SECURE OBJECT TRACKING OF AN OBJECT

Description

[0001] The invention relates to a monitoring device and a method for secure object tracking of an object according to the preamble of claims 1 and 15, respectively.

[0002] Optoelectronic sensors are frequently used in non-contact monitoring to protect against hazards, such as machinery in industrial environments or vehicles in logistics applications. Especially for more complex applications, laser scanners and cameras, especially 3D cameras, are particularly popular.

[0003] In a laser scanner, a light beam generated by a laser periodically sweeps across a surveillance area with the help of a deflection unit. The light is remitted by objects in the surveillance area and evaluated in the laser scanner. The angular position of the deflection unit is used to determine the angular position of the object, and the distance of the object from the laser scanner is also determined from the light travel time, using the speed of light. Using the angle and distance information, the location of an object in the surveillance area is recorded in two-dimensional polar coordinates. The third spatial coordinate can also be recorded by a relative movement in the transverse direction, for example by an additional degree of freedom of movement of the deflection unit in the laser scanner or by moving the object relative to the laser scanner.

[0004] A 3D camera measures a distance and thereby obtains depth information. The captured three-dimensional image data, with distance or range values ​​for the individual pixels, is also referred to as a 3D image, range image, or depth map.

[0005] 3D cameras are available in various technologies, including time-of-flight, stereoscopic, and projection methods, as well as plenoptic cameras. In a time-of-flight (TOF) camera, a scene is illuminated with amplitude-modulated light. The light returning from the scene is received and demodulated at the same frequency used to modulate the transmitted light (lock-in method). The demodulation results in an amplitude measurement value that corresponds to one sample of the received signal. According to the Nyquist criterion, at least two samples are necessary to determine the phase of a periodic signal. Therefore, the measurement is performed with different relative phase positions between the signals for the transmit-side modulation and the receive-side demodulation.This allows the absolute phase shift between the transmitted and received signals to be determined, which in turn is proportional to the object distance in the scene.

[0006] In accident prevention applications, a protective field is monitored that must not be entered by operating personnel during machine operation. If the sensor detects an inadmissible intrusion into the protective field, such as an operator's leg, the machine is switched to a safe state. Sensors used in safety technology must be particularly reliable and therefore meet stringent safety requirements, such as the EN13849 standard for machinery safety and the EN61496 device standard for electro-sensitive protective equipment (ESPE). To comply with these safety standards, a number of measures must be implemented, such as reliable electronic evaluation using redundant, diverse electronics or various function monitoring systems, especially monitoring for contamination of optical components, including a front screen.

[0007] A safety laser scanner or safety camera that complies with these standards and is designed for protective field evaluation works internally with very large amounts of information from the scanned point clouds or depth maps. However, only highly condensed binary information is reliably made available to the outside, namely whether the protective field is violated or not. A safe output (OSSD, Output Signal Switching Device) is typically used for this purpose. More complex safe evaluations such as determining the position of an object or object tracking are not conventionally available.

[0008] While algorithms for object tracking using optical detection have long existed, based on conventional image processing, Kalman filters, or increasingly even artificial intelligence, they are not suitable for safety-related applications due to their complexity and the lack of suitable sensors and control components. Even if it were possible to harden an algorithm through appropriate security measures so that it would meet safety standards, the required scan or image data in a safety laser scanner or safety camera cannot be securely accessed. As already described, there is only one external interface for one bit of secure information about a protective field violation. Even the internal computing resources of the sensor or those of a safety controller are far from sufficient to implement object tracking using any of the known approaches.

[0009] Safety laser scanners and safety cameras that evaluate multiple protective fields simultaneously have recently become known. However, simultaneous evaluation is very computationally intensive, and only a very limited number of parallel protective fields is provided. This ultimately only increases the externally available safe information from one bit to several bits. More than a few position classes corresponding to the areas defined by the respective protective field are therefore indistinguishable, and this is only sufficient for object tracking in special cases. This number may increase in future device generations, but this will not lead to positioning accuracies comparable to the evaluation of scan point clouds or images, and therefore does not change the fundamental limitations compared to object tracking with existing algorithms.

[0010] EP 3 470 879 A1 configures at least partially overlapping monitoring fields in a laser scanner. This creates monitoring segments that differ from one another in the overlapping monitoring fields. The number of monitoring fields provided by the hardware is thus refined to the monitoring segments. With clever configuration, n monitoring fields can result in 2 n< or even slightly more monitoring segments. However, this still does not achieve a particularly fine resolution as long as n remains limited to very small values ​​on the hardware side, and for larger n, the configuration effort would quickly become prohibitive.

