EP2404191A1

EP2404191A1 - Radar sensor having blindness recognition device

Info

Publication number: EP2404191A1
Application number: EP10704106A
Authority: EP
Inventors: Stefan Heilmann; Sonja Eder; Wolf Steffens; Goetz Kuehnle; Dirk Bechler
Original assignee: Robert Bosch GmbH
Current assignee: Robert Bosch GmbH
Priority date: 2009-03-02
Filing date: 2010-01-08
Publication date: 2012-01-11
Also published as: WO2010099988A1; DE102009001265A1; US20120050093A1; US9140780B2

Abstract

The invention relates to a radar sensor, comprising an associated evaluation and control unit (12) which has a measuring mode (MM) for locating radar targets and includes a blindness recognition device which is designed to recognize a blinding of the radar sensor based on the signals received by the radar sensor itself, characterized in that the evaluation and control device (12) has a test mode (TM) in which the radar sensor is controlled on the basis of parameters that are different from the parameters for the measuring mode (MM) and optimized for the blindness recognition device, and that the evaluation and control device (12) has a switching unit (24) for switching between the measuring mode (MM) and the test mode (TM).

Description

description

title

Radar sensor with blindness detection device

State of the art

The invention relates to a radar sensor with an associated evaluation and control device, which has a measurement mode for locating radar targets and includes a blindness detection device which is designed to detect blindness of the radar sensor based on the signals received from the radar sensor itself.

Motor vehicles are increasingly being equipped with driver assistance systems, for example for automatic adaptive cruise control (ACC) or warning of imminent collisions and, if appropriate, for initiating measures to avert the collision or to mitigate the collision sequences (PSS, Predictive Safety System). In these driver assistance systems, a radar sensor is used for detecting the traffic environment and in particular for locating other vehicles. Since the driver will generally rely on the functionality of the assistance system, it is necessary for safety reasons that functional impairments of the radar sensor are detected during operation and can be reported to the driver. For this reason, automotive radar sensors are equipped with a blindness detection device that, while the radar sensor is operating in the measurement mode, specifically evaluates the received radar signals to generate one or more blindness indicators indicative of degradation of the radar sensor - up to complete Blindness - point out.

An obvious criterion for blindness is that the radar sensor does not receive any radar echoes. However, this criterion alone is not very meaningful, since it can not be ruled out that actually no radar targets are present in the detection range of the sensor.

In general, however, a radar sensor installed in a vehicle always receives some reflections of uneven ground. These reflections are referred to as "bottom clutter". The absence of this Bodenclutters is a relatively reliable indication of blindness.

A possible cause of blindness of the radar sensor may be, for example, heavy rain, which reflects and dampens the radar radiation and thereby reduces the sensitivity of the sensor. However, the reflection of the radar radiation in turn leads to a detectable signal, the so-called rain clutter. Through targeted search for this rain clutter, it is therefore possible in principle to detect heavy rainfall and the associated impairment of the radar sensor.

Another possible cause of blindness is a dirt deposit or film of water on the surface of the radar lens or radome. Such a coating leads to the reflection and attenuation of the radar signal and thus to a loss of sensitivity. Again, however, in principle, that the reflection caused by the dirt deposit in turn represents a detectable signal by which the blindness could be detected. However, the distance between the dirt cover and the radar antenna is so small that it normally is outside the range of the distance, which can be monitored by means of the radar sensor.

Finally, a special form of blindness is the so-called angular blindness, which can be caused for example by an ice layer on the radar lens or the radome. Although such an ice layer does not lead to a significant reflection and attenuation of the radar, but caused due to refraction effects a directional deflection of the radar, resulting in angle-resolving radar sensors has the consequence that the angular data of the located objects are no longer reliable. The term "blindness detection" is to be understood in a comprehensive sense and is also intended to include such cases of angular blindness.

The object of the invention is to provide a radar sensor, especially for motor vehicles, which allows a more reliable blindness detection.

