0 does not necessarily apply d 0 » b l , ie the target can be in the near field.

[0062] In this case, the signal model according to the FMCW Range-Doppler measurement principle can be expressed after range compression as: s 1 d t i = A 1 W d d − d rt , 0,1 ⋅ exp j 2 π λ d rt , 1 t i , s 2 d t i = A 2 W d d − d rt , 0,2 ⋅ exp j 2 π λ d rt , 2 t i . d rt, n ( t i ) describes the double path distance from the radar Tx antenna to the target and back to the Rx antenna: d rt , n t i = p Tx , n − p t t i + p t t i − p Rx , n with p t t i = x 0 y 0 + v x v y t i . p Tx, n / p Rx, n is the 2D position of the Tx / Rx antenna of radar unit n. p t is the position of the target in 2D. ∥·∥ denotes the Euclidean norm. For ti = 0 applies d rt,0, n = d rt, n (0).

[0063] Fig. 1 shows a representation of the movement of a target in the near field of a 2-radar unit array as movement along ellipses / hyperbolas. From the geometry in Fig. 3 It can be seen that the position of the target can be determined by the intersection of an ellipse and a hyperbola with both radars at their focal points. The ellipse is determined by the parameter α E = 0.5 d rt , 2 + d rt , 1 2 and the hyperbola by a H = 0.5 d rt , 2 − d rt , 1 2 The resulting ellipse / hyperbola equations are Ellipse : x 2 a E 2 + y 2 b E 2 = 1 mit b E 2 = a E 2 − c HE 2 und c HE = b 1 2 ; Hyperbel : x 2 a H 2 − y 2 b H 2 = 1 mit b H 2 = c HE 2 − a H 2 .

[0064] The intersection point of the ellipse with the hyperbola [ x 0 , y0 ] (strictly speaking, there are two intersection points; one of them can easily be selected by plausibility considerations - the other is behind the radar units) x 0 = − a E a H c HE , y 0 = − c HE 4 + a H 2 + a E 2 c HE 2 − a E 2 a H 2 c HE

[0065] For a moving target you get x = x 0 + vxti and y = y 0 + vyti and thus a E ( ti ) and a H ( ti ) (these linear equations apply to the case ∥ v ∥ T « d rt , where T is the FMCW ramp duration).

[0066] Direct interference of both signals may not be possible here, since d rt,0,1 ≠ d rt,0,2 ("range migration"). Instead, a holographic interference approach is preferred: s Δ x hyp y hyp t i = A 1 s 1 d 1 , hyp x hyp y hyp , t i ∗ ⋅ A 2 s 2 d 2 , hyp x hyp y hyp , t i = A 12 W xy x − x 0 , y − y 0 ⋅ exp j 2 π λ d rt , 2 t i − d rt , 1 t i = A 12 W xy x − x 0 , y − y 0 ⋅ exp j 2 π λ 4 a H t i ,

[0067] With d n , hyp x hyp y hyp = p Tx , n − p t , hyp + p t , hyp − p Rx , n with p t , hyp = x hyp y hyp .

[0068] p t,hypis a hypothesis for the target position in 2D. If both signals are interfered without the complex conjugation, one obtains s Σ x hyp y hyp t i = A 1 s 1 d 1 , hyp x hyp y hyp , t i ⋅ A 2 s 2 d 2 , hyp x hyp y hyp , t i = A 12 W xy x − x 0 , y − y 0 ⋅ exp j 2 π λ d rt , 2 t i + d rt , 1 t i = A 12 W xy x − x 0 , y − y 0 ⋅ exp j 2 π λ 4 a E t i .

[0069] The target movement can therefore be described as a movement a E ( ti ) from ellipse to ellipse along a hyperbola according to a radial velocity and as a movement a H ( ti ) perpendicular to it from hyperbola to hyperbola along an ellipse according to a tangential velocity (cf. Fig. 3 ).

