DE102014003152A1 - Method and device for determining at least one object parameter of a means of transport - Google Patents

Abstract

A method for determining at least one object parameter (1042) of a means of transport (102) is presented. The method comprises a step of reading in (1160) a plurality of first raw targets (210) at a first time (t1) and a plurality of second raw targets (210) at a second time (t2) and a step of releasing (1162) of a mathematical model using the first raw targets (210) and the second raw targets (210) to determine the at least one object parameter (1042). A first raw target (210) and / or a second raw target (210) is a raw target (210) representing a characteristic point of the means of transport (102) determined by a radar measurement, each position having a position at the first time (t1). and / or second time (t2) can be assigned and / or assigned.

Description

State of the art

The present invention relates to a method for determining at least one object parameter of a means of transport, a corresponding device for determining at least one object parameter of a means of transport and a corresponding computer program product.

Since the late 1950s, traffic radars with a narrow opening angle (~ 5 °) of the radar lobe are used. These are installed at the edge of the road at an acute angle to the direction of travel (~ 20 °) in the case of "across-the-road" applications. The velocity is obtained taking into account the acute angle (cosine correction) from the Doppler shift. The speed value is obtained by correcting the Doppler signals detected during a passage of a vehicle by a fixed angle value and evaluating them by means of a histogram.

Over the past decade, traffic radars with tracking radar sensors have found their way into traffic monitoring. These sensors have an opening angle of> 20 ° and can detect not only the speed but also the distance and angle. These sensors can both "a-long-the-road", this means that the center axis of the emitted radar beam is aligned almost parallel to the direction of travel of a road (since a radar sensor is often installed on the roadside, it can also, for example, 5 ° to the direction of travel this means that the central axis of the emitted radar beam is oriented significantly more than 5 ° to the direction of travel, and preferably a marginal beam delimiting the radar beam in the direction of travel near the outer end of the roadway to be detected is located and the other limiting edge beam near the bisector to the vertical to the direction of travel, are used. In "across-the-road" applications, installation angles of the central axis of the radar beam from half of the bisecting line, ie around 22.5 °, have proven to be particularly advantageous. In principle, both applications additionally distinguish between incoming and outgoing traffic. During inbound traffic, the appropriate objects move towards the radar sensor during the measurement process and away from the radar sensor during outgoing traffic during the measurement process. Due to the large opening angle, these sensors can detect vehicles over a longer distance and capture their trajectory over time. The trajectory values are determined by means of a tracking algorithm and output as measured value on previously parameterized virtual trigger lines.

The publication DE 10 2012 107 445 A1 describes a method for classifying moving vehicles.

In the EP 0 286 910 A1 A traffic surveillance apparatus for photographic registration of vehicles is disclosed.

Disclosure of the invention

Against this background, the present invention proposes a method for determining at least one object parameter of a means of transport, a device for determining at least one object parameter of a means of transport and a corresponding computer program product according to the main claims. Advantageous embodiments emerge from the respective subclaims and the following description.

A method for determining at least one object parameter of a means of transport is presented, the method comprising the following steps:
Reading in a plurality of first raw targets at a first time and at least a plurality of second raw targets at a second time, wherein a first raw target and / or a second raw target is a raw target representing a characteristic point of the means of transport determined by a radar measurement, each one characteristic point a position at the first time and / or second time is assignable and / or is assigned; and
Solving a mathematical model using the first raw targets and the second raw targets to determine the at least one object parameter.

The method for determining at least one object parameter of a means of transport may be used for traffic monitoring. An object whose parameter is to be captured can be a Transport (such as a vehicle, in particular a motor vehicle, a track-bound transport, a watercraft or an aircraft) are understood. In particular, a speed of a vehicle can be determined as the object parameter. The method for determining at least one object parameter of a means of transport may be part of a traffic monitoring system or executed in such a traffic monitoring system. Traffic monitoring systems also serve to enforce rules and laws in road traffic. A traffic monitoring system can serve the speed and distance measurement. By means of a radar measurement, a plurality of raw targets can be detected and provided at one time. Generally, raw targets may be understood to mean a plurality of first raw targets and, in addition or alternatively, a plurality of second raw targets. A raw target can be understood as a reflection response, for example, of a radar radiation of a (tracking) radar sensor. The raw targets can be summarized to a point cloud. In this case, a point cloud may comprise a plurality of raw targets at a time. In this context, a transport destination can be summarized in a cloud of points. A raw target can be understood as a measuring point. The first time and the second time may be different. The second time may be later than the first time.

Furthermore, in the step of releasing as object parameter at least one information about a speed and additionally or alternatively a direction of travel and additionally or alternatively a direction error and additionally or alternatively a movement angle and additionally or alternatively an object acceleration and additionally or alternatively a position on a drive axis at a projection time and additionally or alternatively a size of the means of transport can be determined. Advantageously, such a specific object parameter can be used for traffic monitoring.

