DE202016104285U1 - Optoelectronic sensor for detecting an object - Google Patents

Abstract

An optoelectronic sensor (10) for detecting an object (20) in a surveillance area (18) having a light emitter (12) for emitting a light signal (16) into the surveillance area (18) and a light receiver (26) comprising a plurality of avalanche photodiode elements (28) for detecting received light from the monitoring area (18), each biased with a bias voltage above a breakdown voltage and thus operated in a Geiger mode, wherein the light receiver (26) is digital and thus allows individual or groups avalanche photodiode elements (28, 28a), and a control and evaluation unit (30) adapted to evaluate a receive signal of the light receiver (26) and to activate those avalanche photodiode elements (28a) where a light spot (44) of the object (20) remitted or reflected light signal (22) is expected, characterized gekennzei that the light spot (44) is elongate.

Description

The invention relates to an optoelectronic sensor for detecting an object in a surveillance area according to the preamble of claim 1.

Optoelectronic sensors are available in a broad spectrum, ranging from one-dimensional photoelectric sensors and light sensors to laser scanners and cameras. In addition to pure object detection, a distance to the object is also determined in distance-measuring systems. Distance sensors based on the time-of-flight principle measure the propagation time of a light signal that corresponds to the distance over the speed of light. Conventionally, a distinction is made between pulse-based and phase-based measurements. In a pulse transit time method, a short light pulse is emitted and the time to receive a remission or reflection of the light pulse is measured. Alternatively, in a phase method, transmitted light is amplitude modulated and a phase shift between transmitted and received light is determined, wherein the phase shift is also a measure of the light transit time. However, the boundary between the two methods can not always be drawn sharply, because, for example, in the case of complex pulse patterns, a pulse transit time method is more similar to a phase method than to a classical single pulse measurement.

In order to be able to detect even low reception intensities, avalanche photodiodes are conventionally used in some optoelectronic sensors (APD, avalanche photo diode). The incident light triggers a controlled avalanche breakdown. As a result, the charge carriers generated by incident photons are multiplied, and there is a photocurrent, which is proportional to the light receiving intensity, but much larger than a simple PIN diode.

An even greater sensitivity is achieved with avalanche photodiodes, which are operated in the so-called Geiger mode (SPAD, Single Photon Avalanche Diode). Here, the avalanche photodiode is biased above the breakdown voltage, so that even a single, released by a single photon charge carrier can trigger a no longer controlled avalanche, which then recruits all available charge carriers due to the high field strength. Then the avalanche comes to a standstill (passive quenching) and is no longer available for detection for a certain dead time. Alternatively, it is also known to detect and extinguish the avalanche from the outside (active quenching).

A SPAD thus counts as a Geiger counter individual events. Not only are SPADs highly sensitive, but they are also comparatively inexpensive and efficient to integrate into silicon semiconductors. Next, they can then be integrated with little effort on a printed circuit board. A special feature is the fact that even a minimal disturbance event, such as an extraneous light photon or dark noise, generates the same maximum received signal as a useful light signal. In order to counteract these effects, in practice several SPADs are evaluated together.

There are matrix arrays of SPADs that not only provide a sum signal but, similar to a pixel-resolved image sensor, in principle allow the reading of individual SPADs followed by direct quantization. This is referred to herein as a digital SPAD matrix and allows to capture the information of subregions or subgroups.

The use of such a fast and highly sensitive receiving element, in which the information of individual pixels can be processed, requires in itself novel optical concepts with which the performance of the overall system can be optimized. Conventionally, however, only systems are used whose transmitted light using transmit and receive lenses with spherical or aspherical surfaces produces round spots of light in the receiver plane, wherein the light spot also has an inhomogeneous irradiance.

It is therefore an object of the invention to improve the performance of an optoelectronic sensor with a digital SPAD matrix.