[0070] A subsequent Fourier transformation along ti results s Δ x hyp y hyp ω Δ = F s Δ x hyp y hyp t i , t i = A 12 W xy x − x 0 , y − y 0 ⋅ W Δ ω Δ − 2 π λ 4 d a H t i d t i ⋅ exp j 4 π λ d rt , 0,2 − d rt , 0,1 , s Σ x hyp y hyp ω Σ = F s Σ x hyp y hyp t i , t i = A 12 W xy x − x 0 , y − y 0 ⋅ W Σ ω Σ − 2 π λ 4 d a E t i d t i ⋅ exp j 4 π λ d rt , 0,2 + d rt , 0,1 .

[0071] The result of this transformation is in Fig. 4 shown. Fig. 4 shows a resulting 3D spectrum for estimating the tangential / radial velocity at given x and y. Both spectra have the same shape but are located at different positions along the z-axis.

[0072] A maximum search along the ω Δ or ω Σ direction in the resulting 3D spectrum provides d a E t i d t i = λ 8 π ω Σ , max und d a H t i d t i = λ 8 π ω Δ , max .

[0073] Using the partial derivative of (26) and (32), the components of the velocity vector can be calculated as v r , x = ∂ x ∂ a E d a E t i d t i und v r , y = ∂ y ∂ a E d a E t i d t i ; v t , x = ∂ x ∂ a H d a H t i d t i und v t , y = ∂ y ∂ a H d a H t i d t i .

[0074] The velocity vector in Cartesian coordinates is then v x = v r , x + v t , x , v y = v r , y + v t , y . Preservation of the uniqueness range

[0075] The phase addition in (29) leads to a doubling of the measured Doppler frequency, which results in a halving of the unambiguous measurement range. This can preferably be avoided by halving the signal frequency before interference formation. For a generic signal in analytical form s ( t ) = A exp(j φ ( t )) this can be expressed as s f / 2 = A exp j ⋅ unwrap ϕ t 2 .

[0076] (Frequency scaling property of the Fourier transform). This allows the uniqueness region in the Doppler direction to be preserved, at least essentially. Doppler shift compensation

[0077] The Doppler shift can preferably be compensated by determining the Doppler velocity for each target and thus correcting the distance. Alternatively, an optimal filter approach similar to

[12] can be pursued. The algorithms presented in this invention can then preferably be applied unchanged to the result. High bandwidth or high speed

[0078] If the radar has a high bandwidth or the target is moving very fast, so that the approach 4 π B cT v r t i ≈ 0 is no longer met, FFT-based Doppler compression may no longer be applicable because the distance to the target changes noticeably from ramp to ramp during the burst.

[0079] This case can also be covered with an optimal filter approach similar to

[12] . Alternatively, range Doppler compression can be performed using the fractional Fourier transform (FRFT). The algorithms presented in this paper can then be applied to the result unchanged. Sidelobe suppression

[0080] If both radar units have two or more antennas, a range-Doppler-compressed signal can be further compressed in the azimuth direction using a Fourier transform or digital beamforming algorithms (Bartlett, Capon, MUSIC, etc.). The presented method can then be applied unchanged to the result. This leads to a suppression of the side lobes in the resulting [ x , y , v x , v y ] image. Allocation of targets in the 3D spectrum for radial and tangential velocity to each other

[0081] When there is more than one target in a spatial resolution cell, targets can be separated based on amplitude, since the signal originating from one target has the same amplitude in both spectra A 12 , but has different frequency and phase (cf. equations (30) and (31) and Fig. 4 If two or more targets with the same amplitude are present in a spatial resolution cell, target separation can be achieved using a subsequent tracking algorithm. Extension to 3D / more than 2 radar stations

[0082] 3D vector velocity estimation of a target is possible using at least three spatially positioned radar units (transceivers). The result is then obtained by intersecting two pairs, each consisting of an ellipsoid and a hyperboloid.