It is also favorable if, in the reading-in step, at least a plurality of third raw targets are read in at a third time. In the solving step, the mathematical model may be solved using the plurality of third raw targets. The third time may be different from the first time and in addition or alternatively from the second time. Furthermore, a plurality of raw targets may be read in at a variety of different times and used in the solving step. Thus, the accuracy or robustness of the method can be increased.

It is also favorable if in the step of solving the mathematical model is solved using a least squares method. In the solving step, the mathematical model can be solved using a balance calculation, a regression, a fit, or a parameter estimate. In the solving step, the object parameters can be determined using various optimization methods. In doing so, a method can be chosen from a performance or quality point of view. Also, a plurality of optimization methods can be combined. An advantage is an efficient and reliable determination of the at least one object parameter.

In the solving step, the mathematical model may represent a projection of the plurality of raw targets at the respective time to a reference time. Under the reference time can be understood any time. All raw targets can be projected to one, in particular arbitrary, time. In this case, a projection can create a quality feature or a security of the found object parameter. In the solving step, the mathematical model may represent a non-recursive distance-tracking tracking filter or determine the object parameter using a distance non-recursive tracking filter. Advantageously, a robust mathematical model can be used.

Further, the method may include a step of verifying. In the step of verifying, the at least one object parameter may be verified to a reference time using a projection of the plurality of raw targets at the respective time. Using the object parameter found by means of the mathematical model, the majority of the first raw targets and additionally or alternatively the majority of the second raw targets can be projected to a predefined time. An overlay value may represent an indication of a quality of the particular object parameter. Advantageously, the object parameter can be checked.

Further, the method may include a step of estimating a traffic position using a radar signal. Further, the method may include a step of tracking a location area using the transportation location. In this case, the first raw targets and additionally or alternatively the second raw targets can be read using the location area in the step of reading. In particular, the location area may represent an area in which the Means with a predetermined probability at a predetermined time. A residence area can be understood as a bounding box. The step of estimating and the step of tracking can provide a rudimentary tracking of a vehicle for object location determination. In this case, an object can be understood as a means of transport. Advantageously, in the reading step, the raw targets can be limited to those belonging to a cloud of points or which can be assigned to a means of transport. Thus, erroneous raw targets or raw targets of other objects can be filtered out using the location area.

In the read-in step, at least one raw target of the raw targets may include information about an assumed speed and additionally or alternatively an assumed direction of travel. Generally, the raw targets of a plurality of raw targets may be assigned an assumed speed or assumed direction of travel. In this case, the assumed speed can correspond to the real speed of the means of transport at the time or the object parameter. The assumed direction of travel can correspond to the real direction of travel or the object parameter. In this case, a match can also take place within a predefined tolerance range.

In the read-in step, at least one raw target of the plurality of first raw targets and, in addition or alternatively, of the plurality of second raw targets, signal strength information may be assignable and additionally or alternatively assigned. In the solving step, the mathematical model can be solved using the signal strength information. More simply stated, signal strength information may be associated with a raw target of the raw targets. In the solving step, the mathematical model can take into account the signal strength information. Advantageously, strong signal strength information or raw targets based on better or stronger signal strength information can be weighted more heavily than, in comparison, raw targets based on weaker signal strength information. Advantageously, this makes the process more robust or more tolerant of interference.

As an aspect of the idea presented, in the case of passing by vehicles with small distances to a tracking sensor (in particular in the case of "across-the-road" and outgoing traffic), due to recursive filtering, measured values outside the permissible limits in traffic monitoring can occur. For this purpose, a non-recursive filter can be used, which summarizes all recorded measured values to the individual measuring points. In this case, it can be taken into account that each measurement time point due to the passage of a vehicle has a different angle to the sensor, which can be taken into account accordingly. In addition, reflections from different angles can be present at each measurement time since a real vehicle is not an idealized spotlight.

The device for determining at least one object parameter of a means of transport is designed to carry out or implement the steps of a variant of a method presented here in corresponding devices. Also by this embodiment of the invention in the form of a device, the object underlying the invention can be solved quickly and efficiently.

In the present case, a device can be understood as meaning an electrical device which processes sensor signals and outputs control and / or data signals in dependence thereon. The device may have an interface, which may be formed in hardware and / or software. In the case of a hardware-based embodiment, the interfaces can be part of a so-called system ASIC, for example, which contains a wide variety of functions of the device. However, it is also possible that the interfaces are their own integrated circuits or at least partially consist of discrete components. In a software training, the interfaces may be software modules that are present, for example, on a microcontroller in addition to other software modules.

A computer program product with program code which can be stored on a machine-readable carrier such as a semiconductor memory, a hard disk memory or an optical memory and is used to carry out the method according to one of the embodiments described above if the program product is installed on a computer or a device is also of advantage is performed.