The object is achieved by an optoelectronic sensor for detecting an object in a surveillance area according to claim 1. The sensor has a light transmitter for emitting a light signal as well as a light receiver a plurality of operated in Geiger mode avalanche photodiodes or SPADs, which are preferably arranged as pixels to a row or matrix. It is possible, as in a digital SPAD matrix, to selectively activate or deactivate individual avalanche photodiode elements or groups thereof. Only the activated avalanche photodiode elements actually contribute to a measurement. Activation can actually switch individual pixels, so that avalanche breakdown does not even occur in the non-activated pixels, but it is also possible to dispense with reading out non-activated pixels or to disregard their information in the evaluation. An evaluation unit of the sensor activates those pixels that are hit by the light spot, that from the Surveillance area generates the returning light signal on the light receiver, and preferably deactivates the remaining pixels. This selection can, for example, be predicted, parameterized, taught-in or tracked by means of a suitable algorithm. The invention is based on the basic idea of adapting the geometry of the light spot, in particular by means of an elongated light spot in the receiver plane. For this purpose, a corresponding elongate cross-section of the light signal is provided, which arrives at the light receiver.

The invention has the advantage that the elongated light spot is adapted to the specific requirements of a SPAD matrix and typical applications of the sensor. This prevents misdetections, for example, from additional light reflections or diffraction patterns, is electronic and thus much easier to implement than a narrow mechanical aperture. By adjusting the light spot and the activated area, it is possible to increase the irradiance, thus increasing the signal strength and ultimately the range. By deliberately taking into account avalanche photodiode elements which receive useful light, a much better signal-to-noise ratio results. The insensitivity to ambient light is further increased, because the field of view can be set very narrow. Overall, the performance of the sensor increases in terms of range, response time, external light compatibility and pinpoint switching perpendicular to the beam axis significantly.

The light spot preferably has an aspect ratio of at least 2: 1, possibly also 3: 1 or more. Assuming a SPAD matrix with the same pixel geometry in the horizontal and vertical directions, the light spot covers at least twice as many pixels in its longitudinal direction as in the transverse direction. In this case, preferably only one, two or at least very few pixels are provided in the transverse direction. The light spot is advantageously oriented with its longitudinal direction in the column or row direction of the SPAD matrix, but in principle can also be diagonally rotated.

The light spot preferably has a rectangular shape. This allows the best fit to the SPAD or pixel structure of the light receiver. Nevertheless, other geometries such as an ellipse are conceivable, which may be more easily accessible.

The sensor is preferably a distance-measuring sensor, in which the control and evaluation unit is designed to determine a distance of the object from a light transit time between emission of the light signal and reception of the light signal reflected or reflected on the object. When measuring distance, the high sensitivity of a light receiver with avalanche photodiode elements in Geiger mode is particularly advantageous. The invention provides for improved utilization of the high sensitivity. The light transit time can be measured, for example, by TDCs (time-to-digital converter). TDCs can be directly monolithically integrated in a crystal of the light receiver. By combining avalanche photodiode element and TDC, an incident location and time information of the incident photons can be obtained. Herein, in some embodiments, one TDC is provided per pixel, but preferably the association of a TDC to a pixel or group of pixels is adjustable so that fewer circuit elements are needed and the chip area is minimized. The assignment of TDCs is one way to activate avalanche photodiode elements, since an avalanche photodiode element without such assignment does not contribute to the transit time measurement.

The emitted light signal preferably has a light pulse. The sensor thus measures distances according to the pulse method. Even more complicated forms such as double pulses or even pulse codes are conceivable. It is also possible for a plurality of light pulses to be transmitted one after the other, received, and the respective individual results jointly evaluated statistically, for example in a pulse averaging process. Alternatively, a phase method is conceivable.

The light receiver preferably has a receiving optical system which generates the elongated shape of the light spot. By adapting the receiving optics, the light spot can be further optimized. An elongated light spot is also produced by a line arrangement of a plurality of light sources or a corresponding transmission optical system, so that this can be used as an alternative or in addition to a receiving optical system for the generation of the elongate shape.