[0083] The method can be generalized to any number of transmit-receive units and radar arrays, especially if an optimal filter is used instead of Fourier transform processing. For this method, a hypothesis is defined for each [ x, y, z, v x ,v y , v z ] combination in the search area and compared with the measured data. For a target at the hypothetical position p hyp = [x hyp , y hyp , z hyp ] and the radar number n at the position p n the distance dn ( x hyp , y hyp , z hyp ) between radar n and target and the unit vector vn ( x hyp , y hyp , z hyp ) from radar to target to calculate d n x hyp y hyp z hyp = p Tx , n − p t , hyp , v n x hyp y hyp z hyp = p Tx , n − p t , hyp p Tx , n − p t , hyp .

[0084] For the measurement signal sn ( d, ti ) according to the FMCW Range-Doppler principle in radar n the following optimal filter can be formulated a n x hyp y hyp z hyp v x , hyp v y , hyp v z , hyp t i = exp − j 4 π λ v r , hyp , n x hyp y hyp z hyp v x , hyp v y , hyp v z , hyp t i , with the hypothesis v r,hyp, n ( v x,hyp , v y,hyp , v z,hyp ) for the radial velocity (Doppler velocity): v r , hyp , n x hyp y hyp z hyp v x , hyp v y , hyp v z , hyp = v n x hyp y hyp z hyp ⋅ v hyp = v n x hyp y hyp z hyp ⋅ v x , hyp v y , hyp v z , hyp .

[0085] |·| denotes the scalar product of two vectors. The result is thus calculated as m r , hyp , n x hyp y hyp z hyp v x , hyp v y , hyp v z , hyp = ∑ n = 1 N rad ∑ i = 1 N sw s n d x hyp y hyp z hyp , t i a n x hyp y hyp z hyp v x , hyp v y , hyp v z , hyp t i with the number of radar units N rad and FMCW sweeps number N sw . The target position and vector velocity can be determined by maximum search in the resulting 4D pseudospectrum. For 3D problems, a 6D spectrum can be obtained using an analogous procedure.

[0086] This method may be more computationally intensive than the FFT-based method presented previously. Application of the method to a direct path and a cross path obtained by method I and / or II

[0087] If two radars measure to a target using method I and / or II, two direct measurement paths are obtained ( d rt,11 = 2 d 1 : Radar 1 → Target → Radar 1, d rt,22 = 2 d2 : Radar 2 → Target → Radar 2) and two cross paths ( d rt,12 = d 1 + d 2 : Radar 1 → Target → Radar 2, d rt,21 = d 2 + d 1 : Radar 2 → Target → Radar 1). Since method I and / or II enables phase-coherent evaluation of the cross paths, these can also be processed to estimate the vector velocity using the described method.

[0088] If the hyperbolic and ellipse parameters from equations (22) and (23) are d rt,11 and d rt,21, we get: a E = d rt , 21 2 = d 2 + d 1 2 and a H = d rt , 21 − d rt , 11 2 = d 2 − d 1 2 .

[0089] The method can thus be applied unchanged to a combination of a direct path and a cross-path. This has the advantage of revealing targets that are visible in the cross-path and a direct path, but not in both direct paths. Application examples for vector velocity estimation using the proposed method

[0090] Fig. 5shows an automotive application with a car (20) equipped with two radars (1), (2) that are triggered approximately simultaneously by a signal. Another car (30) is driving in the direction of the arrow (4).

[0091] For example, car (20) can be waiting at an intersection or approaching the intersection. With standard processing of the radar data on car (20), the distance and angle to car (30) can be determined. Furthermore, the relative radial velocity can be determined. Since in this case car (30) is moving approximately tangentially to the connecting axis between radar (1), (2) and the target, i.e. car (30), the Doppler frequency would be ≈ 0 Hz. The measured radial velocity would therefore be ≈0 m / s. It is therefore not possible to determine whether the car (30) is moving or stationary. With the method proposed according to the invention, the signals from the radar stations (1), (2) can be processed coherently, even though they are not (phase) synchronized, and from this the (complete) vectorial velocity of the car (30) can be determined. This information can be fused with further algorithms for environment detection, which is advantageous for driver assistance systems and autonomous driving. Using this method, possibly in combination with existing methods, each point in the radar image can be assigned an amplitude (power), a speed, and a direction of movement.