Disclosure of the invention

This object is achieved in that the evaluation and control device has a test mode in which the control of the radar sensor is based on parameters that are different from the parameters for the measurement mode and optimized for the blindness detection device, and that the evaluation and Control device has a switching device for switching between the measuring mode and the test mode.

Advantages of the invention

The invention is based on the consideration that the parameters determining the operation of the radar sensor are optimized with respect to the detection of objects such as vehicles in front and the like, and therefore not necessarily also for the detection of the various forms of blindness have to be optimal. Examples of these parameters include the frequency and power of the transmitted radar signal or, in the case of a Frequency Modulated Continuous Wave (FMCW) radar, where the transmitted frequency is ramp modulated, the modulation swing and the duration of the modulation ramp. According to the invention, the regular operating mode of the radar sensor, ie the measuring mode, is interrupted from time to time during the operation of the radar sensor, in the case of a driver assistance system while driving, and instead it is switched to a test mode in which the parameters with respect to the blindness detection are optimized, so that a more reliable blindness detection is possible.

The necessary interruptions of the measurement mode can be kept as short and timed so that they do not affect the seamless monitoring of the traffic environment.

In the dependent claims advantageous embodiments are given, which relate in particular to the detection of various blindness indicators and the respectively associated parameter optimizations. It goes without saying that different test modes can also be provided for the detection of different blindness indicators.

Brief description of the drawings

An embodiment of the invention is illustrated in the drawings and explained in more detail in the following description.

Show it:

Fig. 1 is a block diagram of a radar sensor according to the invention;

Fig. 2 is a frequency / time diagram for an FMCW radar; 3 and 4 are diagrams for explaining influences of changes of operating parameters of the radar sensor in a blindness detection test mode; and

Fig. 5 is a power / time diagram and a frequency / time diagram.

Embodiments of the invention

The radar sensor shown in FIG. 1 has a transmitting and receiving module 10 and an associated evaluation and control device 12, which supplies information about the traffic environment to a driver assistance system 14, for example a distance control system (ACC) in a motor vehicle. The transmitting and receiving module 10 has at least one antenna 16, which is fed by the evaluation and control device 12 with a signal to be transmitted and the received signals for evaluation to the evaluation and control device 12 passes. At a distance in front of the antenna, a so-called radome 18 is arranged, i. H. a cover designed to protect the antenna and the connected electronic components from the effects of the weather. In some cases, the function of the radome is also fulfilled by a radar lens that focuses the transmitted and received radar radiation.

The evaluation and control device 12 is formed by an electronic data processing system and comprises a Treiberf20 for controlling the transmitting and receiving module 10 and an evaluation part 22 which is operable in at least two different operating modes, namely a measurement mode MM and at least one test mode TM. A switching device 24 controls the switching between the operating modes.

The measurement mode MM is used to locate radar targets. In the example shown, the radar sensor is an FMCW radar in which the frequency f of the radar signal transmitted by the antenna 16 is ramped, such as is shown schematically in Fig. 2. There, the frequency f is plotted against the time t. The echo received by a radar target, such as a preceding vehicle, is mixed within the transmit and receive module 10 by means of a mixer, not shown, with the transmitted signal to provide an intermediate frequency signal whose frequency corresponds to the difference in frequency between the transmitted and received signals. This intermediate frequency signal is transmitted to the evaluation part 22 and is sampled there over the duration T of a frequency ramp. The time signal sampled in this way is then converted into a spectrum by Fast Fourier Transform (FFT). In this spectrum, each located object is distinguished in the form of a peak at a certain frequency.