Preferably, the receiving optics homogenize the light spot. As a result, errors, in particular runtime errors, are avoided by intensity fluctuations. The constant irradiance within the light spot leads to a better utilization of the individual avalanche photodiode elements and thus a higher signal. This increases the range, and more efficient statistics improve the expected response time until a reading is available. The dead time of the avalanche photodiode elements does not have such a strong effect because of the reduced intensity fluctuations, more useful light photons are detected overall than in the case of an inhomogeneous light spot. Also, the variance in the arrival time of the first signal photon decreases, and therefore, in a time of flight measurement, the object distance is more accurately determined. A Transmitting optics can support the homogenization or instead of the receiving optics afford.

The receiving optics has at least one free-form surface. In this case, the freeform surface can be realized on a lens or a mirror. Freeform means that the geometry can not be described as an asphere or its body of revolution, and this allows a precise adjustment of the light spot. The lens or the mirror with the free-form surface can also be combined with a further optical element, in particular likewise with a free-form surface or else with an asphere or cylindrical lens or a correspondingly curved mirror. A special receiving optics, which has as convex-concave lens or mirror system two elements each having at least one free-form surface, as in the still unpublished utility model application with the file number 20 2016 100 006.8. maintain the spot geometry even with temperature changes. As an alternative to a free form, one or more simple optical elements come into consideration, such as the combination of an asphere with a cylindrical lens, but so that the light spot can not be adapted so well. The use of a diffractive optical element instead of the freeform or in combination with a freeform is also conceivable. Again, corresponding explanations also apply to a transmission optics.

The control and evaluation unit is preferably designed for a teach-in mode in which the region of the light receiver to be activated is taught in using the position of a light spot. This can happen in production or later in the field. In manufacturing, for example, tolerances can be compensated for by the decision as to which avalanche photodiode elements are relevant and should be active. In the field, the selection of the area to be activated can be part of a calibration or maintenance process, but also a dynamic adjustment during the measurement or between measurements.

The light receiver is preferably preceded by a microlens array. This microlens field again preferably has a microlens per avalanche photodiode element, wherein alternatively a group-wise assignment is conceivable. Due to the microlenses, the received light exclusively or at least for the most part strikes the avalanche photodiode elements themselves, thus the light-sensitive regions of the light receiver, instead of the intermediate regions, for example with connecting lines or evaluation structures. Thus, an optical sensitivity is achieved as in a light receiver with higher filling factor ideally of 100%. This results in an increased irradiance of the active avalanche photodiode elements under the light spot and thus an increased range of the sensor. The microlens field is particularly effective in combination with a freeform.

The invention will be explained in more detail below with regard to further features and advantages by way of example with reference to embodiments and with reference to the accompanying drawings. The illustrations of the drawing show in:

1 a schematic block diagram of an optoelectronic sensor;

2 a simplified equivalent circuit diagram of avalanche photodiode element in Geiger mode;

3a a schematic representation of a sensor and an object to be detected for illustrating a measured value shift by additional reflexes;

3b a schematic representation of the light receiver and exemplary light spots in the situation of 3a ;

4a an exemplary representation of a round light spot on a SPAD matrix;

4b an exemplary representation of an elongated light spot on a SPAD matrix;

5a a sectional view of a lens with a free form to produce an elongated, homogeneous light spot; and

5b a complementary sectional view of the lens according to 5a in a vertical view

1 shows a simplified schematic block diagram of an optoelectronic sensor 10 in one embodiment as a single-beam light sensor. A light transmitter 12 , For example, an LED or a laser light source, transmits via a transmission optics 14 a light signal 16 into a surveillance area 18 out. Does it meet an object there? 20 , so returns a reflected or reflected light signal 22 via a receiving optics 24 . 25 to a light receiver 26 back. This light receiver 26 has a plurality of light receiving elements called avalanche photodiode elements 28 in Geiger mode or SPADs are formed and which can be understood as pixels. The invention is described using the example of a SPAD matrix, but is also applicable in principle if the light receiver has a pixel structure of light-receiving elements that are not SPADs. The received signals of the avalanche photodiode elements 28 be from a control and evaluation unit 30 read out and evaluated there.