The invention will now be described by way of example with reference to the accompanying drawings. Show it:

1 a schematic representation of a constructed on a roadside traffic monitoring system with an apparatus for determining at least one object parameter of a means of transport according to an embodiment of the present invention;

2 a schematic representation of a means of transport in a radar beam according to an embodiment of the present invention;

3a to 3c a schematic representation of a vehicle with a plurality of raw targets in a theoretical traffic monitoring situation at three successive points in time according to an embodiment of the present invention;

4a to 4c a representation of each vehicle with a plurality of raw targets in a traffic monitoring situation at three consecutive points in time according to an embodiment of the present invention;

5a to 5c a representation of each vehicle with a plurality of raw targets in a traffic monitoring situation at three consecutive points in time according to an embodiment of the present invention;

6a and 6b one projection each of a plurality of raw targets at a reference time in accordance with an embodiment of the present invention;

7 a schematic representation of a Nahbereichsradarmessung with appropriate name or naming of relevant parameters according to an embodiment of the present invention;

8th a representation of a system of equations for an object parameter determination in a Nahbereichsradarmessung according to an embodiment of the present invention;

9 a schematic representation of a Nahbereichsradarmessung with an apparatus for determining at least one object parameter of a means of transport according to an embodiment of the present invention;

10 a block diagram of an apparatus for determining at least one object parameter of a means of transport according to an embodiment of the present invention; and

11 a flowchart of a method for determining at least one object parameter of a means of transport according to an embodiment of the present invention.

In the following description of favorable embodiments of the present invention, the same or similar reference numerals are used for the elements shown in the various figures and similar acting, with a repeated description of these elements is omitted.

1 shows a schematic representation of a built on a roadside traffic monitoring system 100 with a device for determining at least one object parameter of a means of transport 102 according to an embodiment of the present invention. On a street 104 the transport is moving 102 in a direction of travel 106 , From the traffic monitoring system 100 goes a radar beam 108 out, the means of transport 102 detected. The radar beam 108 is partly from the transport 102 reflected and a reflected radar signal is from the traffic monitoring system 100 received and evaluated.

In the origin of the radar beam 108 is a coordinate system of the radar or traffic monitoring system 100 spanned, the abscissa with X _radar and the ordinate with Y _radar is called. At the origin of the means of transport 102 is a coordinate system of the means of transport 102 spanned, the abscissa with X _object and its ordinate with Y _object is called. The origin of the coordinate system of the means of transport 102 indicates a distance d _object to the origin of the traffic monitoring system 100 or the radar beam 108 on. Furthermore, it has an angle α _object to the abscissa X _radar . From a speed v _{object of} the means of transport 102 in the direction of travel 106 results in a measurable radial velocity v _radial .

The transport 102 is also called an object 102 designated. At the transport 102 is it in the in 1 illustrated embodiment to a vehicle 102 or a passenger car 102 ,

In other words, an object moves 102 through a radar beam 108 , The radar beam 108 is, as in the following figures, by two lateral boundary lines of the radar beam 108 shown. The radial velocity v _{radial of} the object 102 leads to a measurable frequency shift of the reflected radar signal. Is the direction of travel 106 of the object 102 known or measurable, may be made of the frequency shift, that is, the measurable radial velocity v _radially, the actual object speed v _object are determined.

2 shows a schematic representation of a means of transport 102 in a radar beam 108 according to an embodiment of the present invention. The radar beam 108 can be from a traffic monitoring system 100 as it is in 1 is shown, go out. In the origin of the radar beam 108 is a coordinate system of the radar or traffic monitoring system 100 spanned, the abscissa with X _radar and the ordinate with Y _radar is called. The radar beam 108 recorded a means of transport 102 , In the in 2 shown embodiment, it is the means of transport 102 around a vehicle 102 , Generally, the transport can 102 also as an object 102 be designated. The radar beam 108 will be at the rear of the vehicle 102 as well as on a longitudinal side of the vehicle 102 reflected. This results in so-called raw targets 210 ,

A raw target 210 represents a characteristic point of the means of transport determined by means of a radar measurement 102 , Each characteristic point is assigned one position at a time.

The object 102 reflects the radar signal at different points 108 , resulting in multiple raw targets 210 or reflection responses 210 result. A raw target 210 is also called an answer of the radar signal 108 or as a reflector 210 designated. Every raw target 210 or any reflection response 210 indicates an angle and a distance to the radar or the origin of the traffic monitoring system 100 on. Due to the different spatial location of the raw targets 210 have the raw targets 210 different measurable radial speeds. This cosine effect is exploited, for example, specifically in the classification.