[0011] EP 3 709 106 A1 proposes a safety system that validates complex, non-safe evaluations with less complex, safe evaluations. In one embodiment, objects are localized in a non-safe manner using scan point clouds. On the other hand, there is a grid of protective fields. The non-safely determined fine position is validated with the safe grid position. As already mentioned, the scan point cloud of a safety laser scanner cannot be read in reality, and there is no non-safe interface for this purpose, so new devices are required. Furthermore, a safe position in the narrower sense remains limited to the grid of protective fields, which must remain coarse due to the small number of possible protective fields.

[0012] EP 1 835 306 A1 discloses a parking aid that illuminates the rear area with Fresnel rings and determines the light level with a receiver. When the Fresnel rings pass over an object while the vehicle is moving, the light level increases or decreases depending on whether a bright or dark zone is currently encountering the object. By counting the light-dark transitions, the relative position of the object to the vehicle can be determined. In another embodiment, alternating Fresnel rings are generated using two light transmitters of different wavelengths, and their transitions are evaluated.

[0013] Completely independent of the technologies discussed so far, rotary encoders or incremental encoders are known. These consist of two parts that move relative to each other. A scanning unit is connected to one part, and a measuring element is connected to the second part. The measuring element has code tracks, and the scanning unit generates scanning signals by scanning the code tracks. From the scanning signals, the respective positions of the two parts relative to each other can be determined.

[0014] It is therefore an object of the invention to extend the safe functionality of a monitoring device that can be used in safety technology.

[0015] This object is achieved by a monitoring device and a method for safely tracking an object according to claim 1 and 15, respectively. The monitoring device has at least one first safe optoelectronic sensor and a safety controller connected to it. A safety controller refers to a safe evaluation unit implemented in any hardware, in particular a safety controller as a controller approved for safety applications in the narrower sense. As used throughout the description, safe and safety mean that measures have been taken to control errors up to a specified safety level or to comply with the regulations of a relevant safety standard for machine safety or non-contact protective devices, some of which are mentioned in the introduction.However, this does not necessarily mean that the monitoring device is used in a security application, even if it could be. For example, it is conceivable that a specific configuration does not provide security in accordance with standards, or that the application is only aimed at observation or, for example, security in the sense of theft protection, not protection against the risk of injury. "Not secure" is the opposite of "secure"; therefore, the aforementioned fail-safe requirements are not met for non-secure devices, transmission paths, evaluations, and the like.

[0016] The first safe optoelectronic sensor comprises a light receiver that generates a received signal from received light from the monitored area. Depending on the sensor, the received signal is, for example, the signal from a photodiode that receives a returning scanning beam, or image data from an image sensor is generally referred to as the received signal. The first sensor further has at least two safe outputs, in particular OSSDs, at each of which a preferably binary safe output signal can be output. The two safe outputs are preferably separate physical connections, but can also be implemented as bits or the like of a more complex safe output. A first control and evaluation unit of the first sensor is designed for protective field monitoring with at least two protective fields: a first protective field or protective field A and a second protective field or protective field B.The protective fields are each individually configurable sub-areas of the monitored zone and are historically named protective field, although in reality, depending on the first sensor, it can be a three-dimensional spatial area. An object entering the first protective field leads to a safe output signal at the first safe output and correspondingly, in the second protective field, to a safe output signal at the second safe output. The affiliation is sometimes expressed below by referring to the safe outputs as OSSD A and OSSD B, for example. OSSD A and B are preferably in the On state when there is an object in the associated protective field A or B, and otherwise in the Off state. The safety controller is designed to evaluate the safe output signals.

[0017] The invention is based on the fundamental idea of ​​dividing the two protective fields into separate partial protective fields, thus achieving a finer spatial resolution. This initially only applies to one dimension, but can be expanded to two and, if necessary, three dimensions in embodiments to be described later. The first and second protective fields are each configured as multiple partial protective fields, and the partial protective fields are lined up along a first line in such a way that the partial protective fields are interlocked, so to speak, with first partial protective fields of the first protective field alternating with second partial protective fields of the second protective field. There may or may not be gaps and / or an overlap in each case.