At least parts of the control and evaluation unit 30 can also with the avalanche photodiode elements 28 integrated on a common chip whose area is then photosensitive areas of avalanche photodiode elements 28 and individual or groups of avalanche photodiode elements 28 share associated circuits for their evaluation and control. Indicates the receiving optics 24 . 25 as shown, a microlens array 24 on, then the light signal 22 be focused in each case on the photosensitive areas. The additional lens 25 can provide for a focusing and shaping of the receiving light and a light redistribution. The microlens field 24 is optional, and the receiving optics 24 . 25 can be designed differently than shown, for example, with multiple lenses, reflexive or with a diffractive optical element.

The light receiver 26 can also be called a digital SPAD matrix. This should mean that the control and evaluation unit 30 Information of individual avalanche photodiode elements 28 or groups of them, or even able to turn them on and off. Both leads to the possibility of subregions of the light receiver 26 to activate.

The sensor 10 is preferably range-measuring. This is determined by the control and evaluation unit 30 a light transit time of emission of the light signal 16 until receipt of the returning light signal 22 and convert this over the speed of light into a distance. Several sensors 10 can be combined to form a grooving light grid with multiple, mostly parallel, beams, which measure or monitor distances in each beam. Mobile systems are also conceivable in which the sensor 10 movably mounted, or scanning systems, in which the light signal 16 . 22 distracted by a moving mirror or light emitter 12 or light receiver 26 to be moved.

The arrangement of the sensor 10 in 1 is purely exemplary to understand. Alternatively, other known optical solutions can be used, such as autocollimation with, for example, a beam splitter and a common optic or the arrangement of the light emitter 12 in front of the light receiver 26 ,

2 shows an exemplary simplified equivalent circuit diagram of a single avalanche photodiode 28 in violin mode. In practice, it is a semiconductor device whose unillustrated structure is assumed to be known here. The avalanche photodiode 28 shows on the one hand the behavior of a diode 32 , It has a capacity through a parallel-connected capacitor 34 is represented. The possible avalanche breakdown generates charge carriers whose origin in the equivalent circuit as a current source 36 is pictured. The avalanche breakdown is caused by an incident photon 38 triggered, this process as a switch 40 acts. There are then different possibilities, the output signal 42 to look at, which will not be discussed here.

In standby mode is above the diode 32 a voltage above the breakdown voltage. Then creates an incident photon 38 a pair of charge carriers, this closes the switch as it were 40 so that the avalanche photodiode 28 via the power source 36 flooded with charge carriers. However, new charge carriers only arise as long as the electric field remains strong enough. Is caused by the power source 36 the capacitor 34 discharged so far that the breakdown voltage is exceeded, the avalanche comes to a standstill by itself to a standstill ("passive quenching"). Then the capacitor charges 34 again, until again a voltage above the breakdown voltage at the diode 32 is applied. There are alternative embodiments in which the avalanche is detected from the outside and then a discharge below the breakdown voltage is triggered ("active quenching").

3a schematically shows the sensor 10 in a typical application situation, in the laterally moving objects as indicated by the arrow 20a -B, where in this example an object 20a a small remission and an object 20b has a high remission. This in 3a The coordinate system shown below will also be used below to denote the x, y, and z directions. 3b shows spots of light 44a -B, by the respective returning light signals 22 on the light receiver 26 be generated.

This results in a so-called lateral black-white shift. This refers to an undesired deviation of the switching point at which the object is detected, due to the object emission. The reason for the deviation is that it is in addition to the wanted light spot 44a of the object in the switch position 20a more light spots 44b can give. For this purpose, various causes come into consideration, such as windshield reflections, reflections in a tube of the light emitter 12 or diffraction structures such as a transmitter aperture. If such a disturbing additional light spot 44b provides sufficient signal components, for example, because of its extent or because the remission of the interfering object 20b is greater than that of the object to be detected 20a , this can lead to premature switching. The object position of the lateral movement direction, at which the sensor 10 switches, therefore becomes unintentionally dependent on disruptive effects and in particular on the remission of the objects 20a b.