Due to the ramp-shaped modulation of the transmitted signal, the frequency difference between received and transmitted signal (and thus the frequency of the intermediate frequency signal) depends (linearly) on the signal propagation time and consequently on the object distance. If the located object has a relative velocity other than zero with respect to the radar sensor, the Doppler effect results in an additional frequency shift. The frequency of a peak in the spectrum is thus dependent both on the distance and on the relative speed of the object and implicitly defines a relationship between the possible distances and relative speeds of the object. By evaluating spectra obtained on several modulation ramps with different slopes, it is then possible to assign a specific distance and a specific relative speed to each object unambiguously.

In practice, the antenna 16 usually consists of several juxtaposed patches, and by evaluating the amplitude and phase relationships between the signals received from the different patches, the azimuth angle of the located object can be determined. In measuring mode MM, the location data thus obtained are transmitted via the object, ie the distance Relative speed and the azimuth angle, reported to the driver assistance system 14.

As shown in FIG. 2, essential parameters for the operation of the radar sensor are the ramp duration T and the modulation stroke H of the modulation ramp. These parameters are optimized in measurement mode MM so that optimum resolution and sensitivity is achieved for objects in the traffic-related distance range.

Although it is possible in principle with the applicable for the measurement mode MM parameter settings to evaluate the signals of the radar sensor so that a blindness or a decrease in the sensitivity of the radar sensor can be determined by different indicators, but the choice of parameters for this blindness detection generally not optimal. For this reason, in the radar sensor described here by means of the switching device 24 is switched from the measurement mode MM to the test mode TM from time to time, and in this test mode, other parameters are used for the control of the transmitting and receiving module 10, namely parameters optimized for blindness detection. In test mode, blindness detection with increased reliability is thus possible. If a blindness or a significant impairment of the sensitivity of the radar sensor is detected, this is reported to the driver assistance system 14, which responds appropriately, for example, by disabling itself and issuing a corresponding warning to the driver.

The switching means 24 may be arranged to periodically switch between the measuring mode MM and the test mode TM. The duration of the test mode TM will be on the order of only a few milliseconds, so that despite the occasional interruptions of the measurement mode still a virtually complete monitoring of the traffic environment is possible. In the example shown, the driver assistance system 14 is also capable of the function of the switching device 24 influence, so that, for example, in critical situations, such as when a collision threatens to switch to the test mode can be prevented.

In the following, it will now be explained with reference to a few examples how the blindness recognition can be improved by a suitable choice of the operating parameters in the test mode TM.

If the radar sensor does not detect any preceding vehicles or other objects of comparable size, this could indicate a blindness of the radar sensor. However, it is also possible that in the detection range of the radar sensor simply no corresponding objects are present. One possible test of whether the radar sensor is blind can now be to temporarily increase the range of the radar sensor.

The range of object distances and relative velocities which can be detected with the aid of the radar sensor is limited in practice by the frequency range on which the spectrum formed from the intermediate frequency signal can be evaluated. This area is largely determined by the hardware of the evaluation part 22. The frequency of the object peaks in the spectrum increases proportionally with the object distance, so that as of a certain object distance, for example of the order of magnitude of about 200 m, the objects no longer lie within the evaluable range of the spectrum. By reducing the frequency deviation H (FIG. 2) with the same ramp duration, however, it can be achieved that the frequency of an object peak grows more slowly as the object distance increases, so that the radar sensor is then sensitive to objects at even greater distances. If no objects can be found even in this enlarged distance range, this is an indicator for a blindness of the radar sensor.

Even if there are no preceding vehicles or other objects in the detection range of the radar sensor, a signal is nevertheless present in a certain frequency range of the spectrum which is caused by reflections of the radar radiation on the road surface. Every little bump causes on the roadway that part of the radar radiation is reflected back to the transmitting and receiving module. The corresponding signals are recognizable in the spectrum as so-called "bottom clutter". If this bottom clutter can be detected in the test mode TM, this means that the sensor is not blinded.