3a to 3c each show a schematic representation of a vehicle 102 with a plurality of raw targets 210 in a theoretical traffic monitoring situation at three successive times t ₁ , t ₂ , t ₃ according to an embodiment of the present invention. This in 3 embodiment shown has similarities to those in the figures 1 and 2 shown embodiments. All three figures 3a to 3a show a vehicle 102 on a street 104 , where the vehicle 102 from a radar beam 108 is detected. Which is due to the reflection of the radar beam 108 resulting raw targets 210 are in the ideal case shown here relative to the position of the vehicle 102 always at the same position. 3a shows the vehicle 102 in the radar beam 108 at a first time t ₁ 3b shows the vehicle 102 in the radar beam 108 at a second time t ₂ and 3c shows the vehicle 102 in the radar beam 108 at a third time t ₃ . In the assumed scenario, the vehicle is moving 102 with a constant speed and a constant direction of movement on the road 104 , The raw targets 210 at the respective time t ₁ , t ₂ , t ₃ are shown as black dots. In 3b and 3c the raw targets of the preceding times t ₁ and t ₁ and t _{2 are shown} as small circles.

If the object is observed over several times (for example, t ₁ , t ₂ , t ₃ ), it can be determined by assigning the raw targets 210 among themselves the ride of the object 102 be recorded (tracking). From the journey, the angle can be estimated and thus the object speed can be determined. Knowing the likely direction of travel, known from the radar-road relationship, helps in the correct assignment of the raw targets 210 , For the assignment, the spatial position of the raw targets 210 and the estimated object speed used so far.

4a to 4c show a representation of each vehicle with a plurality of raw targets in a traffic monitoring situation at three consecutive times in a metrological limit situation according to an embodiment of the present invention. The representation largely corresponds to the representation in the figures 3a to 3c , with the difference that the raw targets 210 relative to Position of the vehicle 102 differ from each other at the three times. This can be due to environmental influences or a surface structure of the vehicle 102 be conditional.

Basis of tracking, that is tracking or tracking the moving vehicle 102 , is a correct assignment of the raw targets 210 among themselves. At the same time, a tracking requires time to correctly develop its model assumptions, such as driving direction or speed, or to correct initial assumptions. The tracking can decrease the actual direction of travel of heavy objects and the accuracy of measurement of the object speed decreases if the assumed (initial) direction of travel of an object 102 does not coincide with the actual (radar is not installed correctly, strong oblique drive of a vehicle 102 ), the object 102 very many raw targets that are very noisy in terms of their spatial position 210 owns (especially at close range, assignment of the raw targets is disturbed) and the observation period of the object 102 is very short (close range, outgoing traffic).

In practice, the assignment of the raw targets over several time steps t ₁ , t ₂ , t _{3 may be} erroneous. For example, due to limited or limited computing resources, only part of the raw targets will 210 used. In the approach presented here, all the raw targets become 210 all time steps t ₁ , t ₂ , t ₃ taken into account without assignment to each other in a closed mathematical model of the moving object. In this case, a plurality of first raw targets 210 at a first time t ₁ and a plurality of second raw targets 210 at a second time t ₂ and a mathematical model using the first raw targets 210 at the first time t ₁ and the second raw targets at the second time t ₂ , to determine at least one object parameter, such as the object speed.

5a to 5c show a representation of each vehicle 102 with a plurality of raw targets 210 in a traffic monitoring situation at three successive times t ₁ , t ₂ , t ₃ according to an exemplary embodiment of the present invention. The representation largely corresponds to the representation in the figures 3a to 3c , with the difference that a representation of the road and the radar beam is omitted and for a spatially fixed coordinate system is shown in which a position of the vehicle 102 and the raw targets 210 is shown. In the in the 5a to 5c In the embodiment shown, the radar beam is at the rear of the vehicle 102 as well as on a longitudinal side of the vehicle 102 reflected. This results in raw targets 210 at the rear of the vehicle 102 as well as on the left side of the vehicle 102 , The raw targets 102 are arranged approximately in an L-shape.

The figures 5a to 5c form the data basis for a projection of the raw targets recorded at a first time t ₁ , second time t ₂ and third time t ₃ 210 to a reference time, as in 6a and 6b is shown. Depending on the embodiment, the reference time is identical or different from one of the times t ₁ , t ₂ , t ₃ .

6a and 6b each show a projection of a majority of the raw targets 210 to a reference time according to an embodiment of the present invention. In the exemplary embodiments illustrated here, t ₂ is chosen as the reference time. As in the description too 4a to 4c is presented a mathematical model using the raw targets 210 solved to determine at least one object parameter such as the object speed. The underlying idea is the projection of all the raw targets 210 to an (arbitrary) time, such as the second time t ₂ chosen here. If the model assumptions such as speed, direction of travel etc. are correct, the raw targets are superimposed 210 the respective reflectors all at a time and form the vehicle shape according to or at least the two sides of the vehicle shape, which cause a reflection of the radar beam. 6a shows a correct projection, that is, the model assumptions or certain object parameters are correct. The raw targets 210 are in 6a almost congruent. In contrast, stands the 6b with a faulty projection. For example, the assumed direction of travel is wrong. The raw targets 210 are in 6b not congruent. The raw targets 210 are in 6b shifted linearly.