In principle, the viewing area can be restricted with a narrow aperture of about <200 μm width and thus the lateral black-and-white displacement can be minimized. On the one hand, however, the mechanical design is complex and would need to be adjusted, and on the other hand, the system would be hardly adaptable in operation. Therefore, instead of choosing the route, only certain avalanche photodiode elements are chosen 28a to activate, as in 3b indicated by a rectangle. As a result, light can be hidden outside the area that is actually to be recognized, and the switching position can also be faded out with differently remitting objects 20a -B kept constant. However, in 3b the light spot 44a very bad to the rectangle of the activated avalanche photodiode elements 28a adapted, and this is now further improved.

4a Each show an exemplary illustration of the light receiver 26 with the light spot 44 , the returning light signal 22 created on it. Here are on the light receiver 26 just a few bold avalanche photodiode elements 28a activated in an elongated rectangle arrangement from here 2 × 8 pixels. In 4a is for comparison a circular and in 4b an elongated, to the arrangement of the active avalanche photodiode elements 28a adapted light spot 44 each represented with the same vertical extent. Such an oblong spot of light 44 can preferably by means of the receiving optics 24 . 25 , especially the lens 25 be generated, alternatively or supportively by beam shaping in the transmission optics 14 or an elongated light transmitter 12 for example, with a line arrangement of light sources.

The oblong light spot 44 according to 4b has an aspect ratio h / b of at least h / b = 2: 1 or more, in the example of 4: 1. As can be clearly seen, this light spot is much better at the activated avalanche photodiodes 28a adapted as the circular light spot 44 according to 4a ,

First, one should assume that the light spot 44 and thus the range of activated avalanche photodiodes 28a for a precise switching point should be as small as possible. However, to ensure some signal dynamics and to compensate for the spurious triggering and deadtime effects of SPADs described in the introduction, a minimum number of individual avalanche photodiode elements must be used 28 activated and from the light spot 44 be lit up. For example, if it's like in 3 If only one lateral position arrives, n> 1 avalanche photodiode elements can be used for this purpose 28a be used in the direction of y x vertical direction y.

The on this area of the activated avalanche photodiode elements 28a adapted elongated, preferably rectangular light spot 44 brings in connection with the designed as a digital matrix light receiver 26 significant advantages, since hereby the entire signal light in the relevant area of the light receiver 26 is projected.

As a numerical example, a SPAD matrix with a grid of 20 μm and individual avalanche photodiode elements or SPADs of size 15 μm × 15 μm. For the sake of simplification assumed single lens of the receiving optics 24 , which is not formed as a microlens array, a focal length of 10 mm in the xz plane is assumed. When using a range of avalanche photodiode elements enabled 28a of 1 × n, the reception angle in the xz plane is approximately ± 0.05 °. Then at a distance of 1 m the viewing area is 1.5 mm wide. An object crossing in the x-direction and to be detected 20 is thus detected with pinpoint accuracy. The switching point in the x-direction is thus almost independent of the remission of the object. By using a different focal length of the receiving optics 24 or another number of activated avalanche photodiode elements 28a In the x-direction, this reception angle can be changed.

If one now compares the irradiance of a circular light spot 44 , which in each case x avalanche photodiode elements in the x and y direction 28 . 28a covering, with a rectangular light spot 44 , which also in the y-direction n, but only a few avalanche photodiode elements in the x-direction 28a covers, then there is a significant increase in the irradiance for the rectangular light spot. In the example of 4 meets the circular light spot 44 all in all 51 and thus three times as many avalanche photodiode elements 28a -B like the adjusted rectangular light spot 44 , The gain in irradiance becomes greater lengthwise, such as a width of only one avalanche photodiode element 28a or a larger extent in the y-direction, even larger and decreases with smaller aspect ratio, but does not disappear as long as the light spot 44 remains elongated.

An effect of the significantly increased irradiance and thus of the improved useful signal is a greater range of the sensor 10 , The avoidance of a switching point shift has already been associated with 3 explained. Another advantage is the better Fremdlichtverträglichkeit. Namely falls outside light not parallel to the yz plane on the receiving optics 24 For example, an area outside the activated avalanche photodiode elements becomes predominantly 28a illuminated, and such extraneous light is not detected. In this case, the respective field of view of the activated avalanche photodiode elements 28a be set variably.

