WO2019180062A1

WO2019180062A1 - Method for detecting an object

Info

Publication number: WO2019180062A1
Application number: PCT/EP2019/056916
Authority: WO
Inventors: Jan Niklas Caspers; Oliver Kern
Original assignee: Robert Bosch Gmbh
Priority date: 2018-03-20
Filing date: 2019-03-20
Publication date: 2019-09-26
Also published as: DE102018204237A1

Abstract

The invention relates to a method for detecting an object by means of a sensor assembly, comprising the steps: first scanning of a predefined angular range by means of an optical phased-array transmitting device of the sensor assembly by emitting electromagnetic radiation from emitters of the optical phased-array transmitting device; identification of objects on the basis of electromagnetic radiation received by the sensor assembly; and verification of the identified objects on the basis of received electromagnetic radiation of a second scanning of the predefined angular range by emitting electromagnetic radiation, the phase of the electromagnetic radiation emitted by each of the emitters being shifted, the mean across all shifted phases being essentially less than or equal to a threshold value that results from 2*pi divided by the number of emitters of the phased-array transmitting device.

Description

Method for detecting an object

Technical area

The invention relates to a method for detecting an object by means of an optical sensor arrangement.

The invention further relates to an optical sensor arrangement.

Although the present invention is generally applicable to any objects, the present invention will be explained in terms of environment recognition of objects in the field of vehicles.

Although the present invention is generally applicable to any optical sensor arrangement, the present invention will be explained with reference to optical sensor arrangements in the form of LIDAR sensors.

State of the art

From DE 10 2006 045 1 15 A1 a target tracking and sensor fusion system for observing a state of at least one object during a cycle has become known. The system includes a plurality of sensors and a controller communicatively coupled to the sensors and configured to more accurately estimate the condition based on sensor fusion.

Disclosure of the invention

In one embodiment, the invention provides methods for detecting an object by means of an optical sensor arrangement, comprising the steps of

First scanning a predetermined angular range by means of an optical phased array transmitting device of the sensor arrangement Emitting electromagnetic radiation from emitters of the optical phased array transmitter,

Identifying objects based on electromagnetic radiation received by the sensor array, and

Verifying (S4) the identified objects based on received electromagnetic radiation from a second scanning (S3) of the predetermined angular range by emission of electromagnetic radiation, wherein the phase of each of the emitters (5) emitted electromagnetic radiation is shifted, the average over all shifted phases is substantially less than or equal to a threshold, which results from 2 ^* Pi divided by the number of emitters (5) of the phased array transmitting device (2).

In a further embodiment, the invention provides an optical sensor arrangement in the form of a LIDAR system or the like comprising

a phased array optical transmitting device which is formed

performing a first scanning of a predetermined angular range by emitting electromagnetic radiation from emitters of the phased array optical transmitting device, and

performing a second scanning of the predetermined angular range by emitting electromagnetic radiation, shifting the phase of the respective electromagnetic radiation emitted by the emitters, the average over all shifted phases being substantially equal to or less than a threshold divided by 2 ^* Pi by the number of emitters of the phased array transmitting device results.

and an evaluation device which is designed

Identify objects based on the sensor array of received electromagnetic radiation, and

- Verify identified objects based on the electromagnetic radiation received during the second scan.

One of the advantages of this is that the noise that is applied to the phase of the emitter makes it easier and more reliable to identify false objects than a false positive object, ie an object falsely identified in a particular position, angle or condition such as speed or the like, the received electromagnetic radiation, and thus the signal-to-noise ratio, would increase, or at least remain constant, and not decrease as expected.

Further features, advantages and further embodiments of the invention are described below or become apparent:

According to an advantageous development, the first scanning is carried out by means of a radiation profile of the emitter, which emits electromagnetic radiation in the predetermined angular range with a main emission direction, wherein light in the remaining angular range - background radiation - homogeneously distributed and emitted. In other words, the background radiation has no significant maxima. In this way, irrespective of the radiation direction, it is substantially equally likely that an object will be detected in a false positive manner, so that overall the detection of objects is more reliable. The term "homogeneous" here is to be understood as meaning a sufficient degree of uniform distribution of the background radiation of the emitters, in particular a scattering by an intensity mean value of +/- 5 dB at least over half of the angular range.