However, the bottom clutter occurs in the spectrum only in a certain frequency range, which depends inter alia on the installation height of the transmitting and receiving module 10 in the vehicle and the vertical opening angle of the radar team. At high installation height, the radar radiation will hit the road at a greater distance, and the ground clutter will accordingly only occur at higher frequencies, but will then become weaker as the distance decreases due to the decrease in the signal intensity. Since the relative speed of the road surface coincides except for the sign with the vehicle's own velocity V, the bottom cutter is also subjected to a corresponding Doppler shift.

A measure of the strength of the bottom clutter and thus the remaining sensitivity of the radar sensor can be obtained by integrating in the spectrum over the frequency range in which bottom clutter is to be expected, but in which normally no "real" objects occur. For a reliable detection, it is advantageous if the bottom clutter is distributed over as large a part of the evaluable range of the spectrum.

FIG. 3 shows how the position of the bottom clutters in the spectrum can be influenced by varying the ramp duration T of the modulation ramp (with a constant frequency deviation) at different intrinsic speeds V of the vehicle. On the horizontal axis in Fig. 3, the numbers k of the so-called frequency bins are plotted, in which the spectrum of the intermediate frequency signal is divided. These numbers k correspond, apart from a normalization constant, to the various frequencies in the spectrum and in practice, for example, range from 0 to 511. FIG. 3 shows only the lower part of the spectrum. The evaluable region 26 of the spectrum is shown hatched in Fig. 3 and begins only at Bin no. 5. The Ground clutter region 28, ie the frequency range in which ground clutter can be detected (and integrated), is shown for different combinations of ramp duration T and airspeed V. If the ramp duration T is only 2 ms, at a speed V of 40 km / h the bottom clutter occurs only in a narrow frequency range at the lower end of the evaluable range 26. Even an increase in the speed V to 100 km / h leads only to a slight elongation and displacement of Bodenclutterbereiches. On the other hand, if the ramp duration T of the modulation ramp is increased to 10 ms, the bottom clutter region 28 is significantly spread, and more so, the higher the intrinsic velocity V is. For the detection of bottom clutter and thus for the verification that the radar sensor is not blind, therefore proves a relatively large ramp duration T advantageous. The longer ramp duration also has the advantage that a more stable result for the bottom clutter indicator is achieved by integration over a larger frequency interval. Furthermore, with the ramp duration T, the sample time in which the spectrum is recorded also increases, and this has a positive effect of reducing the noise, so that the bottom clutter and other evaluable signals receive a more favorable signal-to-noise ratio. At this low airspeed, for example at V = 0, it is not until the extension of the ramp duration T that an appreciable part of the bottom clutter region 28 lies within the evaluatable region 26 is reached.

Another blindness indicator is based on the detection of the so-called rain clutter, which is caused by reflection of the radar signal to raindrops, but only in a distance range of about 0 to 10 m, since at longer distances the weak radar returns of raindrops are no longer detectable.

If rain clutter is detected, this means that the intensity of the radar beam is greatly weakened by the rain and therefore the range and sensitivity of the radar sensor is significantly reduced. The detection of heavy rain is thus synonymous with the statement that the radar sensor is completely or partially blinded. 4 shows how the change in the ramp duration T at different speeds (V = O or 40 km / h) affects the rain-clutter area 30, ie the frequency range in which rain clutter can be detected. It can be seen that the rain-clutter region 30 is shifted to higher frequencies with increasing ramp duration T, but is not spread. That here, unlike the Bodenclutterbereich 28 no spread occurs, is because the radar rays reflected at the raindrops are mainly parallel to the direction of travel and therefore always subject to the same Doppler shift regardless of the distance of the drops, while the radar rays reflected from the ground at the bottom clutter or less run obliquely to the ground, so that the component, which is influenced by the Doppler shift, here depends on the respective distance.

Even with the rain chute, however, the extension of the ramp duration T has the positive effect that the rain clutter region 30 is shifted completely into the evaluable region 26 even at lower speeds.