[0018] This means that the partial protection fields belonging to the same protection field are preferably not contiguous. The sequence is preferably alternating, i.e. if the first line is followed, both protection fields are violated according to the pattern ..ABABABA.. As will be explained later, the protection fields form a type of code track, and more complex code schemes are conceivable, with ..ABBABAAB.. being just one of countless other examples. Furthermore, it is conceivable to add at least one further protection field C, D ... and thus create more powerful codes. All of this is known per se, for example from rotary encoders, and will therefore not be described in detail here, but using the simple representative example ..ABABABA.. The first line can be curved or, preferably, straight; according to the invention, a partial protection field corresponds to a discrete coordinate along the line, whereby the grain can be somewhat finer by switching protection fields, as explained later.

[0019] It should be emphasized that according to the invention, by no means can only a number of positions be distinguished according to the number of protective fields or OSSDs, as with the conventional solutions discussed in the introduction. According to the invention, protective fields are freely configurable, each with, in principle, any number of separate partial protective fields of any shape, as small as desired within the resolution of the first sensor. In contrast to a separate protective field, partial protective fields of the same protective field are interdependent. A violation is only reported jointly at the corresponding safe output; the first sensor therefore initially does not provide any information as to which partial protective field of the affected protective field an object intrusion occurred in. However, the invention makes it possible to determine this through the special configuration of the at least two protective fields.

[0020] The invention has the advantage that discrete, safe object tracking can be implemented in one direction of movement, and later also in multiple directions of movement with significantly increased resolution. This is not tightly tied to the extremely small number of simultaneously monitored protective fields, but rather requires two protective fields per direction of movement. This utilizes the high degree of design freedom in configuring the geometry of a protective field. This opens up a new safety function. It is based on the already safe protective field evaluation and the equally safe outputs and thus, in a sense, inherits functional safety largely without additional effort. Safe, discrete object tracking can be used to check the plausibility or validate an even more precise, non-safe object tracking function, for example, with the concept of EP 3 709106 A1 discussed in the introduction.Thanks to the invention, the validation grid becomes much finer, so that remaining possible errors in the non-safe object tracking function receive a significantly lower upper bound.

[0021] The safety controller is preferably configured to evaluate a temporal sequence of the safe output signals at the first safe output and the second safe output to determine the position of the object along the first line. From the temporal sequence, it is possible to determine which of the partial protection fields is violated, thus yielding a position in the resolution of the partial protection fields, not just the protection fields.

[0022] Preferably, the first partial protection fields form a first code track, and the second partial protection fields form a second code track. This is intended to illustrate an analogy to incremental encoders. The partial protection fields and the gaps between them code one and zero, and the scanning of this special code track occurs when a moving object enters and exits a partial protection field, and the corresponding safe output changes its state accordingly.

[0023] Preferably, the first partial protection fields and the second partial protection fields partially overlap each other in the direction of the first line, in particular by half. This results in the injury sequence ..A AB B.. If an object follows the first line, the injury sequence ..A AB B.. Preferably, a clear area without a partial protection field is arranged along the first line between the partial protection fields. This results in injury sequences such as ...A(empty)B(empty)A.. or ..AB(empty)AB.. Particularly preferably, the overlap and clear area are cumulative, with an injury sequence ..A AB B (empty) A AB .. The clear area can be as large as a partial protection field and / or their mutual overlap. Another preferred extent of the clear area is measured by whether a person can fit within it without triggering a protection field, and this results in a size of 50 cm - 1 m, in particular 60 cm, corresponding to the horizontal extent of a standing person.The overlap area can be as large as the free area, but the actual requirement here is simply that both overlapping protective fields are reliably triggered when the overlap area is entered. A smaller area of, for example, 20 cm is sufficient for this.

[0024] The safety controller is preferably designed to evaluate the temporal sequence of the safe output signals using an incremental encoder method for evaluating the scanning signals of two offset code tracks of a measuring scale. The analogy to an incremental encoder is not merely illustrative; the safe output signals have the same progression as the scanning signals of an incremental encoder. This provides access to the entire powerful arsenal of coding and evaluation methods developed for incremental encoders, which is therefore known per se and can be used according to the invention, without further explanation here.

[0025] The safety controller is preferably designed to determine the direction of the object's movement from the chronological sequence of the safe output signals. This works particularly reliably with the above-specified arrangement of partial protection fields with mutual overlap and clearance. If the safe output signals indicate .. A AB B (empty)..., the object has moved forward; if they indicate B AB A, the object has moved backward.

[0026] The safety controller preferably has a counter that counts up and down according to the direction of movement when entering and exiting a partial protection field. Knowing the respective direction, the position can be recorded incrementally with a counter. It is conceivable to record intrusions into a specific protection field using separate counters, but a single counter that does not distinguish which protection field is affected is sufficient. This could be calculated back from the counter reading anyway. The actual position on the first line can be calculated from the counter reading, knowing the protection field geometry.