According to a further advantageous development, the emitters are randomly shifted relative to one another for the purpose of providing the homogeneously emitted background radiation. This can be provided in a particularly simple manner homogeneous or homogeneous as possible background radiation.

According to a further advantageous embodiment, the emitters are shifted by a maximum of 30% of the original distance to each other. In this way, the background radiation can be provided sufficiently homogeneously in a simple and reliable manner at the same time.

According to a further advantageous embodiment of the invention, the shifting of the phases of the emitter is random. This makes it easy to provide a shifting of the phases. An elaborate control of the phases of each emitter to each other can be dispensed with. According to a further advantageous development, the maximum shift of the phases is% ^* Pi. One of the advantages achieved with this is that a sufficient maximum shift of the phases can thus be provided in order to increase the background radiation.

According to a further advantageous development, the identification of objects comprises a comparison of the intensity of the receiving electromagnetic radiation with a determined distance of an object. This further increases the reliability in the detection of objects. In other words, it is checked here whether a measured signal of an object matches the measured distance of the detected object. Objects that are very close, ie at a short distance to the sensor array, also have a correspondingly stronger measured signal.

According to a further advantageous development, the first scan is repeated M times and the verification is only performed if at least N times an object has been identified during the M-fold first scan. Thus, the overall efficiency is improved, for example, increases the readiness of a sensor array, which is operated by means of the method, because not every goal, but only suspicious targets are checked.

According to a further advantageous development, the ratio of M to N is at least 10. This provides sufficient reliability before objects are verified in a further step.

According to a further advantageous development, the threshold value is essentially 0. In this way, a particularly reliable detection of an object is achieved.

Further important features and advantages of the invention will become apparent from the dependent claims, from the drawings, and from associated figure description with reference to the drawings.

It is understood that the features mentioned above and those yet to be explained below not only in the combination indicated, but can also be used in other combinations or in isolation, without departing from the scope of the present invention.

Preferred embodiments and embodiments of the invention are illustrated in the drawings and will be explained in more detail in the following description, wherein like reference numerals refer to the same or similar or functionally identical components or elements.

Brief description of the drawings

It shows

1 shows a sensor arrangement according to an embodiment of the present invention;

Figure 2 steps of a method according to an embodiment of the present invention;

3a, 3b each show a light distribution of a periodic (Figure 3a) and a non-periodic (Figure 3b) optical phase array according to an embodiment of the present invention; and

4a, 4b each have a radiation characteristic of a phase array for two different deflection directions with phase noise in logarithmic (Figure 4a) and linear (Figure 4b) representation.

Embodiments of the invention

FIG. 1 shows a sensor arrangement according to an embodiment of the present invention.

FIG. 1 shows in schematic form a sensor arrangement 1. The sensor arrangement 1 comprises a phase array 2 for transmission and reception electromagnetic radiation, for example an optical signal, with a plurality of emitters 5, for example 100 or 1000 individual emitters. The phase array 2 is connected to an evaluation device 6, for example in the form of a computer or the like. If electromagnetic radiation is now sent by the phase array 2, more precisely by its emitter 5, into a specific angular range, for example, the angular range is scanned by scanning the corresponding radiation in a grid pattern over the entire angular range, objects 10, 11, for example, can be detected. In the example of Figure 1, only the vehicle 10 is a real target, whereas the traffic sign 1 1 is a so-called false positive object.

In general, false-positive objects here are points which the sensor arrangement 1 identifies and outputs in a point cloud composed of the signals of the received electromagnetic radiation corresponding to the respective directions and distances, which in reality, however, do not exist, at least at this determined position. In other words, the sensor arrangement 1 detects an object 1 1, but in fact there is no object there. In phase arrays are shown by a relatively large optical power, which is not emitted in the main beam / main lobe or the respective scanning direction, but in the background, objects at a wrong angle, shown in Figure 1. Here, the sensor arrangement 1 detects an area where a vehicle 10 is located at a greater distance, here for example> 100 m. At the same time, another object, for example a shield 1, is located in the visible region of the sensor arrangement 1, that is to say in the scanned angular range, but not where the main lobe / current emission direction / scanning beam is aligned. Although this shield 11 is located at a wrong angle, but only a short distance, here a few meters, for example, 1 m from the sensor assembly. 1

Due to the greater distance in the forward direction, the backscattered light from the vehicle 10, for example, has a power reduced by up to 40 dB compared to the backscattered light from the shield 1 1

assuming the same reflectivity of the targets / objects.