Fig. 5 shows an example of a possible modulation scheme in the alternate operation of the radar sensor in the measurement mode MM and in the test mode TM. In the lower part of the diagram in FIG. 5, the modulated frequency f of the transmitted signal is plotted against time t. In the measuring mode, the modulation takes place alternately with rising and falling ramps with a constant, relatively short ramp duration. The modulation stroke is continuously varied in the example shown during the measurement mode, for example, to adjust the range of the radar sensor to the respective traffic situation.

In test mode TM, on the other hand, only falling frequency ramps are driven, with significantly longer ramp times and also (slightly) increased modulation stroke. The use of falling ramps is for blindness detection, in particular for the detection of rain clutter advantage, since then the distance and relative speed-dependent components of the frequency shift so cooperate, that the rain clutter area is moved to the evaluable area.

In addition, the modulation period T is kept constant in the test mode, which simplifies the evaluation of the received signals.

In the upper diagram in Fig. 5, the radiated from the antenna 16 power P is plotted against time. It can be seen that in the test mode TM this power is varied between at least two values, in the example shown between a higher value on a modulation ramp and a lower value on a different modulation ramp. This power variation allows even more sensitive and reliable detection of conditions in which the sensitivity of the radar sensor is degraded. Factors that lower sensitivity will have a greater impact on performance. A comparison between the signals that are received in the test mode on the one hand at high power and the other at reduced power, therefore, makes the blindness indicators more prominent and also makes the blindness detection robust against temperature and aging effects that affect the behavior of the transmitting and receiving electronics.

If the modulation stroke H is significantly increased in the test mode, cases of blindness attributable to a dirt deposit or water film on the radome 18 (FIG. 1) can also be detected. Namely, such a coating leads to the fact that a part of the radiation emitted by the antenna 16 is already reflected at the radome 18. This deposit can therefore be considered as a "radar target" which is located at an extremely short distance from the antenna 16. With the normal evaluation method, which is based on an analysis of the spectrum of the intermediate frequency signal, however, objects at such close intervals are not detectable. An increase in the modulation H in these cases brings the frequency of the intermediate frequency signal closer to the evaluable range. If one does not evaluate the spectrum, but directly the time signal from which the spectrum is obtained, the radiation reflected at the radome 18 results in a sinusoidal signal with a characteristic frequency (very small corresponding to the small distance). With a sufficiently large modulation H, this frequency is increased so that the sinusoidal signal within the relatively small time window (of the duration of the modulation ramp T) is recognizable as a sine wave with the frequency characteristic for the Radomabstand, so that reflections can be directly detected at Radombelag ,

Claims

claims

A radar sensor having associated evaluation and control means (12) having a measurement mode (MM) for locating radar targets and including blindness detection means adapted to detect blindness of the radar sensor based on the signals received from the radar sensor itself, characterized in that the evaluation and control device (12) has a test mode (TM) in which the control of the radar sensor is performed on the basis of parameters different from the parameters for the measurement mode (MM) and optimized for the blindness recognition device, and in that the evaluation and control device has a switching device (24) for switching between the measuring mode (MM) and the test mode (TM).

2. Radar sensor according to claim 1, characterized in that the evaluation and control device (12) is adapted to ramp the frequency of the transmitted radar signal, and that in the test mode (TM) the ramp duration (T) compared to the ramp duration in the measuring mode ( MM) is changed.

3. Radar sensor according to claim 1 or 2, characterized in that the evaluation and control device (12) is adapted to ramp the frequency of the transmitted radar signal, and that in the test mode (TM) of the modulation (H) over the Modulationshub in Measuring mode (M) is changed.

4. Radar sensor according to one of the preceding claims, characterized in that the evaluation and control device (12) is adapted to control the radar sensor in the test mode (TM) with a different transmission power (P) than in the measuring mode (MM).

5. Radar sensor according to claim 4, characterized in that the evaluation and control device (12) is adapted to vary the transmission power (P) in the test mode (TM).