[0027] The safety controller is preferably designed to only process a safe output signal at one of the safe outputs as a new safe output signal when a safe output signal is temporarily present at the other safe output. This provides a concrete example of how a procedure developed for incremental encoders can be advantageously applied to the invention. Incremental encoders are prone to bouncing. The term originally comes from a switch or button that triggers several times in quick succession due to a malfunction. A similar situation can occur if an object is located at the boundary of a protective field. This measure is used for debouncing. A largely stationary object at the boundary of the protective field only triggers a further violation of the same protective field when it bounces, whereas a moving object will inevitably violate the other protective field.

[0028] The partial protection fields preferably have the same shape and / or size. This results in uniform code elements, which significantly simplify both configuration and evaluation and significantly reduce their susceptibility to errors. For the same reasons, the partial protection fields preferably have a simple shape, for example, rectangular or cuboid. This requirement is not necessary for the functional principle.

[0029] The safety controller is preferably designed to shift the protective fields along the first line as soon as a safe output signal is applied to one of the safe outputs. Thus, the partial protective fields effectively track the object, which repeatedly breaches the shifted protective field boundary during its movement. The shift preferably occurs by a fraction of the distance between two partial protective fields. A configuration with equally sized partial protective fields and spacings corresponding to the extent of the partial protective fields (A(empty)B(empty)) is preferably provided. The shift preferably occurs by switching to a staggered arrangement of the protective fields. This preferred configuration can be achieved with four switchable protective field sets. The shifts are preferably counted to determine the position incrementally.

[0030] The safety controller is preferably designed to move the protective fields in one direction when a safe output signal is present at the first safe output and in the other direction when a safe output signal is present at the second safe output. The first protective field then serves to detect forward movement, and the second protective field serves to detect backward movement along the first line. A counter is incremented or decremented depending on the direction of movement.

[0031] The first safe optoelectronic sensor is preferably a 3D camera, particularly a time-of-flight camera. This allows a large monitoring area to be captured, and the configuration of protective fields and partial protective fields is possible in a fine grid. The 3D camera, originally designed only for protective field evaluation, is given the additional function of safe object tracking by the invention. Alternatively, the first safe optoelectronic sensor can be a laser scanner, but this requires a suitable perspective.

[0032] The safety controller is preferably designed to trigger a safeguarding of a machine monitored by the monitoring device if the object is in a dangerous position and / or in a dangerous movement. The assessment of whether an object position or movement is dangerous can be performed in a downstream processing unit, which for the sake of simplicity is assigned to the safety controller. A dangerous position can be too close to a machine or a machine part, with possible time dependencies or the consideration of the machine's work processes. Movements enable additional evaluations; for example, a movement parallel to the machine or even with a subcomponent away from it is less critical than a movement directly towards the machine. Speed ​​can also play a role (speed and separation monitoring).The safeguarding may consist of swerving, slowing down or stopping the machine or assuming another safe state.

[0033] The monitoring device preferably comprises a second optoelectronic sensor with a second light receiver, at least one third safe output and one fourth safe output, as well as a second control and evaluation unit, which is designed for protective field monitoring of at least one third protective field with a plurality of third partial protective fields and a fourth protective field with a plurality of fourth partial protective fields with output of a safe output signal at the third safe output or the fourth safe output in the event of object intervention, wherein the third partial protective fields and fourth partial protective fields are arranged alternatingly one after the other along a second line transverse to the first line, and wherein in particular the safety controller is designed to evaluate a temporal sequence of the safe output signals at the third safe output and the fourth safe output,to determine the position of the object along the second line. The second optoelectronic sensor is thus also capable of dividing two protective fields into partial protective fields, monitoring them, and providing corresponding safe output signals. It thus corresponds to the first optoelectronic sensor, at least with regard to the features essential to the invention, and is preferably functionally or even structurally identical. The two sensors monitor partial protective fields lined up on lines oriented perpendicular to each other. This enables object tracking in two dimensions. The second line, like the first line, is preferably straight and even more preferably oriented perpendicular to the first line. This then results in Cartesian coordinates. At least one line can be curved, resulting in correspondingly distorted coordinates.which can be transformed into Cartesian coordinates at any time, given knowledge of the protective field geometry. By adding a third optoelectronic sensor with a fifth and sixth protective field with partial protective fields along a third line perpendicular to the other two lines, object tracking can be extended into the third dimension.