In the 6a and 6b projection shown dispenses with an assignment of the raw targets with each other. Advantageously, the projection also deals with large, noisy raw target quantities or point clouds. The projection is formulated via a system of equations. The goal of the system of equations is to determine object parameters that allow the most ideal projection possible.

In an ideal projection, all the raw targets are projected onto an "L", ie the vehicle surface. The long side and the short side of the "L" are perpendicular to each other. The scattering or the spatial distance of the raw targets from the axles is minimal. It is not mandatory that an "L" is formed. Depending on the reflective properties of the object, only the short or long side of the object can be used L's are captured. Another condition is that all raw targets, since they belong to the same object, can be projected onto a common object velocity.

Depending on the exemplary embodiment, the unknowns of the system of equations and thus object parameters are an object speed v ₀ , a direction error Δδ, an object acceleration a, a transverse distance D of the passage of the vehicle to the radar and a position of the vehicle on the drive axis at the projection time s ₀ . The mathematical model addressed here will be explained in more detail with reference to an exemplary embodiment in the following figures.

7 shows a schematic representation of a Nahbereichsradarmessung with corresponding designation or naming of relevant parameters according to an embodiment of the present invention. In doing so, the in 1 already illustrated geometric relationships and illustrates the errors occurring in practice. In a coordinate system of the radar system or the traffic control system, the abscissa with X _radar and the ordinate with Y _radar is called, moves a means of transport 102 with a speed v ₀ and a (real) direction of travel 106 , In this case, a direction of travel error Δδ occurs between a real direction of travel 106 of the means of transport 102 and an assumed direction of travel 720 of the means of transport 102 on. The negative installation angle δ is 202 ° in the illustrated embodiment. The negative installation angle δ results δ = -158 ° = 360 ° - 158 ° = 202 °

From the assumed direction of travel 720 The result is a world coordinate system whose abscissa is called X _world and whose ordinate is Y _world . The origin of the world coordinate system lies at the origin of the coordinate system of the radar system. The ordinate Y _world is perpendicular to the assumed direction of travel 720 ,

D = 3,7087 m bei 4 m Vorbeifahrt unter 22°.The position of the means of transport 102 on the drive axle at a time is denoted by s (t). The distance between the position s (t) and the intersection of the transverse distance D with a direction of travel straight line is denoted by l (t). The following results in the following relationships or equations.

D = 3.7087 m at 4 m Passage below 22 °.

Problem: D and l (t) are ideal for a point reflector only. In a real vehicle 102 There is still an offset on the vehicle surface. Therefore, approximately the following is assumed: Targets are distributed either on the side (that is, D = constant) or on the front or rear (that is, l (t0) = constant). For each measured value, an affiliation to the side or front / rear can be estimated. In this case, in one exemplary embodiment, the angle between the measurement and the direction of travel is used as a measure of the weighting.

8th shows a representation of a system of equations for an object parameter determination in a Nahbereichsradarmessung according to an embodiment of the present invention. From the description too 7 results in one embodiment, the in 8th illustrated equation system:

The system of equations can be divided into three parts T1, T2 and T3, whereby in the representation of the equation system only one line of a part is ever formulated as a representative. The first part T1 advantageously creates an optimal projection of all raw targets to an object speed v ₀ , taking account of acceleration a and direction error Δδ. The object shape ("L") is neglected. In the embodiment shown, the first part of the formula work is weighted by a factor of 1.

The second part T2 advantageously creates an optimal projection of all the raw targets on the long side of the L's for estimating the transverse distance. Due to the weighting w _s , raw targets of the short side lose weight. Since the position of the long side, the speed can hardly be estimated, it is omitted in this part of the formula.

The third part T3 advantageously provides an optimal projection of all raw targets on the short side of the L's for estimating all parameters except the transverse distance. Due to the weighting w _c , raw targets on the long side lose weight.

All raw targets flow into all parts T1, T2 and T3 of the formula with a different weight. The object velocity v ₀ (sought major unknown) can be determined solely with the first part and the third part. In its entirety, the equation system contributes to an optimal projection of the raw targets into the L-shape, taking into account that all the raw targets should have the same object velocity.

In a preferred embodiment, the weighting is enhanced by the signal strength to reward strong reflectors and to punish weak reflectors as well as interferences.

9 shows a schematic representation of a Nahbereichsradarmessung with a device for determining at least one object parameter of a means of transport 102 according to an embodiment of the present invention. As in 1 illustrated moves on a street 104 the transport 102 in a direction of travel 106 , From the traffic monitoring system 100 goes a radar beam 108 out, the means of transport 102 detected. The radar beam 108 is partly from the transport 102 reflected and a reflected radar signal is from the traffic monitoring system 100 received and evaluated. The radar beam 108 will be at the rear of the vehicle 102 as well as on a longitudinal side of the vehicle 102 reflected. An evaluation of the reflections becomes raw targets 210 certainly. In the origin of the radar beam 108 is a coordinate system of the radar or traffic monitoring system 100 spanned, the abscissa with X _radar and the ordinate with Y _radar is called.