[0092] Fig. 6shows another application with a car (20) equipped with two radars (1), (2) that are triggered approximately simultaneously by a signal, with the car driving in the direction of arrow (4). Parked cars (30) are present at the side of the road. A pedestrian (5) walks onto the road in the direction of arrow (6).

[0093] If the car (20) is moving in the direction of arrow (4) and there are parked cars (30) at the side of the road, the car (20) would not detect the pedestrian (5) moving in the direction of arrow (6) until the last moment, as it would be obscured by the parked cars. An additional measurement of the tangential speed once the pedestrian can be seen by the radar can enable a faster response and possibly prevent an accident.

[0094] Overall, the present invention also encompasses, in particular, a particularly accurate estimation of the (2D or 3D) vectorial velocity of objects (targets). For this purpose, at least two (spatially separated) FMCW radars can be provided, which measure in particular according to the range Doppler principle. These FMCW radars are preferably only roughly time-synchronized. Both stations can be triggered (approximately) simultaneously and transmit and receive FMCW bursts (with a known frequency offset from each other). The complete vectorial velocity of the objects (targets) in the environment can then be estimated from an interference of resulting (baseband) signals.

Claims

1. Radar method for determining the angular position, the location and the vectorial speed of a target, wherein a first transceiver unit (1) and at least one second transceiver unit (2) spatially separate from the first transceiver unit (1), which are not synchronized, a start of measurement of the first transceiver unit (1) and of the second transceiver unit (2) are triggered wirelessly or by wire, but with a time deviation Δtn, wherein measurements of the transceiver units (1, 2) are processed coherently, wherein a determination of elliptic / hyperbolic parameters are carried out, in particular in several or all pixels, and a transformation of the elliptic / hyperbolic velocity into a vectorial, preferably Cartesian, velocity is carried out, wherein the elliptic / hyperbolic parameters are determined on the basis of a double-path distance drt,n(ti) from the radar transceiver unit (1,2) to the target and back to the transceiver unit (1,2) by the following formulas: the elliptic parameter a E = 0.5 d rt , 2 + d rt , 1 2 and the hyperbolic parameter a H = 0.5 d rt , 2 − d rt , 1 2 , wherein the double-path distance drt,1(ti) refers to the first transceiver unit (1,2) and drt,2(ti) refers to the second transceiver unit (1,2), wherein ti is a slow time.

2. Method according to claim 1, wherein a) at least two measurement signals from the first transceiver unit (1) and from the second transceiver unit (2), which have been reflected by a common target, are interfered with one another, in particular by complex-conjugate multiplication, and / or b) range-Doppler-compressed signals are interfered with one another, in particular by convolution and / or c) holography is carried out, in particular in the xy direction, and interference is along a Doppler direction.

3. Method according to one of the preceding claims, wherein the target is located in the near field, wherein the near field is that region in which the distance of the target is less than 10 times a distance between the furthest-spaced transceiver units of the system.

4. Method according to one of the preceding claims, wherein the transceiver units (1, 2) form a distributed aperture.

5. Method according to one of the preceding claims, wherein a location of the target is carried out according to the holography principle, by holographically interfering at least two measurement signals of the first transceiver unit (1) and of the second transceiver unit (2), respectively.

6. Method according to one of the preceding claims, wherein the method is an FMCW radar method and / or is a method that operates according to the range-Doppler principle.