[0034] The first control and evaluation unit is preferably configured to switch the protective field monitoring to a third protective field with a plurality of third partial protective fields and a fourth protective field with a plurality of fourth partial protective fields. The third partial protective fields and fourth partial protective fields are arranged alternatingly along a second line transverse to the first line. In particular, the safety controller is configured to evaluate a temporal sequence of the safe output signals to determine a position of the object along the second line. Violations of the third and fourth protective fields preferably lead to a corresponding safe output signal at the first and second safe outputs, respectively.Switching is preferably very fast (toggling), and in particular faster than the response time of the protective fields required by the safety concept, so that all protective fields are effectively monitored simultaneously. This is an alternative to the embodiment described in the previous paragraph, which emulates the second safe optoelectronic sensor through switching, so that only one sensor is required.

[0035] The method according to the invention can be further developed in a similar manner and thereby exhibits similar advantages. Such advantageous features are described by way of example, but not exhaustively, in the subclaims following the independent claims.

[0036] The invention will be explained in more detail below with regard to further features and advantages, using exemplary embodiments and with reference to the accompanying drawings. The figures of the drawing show: Fig. 1 a schematic block diagram of a 3D time-of-flight camera; Fig. 2 a schematic diagram of a monitoring device with two cameras and a safety controller; Fig. 3 an exemplary depth map with two protective fields; Fig. 4a an alternative representation of the protective fields from Figure 3 and in the lower part a signal curve at assigned safe outputs during an object movement through the protective fields; Fig. 4b a representation according to Figure 4a when object moves in the opposite direction; Fig. 5 an example depth map similar Figure 3 now with several partial protection fields; Fig. 6 an alternative representation of the protection fields from Figure 5and in the lower part a signal curve at assigned safe outputs during an object movement through the protective fields; Fig. 7 a schematic representation of two pairs of protective fields arranged crossed to each other for object tracking, now in two dimensions; Fig. 8 an exemplary depth map with protective fields that are moved in a further embodiment; Fig. 9 an alternative representation of the protective fields from Figure 8 with illustration of the switching of protective fields when a moving object intervenes in the protective field; and Fig. 10 a schematic representation of four protective field sets, between which switching takes place cyclically when an object moves.

[0037] Figure 1shows a schematic block diagram of a camera 10, which is preferably designed as a 3D time-of-flight camera and which is described as representative of an optoelectronic sensor that can be used in connection with the invention. An illumination unit 12 emits transmitted light 16, modulated by a transmitting optics 14, into a monitoring area 18. LEDs or lasers in the form of edge emitters or VCSELs can be considered as the light source. The illumination unit 12 can be controlled such that the amplitude of the transmitted light 16 is modulated at a frequency typically in the range of 1 MHz to 1000 MHz. The modulation is, for example, sinusoidal or rectangular, or in any case, a periodic modulation. The frequency creates a limited unambiguous range for the distance measurement, so that low modulation frequencies are required for long ranges of the camera 10.Alternatively, measurements are carried out at two to three or more modulation frequencies in order to increase the unambiguousness range by combining the measurements.

[0038] If the transmitted light 16 strikes an object or a person 20 in the surveillance area 18, a portion is reflected back to the camera 10 as received light 22 and there guided through a receiving optics 24, for example a single lens or a receiving objective, to an image sensor 26. The image sensor 26 has a plurality of receiving elements or receiving pixels 26a, arranged, for example, in a matrix or row. The resolution of the image sensor 26 can range from two or a few to thousands or millions of received pixels 26a. Demodulation takes place therein according to a lock-in method. By repeated acquisition with modulation of the transmitted light 16 that is slightly offset across the repetitions, several sample values ​​are generated, from which the phase shift between the transmitted light 16 and the received light 22, and thus the light propagation time, can ultimately be measured.The pixel arrangement is typically a matrix, resulting in lateral spatial resolution in an X-direction and a Y-direction, which complements the Z-direction of the distance measurement to the three-dimensional image data. This 3D acquisition is preferably meant when referring to a 3D camera, a 3D time-of-flight camera, or three-dimensional image data. In principle, however, other pixel arrangements are also conceivable, such as a pixel row selected in a matrix or forming the entire image sensor of a line-scan camera.