To the means of transport 102 is a common area 930 Are defined. The lounge area 930 is also called Boundingbox 930 designated. The lounge area 930 describes an area in which with a predefined probability the means of transport 102 is staying. This is a simplified tracking of the means of transport 102 provided for an approximate object position determination. Using the approximate object position determination becomes the lounge area 930 defined and the raw targets 210 within the common area 930 detected. The raw targets 210 within the common area 930 are combined in one embodiment at any time in each case as a point cloud. Advantageous creates the definition of the residence area 930 the vehicle a filter to filter out any means of transport or objects that should not be detected at this time.

Will the vehicle 102 observed over a longer period of time, the assumed constants of the equation system can change. For example, the vehicle changes 102 the acceleration or the direction of travel. For this case, it makes sense only the raw targets 210 up to a limited past in the equation system or to let the coefficient matrix age. Here are the oldest raw targets 210 rolling through the latest raw targets 210 be replaced.

The Kalman filtering has a recursive behavior of the filtering, which leads to the fact that the past information of the received measured values is gradually lost. This is particularly problematic if only a few measuring cycles are available and thus the Kalman filter does not have enough time to adapt the measuring and process behavior adaptively with its built-in adaptation method. For this reason, a measuring method and evaluation method using a non-recursive filter is proposed here. This method is used in radar measurements for so-called over-the-road scenarios.

The following embodiment sets this method prototypically for an across-the-road measurement with 22 ° outflowing. As input values time-stamped radar measurements are assumed. From a radial velocity, an alpha measurement angle and the installation angle to the main travel direction, an estimation speed for the vehicle heading is calculated as follows: ν _Heading = ν _Rad / cos (α _Heading )

The angle (α _heading ) is calculated from the single measurement and the installation angle: α _Heading = α _Installation + α _Measurement

For left-hand traffic and outgoing measurement, the installation angle is typically α _Measurement = 158 ° = 180 ° - 22 ° = 2.75762 rad.

If the vehicle is slightly inclined, the result is a constant angular error Δδ, which causes the heading speeds to be misjudged. If the angle error Δδ is known, this angle error Δδ can be corrected. Assuming that a vehicle 102 obliquely with steady velocity v and without cornering, the result is the following relationship: ν _{Heading, skew} = ν _Rad / cos (α _{Heading, skew} ) = ν _Rad / cos (α _Heading + Δα _skew )

According to Taylor series development applies:

After insertion of the derivative follows from this as an approximation:

Converting the equation given above and comparing it with the straight-line equation y = mx + b yields:

The overdetermined system of equations can be solved with a linear regression.

y_i = a₀k_i,0 + a₁k_i,1 + a₂k_i,2 mit
a₀ = Δα_Skew
a₁ = ν_t0
a₂ = aIf you also want to estimate the acceleration at the same time, you have to include the age of the calculation:

y _i = a ₀ k _{i, 0} + a ₁ k _{i, 1} + a ₂ k _{i, 2} With
a ₀ = Δα _skew
a ₁ = ν _t0
a ₂ = a

If you capture the vehicle 102 Over several measuring cycles, one can let in assuming a constant speed all measurements in a compensation calculation. v (t) = v (t0) + a * (t - t0)

The illustrated nonrecursive tracking filter can be extended with a distance view. The following considerations are made from the point of view of the radar coordinate system. The aim is to describe a closed approach, in which all measurement parameters of the radar are taken into account.

A point x (t) / y (t) can be calculated from a measurement at time t as follows:

The following is assumed that the vehicle 102 with the preferred direction δ (for example, 22 °) through the radar beam. The transverse distance D to this passage is then determined according to the following formula:

The transverse distances D are always the same when driving straight ahead at a point reflector. This also applies to reflectors distributed over the longitudinal axis of the vehicle (ie reflectors on the vehicle side).

Assuming now that the real direction of travel δ deviates from the preferred direction of travel δ ₀ by Δδ, the following relationships can be established: δ = δ ₀ + Δδ -sin (δ) ≈ - (sin (δ ₀ ) + Δδ cos (δ ₀ )) cos (δ) ≈ cos (δ ₀ ) - Δδsin (δ ₀ ) D ₀ = -x (t) sinδ ₀ + y (t) cosδ ₀ D ≈ D ₀ + (x (t) cosδ ₀ + y (t) sinδ ₀ ) Δδ

Furthermore, for a reflector with continuous acceleration, the following general equation of motion applies: s (t) = s ₀ + ν ₀ t + 1 / 2at ²

The value s ₀ describes the deviation in the direction of travel with respect to the perpendicular of the passage to the origin of coordinates (speaks: distance footpoint to the observation point).