On the other hand, the phase array acts like a spatial filter in the forward direction. For example, a phase array has a suppression of 20-30 dB from the main lobe / main emission direction with respect to the background, or only a few dB between main lobe and sidelobes. This filtering would take place twice in a coaxial arrangement of transmitting and receiving device of the sensor arrangement 1, both in the transmitting and receiving direction, so that the suppression of arbitrary targets would not be in the target direction, ie the main emission direction, but in the general angular range about 40-60 dB. This difference in performance is comparable to the suppression by the range difference. The sensor assembly 1, assuming that it is currently scanning in a first direction, would also identify a target at a very short distance (1 m), which in reality is in a different direction. However, the sensor arrangement 1 can not distinguish this. This object does not exist where the sensor arrangement detects it and is thus a false positive object.

FIG. 2 shows steps of a method according to an embodiment of the present invention.

In a first step S1, a first scanning of a predetermined angular range is effected by means of an optical phased array transmission device of the sensor arrangement by emission of electromagnetic radiation from emitters of the optical phased array transmission device.

In a second step S2, objects are identified on the basis of electromagnetic radiation received by the sensor arrangement.

In a third step S3, a second scanning of the predetermined angular range by emission of electromagnetic radiation, wherein the phase of each emitted by the emitters electromagnetic radiation is shifted, wherein the average over all shifted phases in the Is substantially less than or equal to a threshold resulting from 2 ^* Pi divided by the number of emitters of the phased array transmitter.

In a fourth step S4, the identified objects are verified on the basis of received electromagnetic radiation from the second scanning according to step S3.

In addition, the most uniform possible spatial distribution of the background radiation of the phase array 2 can be provided. In other words, the background radiation is not suppressed, but evenly distributed. If a lateral object, such as the shield 11 according to FIG. 1, is detected, then this is always detected, regardless of where the sensor arrangement 1 directs its main emission direction. This false positive target is then identified with otherwise identical characteristics, such as range, but also speed or signal strength. If the sensor arrangement 1 has now always detected an object in a short distance with the same properties but at different angles after several directions, it marks this object and checks it. In this check, the electrical control of the phase shifter is provided with a noise. This noise must be free of expectation, so the mean corresponds to a technical zero, but in detail can be several p. As a result of the noise, the efficiency of the phase array is intentionally worsened, so that less light goes into the main lobe or main emission direction and more into the background radiation. Although a "technical zero" is advantageous as an average, it is sufficient if the average is substantially less than or equal to a threshold resulting from 2 ^* Pi divided by the number of emitters of the phased array transmitter.

If the signal now decreases at the marked point 3, this is assumed to be a true, ie correctly identified, object 10 and the sensor arrangement can continue to scan. If, however, the signal at the marked point / object 11 increases or remains stable, then this is a false positive target / object.

3a, 3b each show a light distribution of a periodic (FIG. 3a) and a non-periodic (FIG. 3b) optical phase array according to an embodiment of the present invention. FIG. 3 a shows a periodic phase array with two deflection directions (0 °, 10 °). Also shown are sidelobes at the angles +/- 50 ° and +/- 38 °. The emitters 5 of the phase array have a spacing of 2 μm and the phase array emits light at a wavelength of 1.55 μm. In Figure 3b, the emitters are now randomly shifted by a maximum of 30% of the original distance to each other. This results in a sufficiently uniform distribution of the background light, as shown in Figure 3b. Over the entire field of view of 120 ° no sidelobes exist anymore. Due to the relatively uniform distribution over the entire angular range, regardless of where the main lobe of the transmitting device is directed, it is equally probable that it detects a false positive target. If a target with the same characteristics, in particular distance, velocity, signal occurs under different directions, this is identified several times. If a target is identified N times within M scans, it will be marked as above. The mark here means that this target is now verified, so it is checked that it is not a false positive target. For the purpose of verification, a noise is deliberately added to the phase setting of the individual phase shifters in the phase array 2, ie the phase adjustment is shifted at each phase shifter by a random value, the mean value of the shift being zero. The size of the shift can be adjusted.