[0039] At least two protective fields are configured in a control and evaluation unit 28 with at least one digital processing unit, such as a microprocessor or the like. These are geometric specifications for a sub-area of ​​the monitored area 18, which are configured, for example, in a CAD program or in any other way using the control and evaluation unit 28 or imported via an interface 30. The protective fields are monitored for object intrusions, and in the event of a protective field violation, a safe output signal is output at a safe output 32, 34 assigned to the protective field. The state of the safe output therefore reflects the presence or absence of an object in the associated protective field in binary form. The two protective fields are referred to below as protective field A and B, and the associated safe output 32, 34 is referred to as OSSD A and OSSD B.The camera 10 and in particular the protective field evaluation including the output signals or OSSD states are safe in the sense defined above.

[0040] Figure 2 shows a schematic representation of a monitoring device with two cameras 10a-b and a safety controller 36, which can be used, for example, at an intersection or in connection with the control of a robot. In an alternative embodiment, only one camera is provided, although other optoelectronic sensors are also possible, in particular a laser scanner.

[0041] The evaluation of the OSSD states, which will now be described, takes place in a safety controller 36. This is initially generally understood as the safe evaluation of a computing unit in any hardware and only preferably as a safety controller in the narrower sense. The safety controller 36 performs safe object tracking, the result of which can be, for example, safe values ​​for the position and, if necessary, also speed at a given point in time. This, in turn, can be used for a risk assessment in a higher-level evaluation (not shown), which is alternatively also implemented in the safety controller 36. If a hazard is detected, a safety-related signal is output to a monitored machine or robot.The machine or robot then slows down or switches to work steps that are safe, at least given the detected object movement. Only in an emergency is the machine transferred to a safe state. This ensures high overall availability and productivity.

[0042] Safe object tracking is an important basic function for a wide variety of automation applications in manufacturing and logistics, for example, when a vehicle is required to avoid a person rather than simply stop. Likewise, robots should use knowledge of the precise position of a nearby person to divert to other work areas and maintain productive processes. Future safety solutions that influence automated processes at a higher level in a larger area, up to an entire hall or factory, will also rely on knowledge of the positions of all people.

[0043] Figure 3 shows an example depth map with two protective fields. According to a top-down perspective, parts of the ground and the upper part of a person's body 20 are recorded as a 3D point cloud. Two overlapping protective fields 38A-B are configured at head height. Figure 4a shows in the upper part an alternative representation of the protective fields 38A-B from a top view and in the lower part a temporal signal curve at the associated safe outputs 32, 34, which are designated OSSD A and OSSD B, during a movement of the person 20 through the protective fields 38A-B, hereinafter referred to as protective fields A and B. Figure 4b This is shown when person 20 moves in the opposite direction.

[0044] The protective fields A and B can be understood as code elements of a code track of an incremental encoder. An incremental encoder is used to measure a linear movement or a rotary movement in discrete steps. For this purpose, it uses at least two separate code tracks with incremental code elements, each of which generates a binary periodic signal upon movement. The signals of both code tracks are phase-shifted from each other, which makes it possible to determine the direction of movement. The principle can be applied to discrete object tracking with an optical sensor and protective field monitoring. With an incremental encoder, the scanning of a respective code element generates a pulse in the scanning signal. In a similar way, the movement of the person 20 through a protective field A or B generates a temporary on state at the associated OSSD A or B, which is also pulse-shaped.

[0045] As in Figure 4aTo detect this, when a movement occurs in one direction, here upwards, protective field A is triggered first and then shortly after protective field B. Accordingly, two pulses occur at OSSD A and OSSD B in quick succession. When the movement direction is reversed according to Figure 4b the temporal behavior is reversed. The overlap provides an additional consistency condition: There must always be an intermediate state in which both protective fields A and B are violated and thus a pulse is present at both OSSDs A and B at the same time. With the constellation of Figures 3 and 4a-b Initially, only the direction of movement of the penetration of this pair of protective fields A and B can be determined. Since the protective field evaluation is reliable, the information on which side of the protective fields A, B the person 20 is located is reliably available. The constellation of the Figures 3, 4a-b but also corresponds to only a single code element or increment of an incremental encoder.

[0046] Figure 5 shows an example depth map similar Figure 3 now with several partial protection fields. This takes advantage of the fact that the protective fields 38A-B can be configured almost arbitrarily. In terms of protective field evaluation, there are still only the two protective fields A and B. However, each of the protective fields A and B is now divided into a plurality of non-contiguous partial protection fields, with the partial protection fields being arranged alternately along a line. This enables object or movement tracking along the line and thus in one dimension. In a further embodiment, which will be described later with reference to the Figure 7 described, this can be extended to two or three dimensions.