s(t) = x(t)cosδ + y(t)sinδ = x(t)cos(δ₀ + Δδ) + y(t)sin(δ₀ + Δδ) ⇒ x(t)cosδ₀ + y(t)sin(δ₀) + [–x(t)sinδ₀ + y(t)cosδ₀]Δδ = s₀ + ν₀t + 1 / 2at² The following relationship applies formulaically:

s (t) = x (t) cosδ + y (t) sinδ = x (t) cos (δ ₀ + Δδ) + y (t) sin (δ ₀ + Δδ) ⇒ x (t) cosδ ₀ + y ( t) sin (δ ₀ ) + [-x (t) sinδ ₀ + y (t) cosδ ₀ ] Δδ = s ₀ + ν ₀ t + 1 / 2at ²

The track can be better appreciated at the front and rear. The transverse distance is better appreciated when looking at the page. Therefore, the corresponding equations are provided with a weighting factor. Which results from the viewing angle to the estimated object. With these derivations an overdetermined equation system is set up:

Let α _{i be} the angle of measurement i, it is proposed to determine the weights w _s and w _c as follows:

10 shows a block diagram of a device 1040 for determining at least one object parameter 1042 a means of transport according to an embodiment of the present invention. The means of transport may be an embodiment of a means of transport described in the preceding figures 102 act. The device includes an interface 1044 for reading in a plurality of first raw targets 210 at a first time t ₁ and at least a plurality of second raw targets 210 at a second time t ₂ and a facility 1046 for solving a mathematical model using the first raw targets and the second raw targets, around the at least one object parameter 1042 to determine. A first raw target 210 and / or a second raw target 210 is a raw target 210 representing a characteristic point of the means of transport determined by a radar measurement. Each characteristic point is assigned a position at the first time and / or second time.

Optionally, the device has 1040 An institution 1048 for verification. The device 1048 for verification, the object parameter is formed 1042 using a projection of the plurality of raw targets at the respective time point to verify a reference time.

Advantageously, the device 1040 another interface 1050 for providing the at least one object parameter 1042 on. the interface 1050 for providing is designed to provide the object parameter as a signal representing the object parameter or digital data word.

Optionally, the device comprises 1040 An institution 1052 for estimating a traffic position using a radar signal 108 and a facility 1054 for tracking a lounge area 930 using the transport location. When the device 1040 An institution 1052 for treasure and a facility 1054 to track, so is the interface 1044 for reading trained, the first raw targets 210 and the second raw targets 210 using the lounge area 930 read.

the interface 1044 for reading is formed in one embodiment, to read in the majority of the first raw targets and the plurality of second raw targets signal strength information and assign the respective raw targets. In this case, the device is 1046 for solving, solve the mathematical model using the signal strength information.

11 shows a flowchart of a method for determining at least one object parameter of a means of transport according to an embodiment of the present invention. The method comprises a step 1160 reading in a plurality of first raw targets at a first time and at least a plurality of second raw targets at a second time and one step 1162 solving a mathematical model using the first raw targets and the second raw targets to determine the at least one object parameter. The first time differs from the second time. The raw targets are location-specific data determined by reflection from a radar signal at the respective assigned time. The object parameter includes, for example, information such as a speed, a heading direction, a heading error, a movement angle, an object acceleration, a position on a travel axis at a projection time, or a size of the vehicle.

In one embodiment, in step 1160 reading at least a plurality of third raw targets read at a third time. Then in step 1162 solving the mathematical model using the plurality of third raw targets, wherein the third time is different from the first time and the second time.

In order to solve the mathematical model, in one embodiment a least squares method is used. Alternatively, in one embodiment, the mathematical model is solved using a balance calculation, a regression, a fit, or a parameter estimate. The at least one object parameter is determined in one embodiment using an optimization method.

The mathematical model represents in step 1162 solving a projection of the plurality of raw targets at the respective time to a reference time.

Optionally, the method includes a step 1164 of verifying. In the step of verifying, the at least one object parameter is verified to a reference time using a projection of the plurality of raw targets at the respective time. Depending on the embodiment, the reference time corresponds to a time of a plurality of raw targets or is different from these.

Optionally, the method includes a step 1166 estimating a traffic position using a radar signal and a step 1168 tracking a location area using the transportation location. If the procedure is a step 1166 of the estimation and a step 1168 of tracing will be in step 1160 reading in the first raw targets and the second raw targets using the location area. In this case, the occupied area represents an area in which the means of transport is located with a predetermined probability at a predetermined time.

Optionally, at least one raw target of the raw targets is assigned information about an assumed speed or assumed direction of travel when the raw target in step 1160 read in. As an alternative, one in step 1166 estimating or step 1168 the tracking speed or assumed direction of travel are assigned to one or more raw targets.

In a favorable embodiment is in step 1160 of reading in the plurality of first raw targets and the plurality of second raw targets, signal strength information. The signal strength information is in the favorable embodiment in step 1162 of solving for solving the mathematical model.