An additional homogenization of the irradiance within the light spot 44 For example, by suitable design of the receiving optics 24 . 25 and in particular the lens 25 , alternatively or supporting the transmission optics 14 can also be achieved and brings other significant benefits. Because of the dead time of SPADs after a receive event, a homogeneous irradiance within the light spot increases 44 the total number of detected photons compared to an inhomogeneous irradiation distribution, as would occur in receiving lenses with aspheric surfaces. This also increases the received signal and thus the range of the sensor 10 , Another advantage of homogeneous illumination is the minimization of the variance in the arrival time of the first signal photon on a respective avalanche photodiode element 28 . 28a in a distance-measuring system, which therefore determines the object distance more accurately.

5 shows a lens with a freeform that as a lens 25 the receiving optics 24 . 25 for producing an elongate, homogeneous light spot 44 can be used. It is 5a a section in the yz plane and 5b in the xz plane perpendicular to it. An area 46 This lens is a freeform surface, the second surface 48 an asphere. The result is a rectangular light spot with the illustrated geometry 44 with an extension of 30 μm × 250 μm (b × h).

Such a free-form lens or a likewise usable free-form mirror in a reflexive arrangement can be adapted particularly precisely to the respective requirements, because the light spot 44 can specifically shape and at the same time the intensity distribution within the light spot 44 to get voted. Examples of free-forms are polynomial surface representations, finite sums with basis functions, such as Zernike polynomials or Legendre polynomials, or NURBS surfaces (non-uniform Rational B-splines). The concrete free-form geometry is obtained for example by simulation in an optical program or by explicit calculation such as the so-called tailoring of optical surfaces, which is based on the solution of Monge-Ampere differential equations. 5 gives an example, but the actual best fitting freeform depends on numerous factors like the sensor 10 , its dimensions and internal structure as well as the specific detection application.

Claims (9)

Optoelectronic sensor ( 10 ) for capturing an object ( 20 ) in a surveillance area ( 18 ) with a light transmitter ( 12 ) for emitting a light signal ( 16 ) into the surveillance area ( 18 ) and with a light receiver ( 26 ) containing a plurality of avalanche photodiode elements ( 28 ) for detecting received light from the surveillance area ( 18 ), each biased with a bias voltage above a breakdown voltage and thus operated in a Geiger mode, wherein the light receiver ( 26 ) is digitally configured and thus enables individual or groups of avalanche photodiode elements ( 28 . 28a ), or with a control and evaluation unit ( 30 ), for evaluating a received signal of the light receiver ( 26 ) as well as being adapted to those avalanche photodiode elements ( 28a ), where a light spot ( 44 ) of the object ( 20 ) reflect or reflect light signal ( 22 ), characterized in that the light spot ( 44 ) is elongated. Sensor ( 10 ) according to claim 1, wherein the light spot ( 44 ) has an aspect ratio of at least 2: 1. Sensor ( 10 ) according to claim 1 or 2, wherein the light spot ( 44 ) has a rectangular shape. Sensor ( 10 ) according to one of the preceding claims, which is a distance-measuring sensor, in which the control and evaluation unit ( 30 ) is adapted to a light transit time between transmission of the light signal ( 16 ) and receiving at the object ( 20 ) reflected or reflected light signal ( 22 ) a distance of the object ( 20 ). Sensor ( 10 ) according to one of the preceding claims, wherein the light receiver ( 26 ) a receiving optics ( 25 ), which the elongated shape of the light spot ( 44 ) generated. Sensor ( 10 ) according to claim 5, wherein the receiving optics ( 25 ) the light spot ( 44 ) homogenized. Sensor ( 10 ) according to claim 5 or 6, wherein the receiving optics ( 25 ) at least one free-form surface ( 46 ) having. Sensor ( 10 ) according to one of the preceding claims, wherein the control and evaluation unit ( 30 ) is designed for a learning mode in which the region of the light receiver ( 26 ) based on the position of a light spot ( 44 ) is taught. Sensor ( 10 ) according to one of the preceding claims, wherein the light receiver ( 26 ) a microlens field ( 24 ) is arranged upstream.

Images