[0047] Figure 6 shows in the upper part the protection fields A and B divided into partial protection fields of the Figure 5again from the top view and in the lower part a signal curve at OSSD A and OSSD B, while a person 20 moves in the direction shown through the protective fields. The periodic continuation of the constellation from the Figures 3, 4a-b expands individual code elements to form a code track. Person 20 generates a sequence of pulse pairs by moving through the protective fields A and B and their partial protective fields. Such a signal pattern corresponds to the scanning signals of an incremental encoder. It has already been Figure 4a-bexplains how each pulse pair can be evaluated to reliably determine the direction of movement. With the help of a counter and the partial protection fields, the linear movement of person 20 can be measured. For movement in one direction, here for example left to right, the counter is incremented, and for movement in the opposite direction it is decremented. The counter value can be converted into a discrete linear position using the specific protection field geometry. Initializing the counter is unproblematic as long as persons 20 can only enter the area of ​​protection fields A and B from the left or right and as long as the direction of movement can be used to differentiate between right and left. For the one-dimensional case, structural or other measures may be necessary to prevent a person 20 from entering from the side.In the later two-dimensional expansion, this is also eliminated because a person 20 is not created in the middle of a room. Only a protocol is required when activating and releasing the system to ensure that no person 20 is initially in or between the protected fields, but these are common measures in safety technology.

[0048] Since the protective field evaluation and the signals at the OSSDs are reliable in terms of functional safety, only the simple binary counter evaluation needs to be implemented in accordance with functional safety specifications. A safety controller 36 with very moderate performance is sufficient for this. The resulting counter reading can be evaluated using internal logic operations and, in turn, used as a control signal for a machine. This can be used to prompt a machine or robot, in particular, to take appropriate measures to eliminate a hazard, including complete deceleration or shutdown. However, safe position information or object tracking can also be evaluated in a much more complex manner, for example, by determining the appropriate safety measure based on the trajectories, possibly also speeds and accelerations, of monitored persons, or the machine status.

[0049] Motion tracking, analogous to an incremental encoder, was described using a very simple configuration with regular, similar, alternating, and overlapping partial protection fields. Numerous other coding and evaluation methods are known for incremental encoders. Such code tracks can be simulated with partial protection fields, and the incremental encoder's evaluation method can then be applied to the corresponding signals at the OSSDs. This particularly affects the robustness of the evaluation. One example is debouncing during switching. During bouncing, person 20 is located at a protective field boundary and repeatedly triggers the protective field through small movements. One possible debouncing method is to require that after a protective field intervention in protective field A, a protective field intervention must first occur in protective field B, and vice versa; otherwise, the protective field intervention is ignored.However, incremental encoders have even more complex debouncing methods.

[0050] Figure 7 shows a schematic representation of two pairs of protective fields 38Ax, 38Bx, 38Ay, 38By arranged in a crosswise arrangement for object tracking, now in two dimensions. This doubles the protective field configuration explained so far, with one pair of protective fields 38Ax, 38Bx responsible for one direction of movement or dimension, and the other pair of protective fields 38Ay, 38By responsible for the other direction of movement or dimension. Furthermore, the respective widths are preferably adjusted so that the flat area is covered by the resulting grid of partial protective fields.

[0051] In one direction, i.e., X or Y, motion tracking proceeds as described. Now, only two counters, one per direction or dimension, need to track the positions along their respective directions by counting up or down. The orientation of the two protective field pairs 38Ax, 38Bx, 38Ay, 38By is preferably perpendicular to each other; however, skewed coordinates can be handled analogously with knowledge of the protective field geometry. This also applies to curved partial protective fields or curved lines along which they are arranged; in this case, a corresponding coordinate transformation into Cartesian coordinates is possible with knowledge of the protective field geometry.

[0052] The additional protective field pair can be monitored by using a second sensor, as shown in Figure 2as already shown. Alternatively, a single sensor remains, which quickly switches its protective fields between the configurations for a respective direction ("toggling"). It should be noted that difficulties can arise if more than one person is in the area covered by the protective fields. With the few protective fields, four in the example, it is not possible to clearly assign a protective field intervention to one of the people. However, this will then very quickly lead to signal curves on the OSSDs that a single person could not trigger. Thus, while the system is not capable of tracking multiple people, it is very capable of detecting the error state with multiple people and, if necessary, initiating a safety measure.This safety measure can consist of switching to a conventional protective field, which only protects the actual source of danger and cannot provide reliable object tracking.