The embodiments described and shown in the figures are chosen only by way of example. Different embodiments may be complete or in relation to individual features be combined. Also, an embodiment can be supplemented by features of another embodiment.

Furthermore, method steps according to the invention can be repeated as well as carried out in a sequence other than that described.

If an exemplary embodiment comprises a "and / or" link between a first feature and a second feature, then this is to be read so that the embodiment according to one embodiment, both the first feature and the second feature and according to another embodiment either only first feature or only the second feature.

LIST OF REFERENCE NUMBERS

100

Traffic monitoring system

102

Transportation, vehicle, object

104

Street

106

direction of travel

108

Radar beam, radar signal

210

Rohziel

t ₁

first time

t ₂

second time

t ₃

third time negative installation angle

Δδ

Heading error

D

Transverse distance of the passage of the vehicle to the radar

D '

transverse distance

l

longitudinal distance

t

Projecting time

a

object acceleration

s ₀

Position of the vehicle on the drive axle at the time of projection

v ₀

object speed

720

assumed direction of travel

w _s

weighting

w _c

weighting

930

lounge area

1040

contraption

1042

object parameters

1044

Interface for reading

1046

Device for releasing

1048

Device for verification

1050

Interface for providing

1052

Device to appreciate

1054

Device for tracking

1160

Step of reading in

1162

Step of solving

1164

Step of verifying

1166

Step of estimating

1168

Step of tracking

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• DE 102012107445 A1 [0004]
• EP 0286910 A1 [0005]

Claims (11)

Method for determining at least one object parameter ( 1042 ) of a means of transport ( 102 ), the method comprising the following steps: reading in ( 1160 ) a plurality of first raw targets ( 210 ) at a first time (t ₁ ) and at least a plurality of second raw targets ( 210 ) at a second time (t ₂ ), wherein a first raw target ( 210 ) and / or a second raw target ( 210 ) a raw target ( 210 ), which is a characteristic point of the means of transport determined by means of a radar measurement ( 102 ), wherein each characteristic point is assigned and / or assigned a position at the first time (t ₁ ) and / or second time (t ₂ ); and solve ( 1162 ) of a mathematical model using the first raw targets ( 210 ) and the second raw targets ( 210 ), the at least one object parameter ( 1042 ). Method according to claim 1, wherein in step ( 1162 ) of solving as object parameter ( 1042 ) at least one information about a speed (v) and / or a direction of travel ( 106 ) and / or a direction of travel error (Δδ) and / or a movement angle and / or an object acceleration (a) and / or a position (s ₀ ) on a drive axis at a projection time and / or a size of the means of transport ( 102 ) is determined. Method according to one of the preceding claims, in which in step ( 1160 ) of reading at least a plurality of third raw targets ( 210 ) is read in at a third point in time (t ₃ ), and wherein in step ( 1162 ) solving the mathematical model using a plurality of third raw targets ( 210 ), wherein the third time (t ₃ ) is different from the first time (t ₁ ) and / or the second time (t ₂ ). Method according to one of the preceding claims, in which in step ( 1162 ) of solving the mathematical model is solved using a least squares method. Method according to one of the preceding claims, in which in step ( 1162 ) of solving the mathematical model a projection of the plurality of raw targets ( 210 ) at the respective time (t ₁ , t ₂ , t ₃ ) to a reference time. Method according to one of the preceding claims, with a step ( 1164 ) of verifying, wherein in step ( 1164 ) of verifying the at least one object parameter ( 1042 ) using a projection of the plurality of raw targets ( 210 ) at the respective time (t ₁ , t ₂ , t ₃ ) is verified at a reference time. Method according to one of the preceding claims, with a step ( 1166 ) of estimating a traffic position using a radar signal ( 108 ) and in one step ( 1168 ) of tracking a location area ( 930 ) using the transport position, wherein in step ( 1160 ) of reading the first raw targets ( 210 ) and / or the second raw targets ( 210 ) using the residence area ( 930 ), in particular where the residence area ( 930 ) represents an area in which the means of transport ( 102 ) with a predetermined probability at a predetermined time. Method according to one of the preceding claims, in which in step ( 1160 ) of reading at least one raw target ( 210 ) of the raw targets ( 210 ) comprises information about an assumed speed and / or an assumed direction of travel. Method according to one of the preceding claims, wherein in step ( 1160 ) of reading at least one raw target ( 210 ) of the majority of the first raw targets ( 210 ) and / or the majority of the second raw targets ( 210 ) signal strength information is assignable and / or is assigned, and wherein in step ( 1162 ) solving the mathematical model using the signal strength information is solved. Contraption ( 1040 ) for determining at least one object parameter ( 1042 ) of a means of transport ( 102 ) configured to perform the steps of a method according to any one of claims 1 to 8 in corresponding devices. Computer program product with program code for carrying out the method according to one of claims 1 to 8, when the program product is stored on a device ( 1040 ) is performed.

Images