For example, FIGS. 4a, 4b show the effect of a maximum shift of +/- 0.75p per phase shifter. By this random shift so more light is scattered in the background and less in the main lobe. The signal strength for a real target / object is reduced. Thus, if the marked object exists in the "viewing direction" of the sensor array, then the signal strength will decrease. If the marked target is a false positive target, then the signal strength is constant or even greater. The noise on the phase shifters / emitters makes it possible to reliably differentiate real from false positive targets. In addition to the steps described above, a pre-mark may be used based on whether the signal being measured matches the measured distance of the object. Objects that are very close generally also have a stronger signal. 4a, 4b each show a radiation characteristic of a phase array for two different deflection directions with phase noise in logarithmic (Figure 4a) and linear (Figure 4b) representation. FIGS. 4a, 4b show the effect of the phase noise on the deflection characteristic of a phase array for two different deflection directions, once 0 °, once 10 ° (deflection angle reference 20). The intensity 21 is plotted on the vertical axis. In the logarithmic representation in FIG. 4 a, an increase of the bottommost light in the region of the emission angle 20 around the 5 ° is shown with reference to the arrow 30. The reduction in the two

Viewing directions, ie once 0 °, once 10 °, is shown by the two arrows 31. This can be seen more clearly in the linear representation of FIG. 4b.

In summary, at least one of the embodiments has at least one of the following advantages:

• Reliable identification of wrong targets.

• Cost-effective production.

• Reduced sensitivity to mechanical shocks.

• Compact design.

Although the present invention has been described with reference to preferred embodiments, it is not limited thereto, but can be modified in a variety of ways.

Claims

1. A method for detecting an object (10, 11) by means of an optical sensor arrangement (1),

comprising the steps

First scanning (S1) of a predetermined angle range by means of an optical phased array transmitting device (2) of the sensor arrangement (1) by emitting electromagnetic radiation (3, 4) from emitters (5) of the optical phased array transmitting device (2),

Identifying (S2) objects (10, 11) on the basis of electromagnetic radiation received by the sensor arrangement (1), and

Verifying (S4) the identified objects (10, 11) on the basis of received electromagnetic radiation from a second scanning (S3) of the predetermined angular range by emission of electromagnetic radiation, wherein the phase of the respective emitted by the emitters (5) electromagnetic radiation where the average over all shifted phases is substantially less than or equal to a threshold resulting from 2 ^* Pi divided by the number of emitters (5) of the phased array transmitter (2).

2. The method according to claim 1, wherein the first scanning (S1) by means of a radiation profile of the emitter (5), which emits electromagnetic radiation in the predetermined angular range with a main emission direction, wherein light in the remaining angular range - background radiation - homogeneously distributed and emitted becomes.

3. The method according to claim 2, wherein the emitters (5) for providing a homogeneously radiated background radiation are randomly shifted from each other.

4. The method according to claim 3, wherein the emitter (5) are shifted by a maximum of 30% of the original distance to each other.

5. The method according to any one of claims 1 -4, wherein the shifting of the phases of the emitter (5) takes place at random.

6. The method according to claim 5, wherein maximum displacement of the phases ^Z Ä ^* Pi.

The method of any of claims 1-6, wherein identifying objects (10, 11) comprises balancing the strength of the received electromagnetic radiation with a determined distance of an object (10, 11).

8. The method according to any one of claims 1-7, wherein the first scanning (S1) is repeated M times and the verification (S3) is performed only when at least N times an object (10, 1 1) during the M first scans was identified.

The method of claim 7, wherein the ratio of M to N is at least 10.

10. The method of claim 1, wherein the threshold is substantially zero.

1 1. Optical sensor assembly (1) in the form of a LIDAR system or the like the same

a phased array optical transmitting device (2) which is formed

performing a first scanning (S1) of a predetermined angular range by emitting electromagnetic radiation from emitters (5) of the phased array optical transmitting means (2), and second scanning (S3) the predetermined angular range by emitting electromagnetic radiation (3, 4) ), whereby the phase of the respective electromagnetic radiation emitted by the emitters (5) is shifted, wherein the average over all shifted phases is substantially less than or equal to a threshold value consisting of 2 ^* Pi divided by the number of emitters (5 ) gives the phased array transmitter (2)

and an evaluation device (6), which is designed To identify objects (10, 11) by means of electromagnetic radiation (3, 4) received by the sensor arrangement (1), and

To verify identified objects (10, 11) on the basis of the electromagnetic radiation received in the second scanning (S3).