7. Method according to one of the preceding claims, wherein the at least two transceiver units (1, 2) are monostatic.

8. Method according to one of the preceding claims, wherein a signal frequency is halved before interference is formed and / or wherein a Doppler velocity is determined, in particular for at least partially compensating a Doppler shift and / or wherein an optimal filter approach is used, in particular for at least partially compensating a Doppler shift and / or in the case of a comparatively high bandwidth of the radar and / or a comparatively high velocity of the target, and / or wherein a range-Doppler compression is carried out by means of a fractional Fourier transform (FRFT), in particular in the case of a comparatively high bandwidth of the radar and / or a comparatively high velocity of the target and / or wherein one or more of optionally range-Doppler-compressed signals is additionally compressed in the azimuth direction, preferably by means of a Fourier transform and / or digital beamforming algorithms, in particular for side lobe suppression.

9. Method according to one of the preceding claims, wherein at least three transceiver units are used, wherein preferably a 3D velocity determination is carried out, in particular by forming an intersection of two pairs, each of which consists of an ellipsoid and a hyperboloid, and / or wherein an optimal filter approach is used, wherein a hypothesis for a plurality or all of [x, y, z, vx, vy, vz] combinations in a search area are formed and compared with measurement data.

10. Method according to one of the preceding claims, wherein - a first signal is generated in the first transceiver unit (1) and transmitted, in particular broadcast, via a path, - a further first signal is generated in the second transceiver unit (2) and transmitted, in particular broadcast, via the path, - a first comparison signal is formed from the first signal of the first transceiver unit (1) and from a first signal received via the path from the second transceiver unit (2), and - a further comparison signal is formed from the first signal of the second transceiver unit (2) and from a first signal received via the path by the first transceiver unit (1), - wherein the further comparison signal is preferably transmitted, in particular communicated, from the second transceiver unit (2) to the first transceiver unit (1), and / or wherein a comparison-comparison signal is preferably formed from the first comparison signal and the further comparison signal and / or wherein deviations of the comparison signals, which are caused by systematic deviations in the transceiver units, are compensated in a first step and at least one complex value from a first of the two comparison signals or from a signal which has been derived from this first comparison signal is used to adjust at least one complex value of the second of the two comparison signals or a value of a signal that was derived from this second comparison signal, and to thus form an adjusted signal, wherein the adjustment is carried out in such a way that the vectorial sum or the difference of the complex values is formed by means of a mathematical operation, or the sum or the difference of the phases of the complex values is formed.

11. Radar system for determining the angular position, the location and the vectorial speed of a target, in particular for carrying out the method according to one of the claims 1 to 10, wherein a first transceiver unit (1) and at least one second transceiver unit (2) are provided, which are not synchronized with each other, wherein a control device is provided, which is configured to trigger a start of measurement of the first transceiver unit (1) and the second transceiver unit (2) with a time deviation Δtn, wirelessly or by wire, wherein a computing and / or evaluation unit is provided which is configured in such a way that measurements of the transceiver units (1, 2) are processed coherently, wherein a determination of elliptic / hyperbolic parameters, in particular in several or all pixels, is carried out and a transformation of the elliptic / hyperbolic velocity into a vectorial, preferably Cartesian, velocity is carried out, wherein the elliptic / hyperbolic parameters are determined using a double-path distance drt,n(ti) from the radar transceiver unit (1, 2) to the target and back to the transceiver unit (1, 2) by the following formulas: the elliptic parameter a E = 0.5 d rt , 2 + d rt , 1 2 and the hyperbolic parameter a H = 0.5 d rt , 2 − d rt , 1 2 , wherein the double-path distance drt,1(ti) refers to the first transceiver unit (1,2) and drt,2(ti) refers to the second transceiver unit (1,2), wherein ti is a slow time.

12. Use of the method according to one of claims 1 to 10 and / or of the system according to claim 11, for a mobile device, preferably a vehicle, in particular a passenger car and / or truck.

13. Mobile device, in particular a vehicle, preferably a passenger car and / or truck, comprising a system according to claim 11.