[0053] Figure 8 To illustrate another embodiment, FIG. 1 shows an exemplary depth map with a person 20 and protective fields 38A-B. In this embodiment, the protective fields 38A-B are switched synchronously with the movement. Figure 9 shows the protective fields 38A-B of the Figure 8 again in a top view in three different stages of a movement. Only two partial protection fields are shown, this will be explained later with Figure 10 Analogous to the previous procedure, periodically with a large number of such partial protection fields.

[0054] Links in Figure 9At a time t = t 0 , a person 20 is located between two partial protection fields of the protection fields 38A-B. A counter for the position is in a state n. The two partial protection fields belong to different protection fields 38A-B in order to distinguish between a forward and backward movement. In the middle of the Figure 9 Person 20 has moved forward into the partial protection field of the protection field 38A at a time t = t 0 + dt. The counter for the position is incremented accordingly to n + 1. On the right in Figure 9The protective fields are then switched to another configuration in which the partial protective fields are slightly offset upwards. This eliminates the protective field violation. As the movement continues, protective field 38A for the forward movement is violated again, the counter increments to n+2, and the switchover to a slightly upwardly offset protective field set occurs once more. During each reverse movement, the protective field switchover occurs in the reversed manner, with a downwardly offset protective field set, and the position counter decrements by one.

[0055] Figure 10shows a schematic representation of four protective field sets, between which a cyclic switchover occurs during an object movement. In these protective field sets, the extent in the monitored direction of movement is the same for the partial protective fields and the gaps between them, and the offset is half of this common extent. Thus, the cycle is completed after four switchovers. If necessary, the roles for forward and backward, i.e. protective field A and B, can be swapped when switching between the first and last protective field set at the end of a cycle. Figure 10 is exemplary, many alternative protective field sets are conceivable with which a counter for the position and thus safe object tracking can be implemented by automatically switching to slightly offset partial protective fields.

[0056] The alternative embodiment with switching to offset protective fields is analogous to the explanations for Figure 7 It can be extended to the multidimensional case. It requires somewhat more implementation effort, as multiple protective field sets must be configured, although this can be significantly simplified with at least partially automated support, at least on the user side. The required rapid switching between protective field sets also leads to a certain amount of additional effort and higher demands on the hardware and sensor.

[0057] In all embodiments, it can be useful to use multiple sensors. On the one hand, this allows a larger overall monitoring area 18 to be captured. Multiple sensors can serve alternatively or additionally to achieve a grid resolution. The resolution of object tracking is limited by a minimum size and a minimum distance between partial protection fields. This is due to safety-related reasons that only allow reliable protection field evaluation, depending on the original sensor resolution, for protection fields with a certain resolution, for example, a minimum size of 30 cm and a minimum distance of 50 cm. By slightly offsetting the protection fields of two or more sensors, an overall higher resolution can be achieved. Here, again, there is an analogy to incremental encoders with more than two code tracks.For incremental encoders, arrangements of the code elements and evaluations of their scanning signals are also known, which can be transferred to partial protection fields and an evaluation of associated OSSD signals and which are more robust and / or have better resolution than naive approaches.

[0058] The described 3D camera 10, in particular a TOF camera, is particularly suitable for the safe object tracking according to the invention. However, other sensors are also possible, in particular a safety laser scanner. Certain limitations, for example, due to constraints on the application geometry, must be accepted. The movement should preferably be perpendicular to the beam direction of the safety scanner and tracked by a corresponding arrangement of two-dimensional, overlapping protective fields.

Claims

1. A monitoring device for safe object tracking of an object (20) in a monitoring area (18), having at least one first safe optoelectronic sensor (10, 10a) and a safety controller (36) connected to the first optoelectronic sensor (10, 10a), the first optoelectronic sensor (10, 10a) comprising a first light receiver (26) for generating a received signal from received light (22) from the monitoring area (18), at least a first safe output (32) and a second safe output (34), each for outputting a safe output signal, and a first control and evaluation unit (28) configured to safely monitor at least a first protective field (38A) and a second protective field (38B) in the monitoring area (18) for object intrusions using the received signal and, in the event of an object intrusion into the first protective field (38A), to output a safe output signal at the first safe output (32) and, in the event of an object intrusion into the second protective field (38B), to output a safe output signal at the second safe output (34), wherein the safety controller (36) is configured to evaluate the safe output signals, wherein the terms "safe" and "safety" mean that measures are taken to control faults up to a specified safety level and to comply with regulations of a safety standard for machine safety or electro-sensitive protective equipment, and wherein the first protective field (38A) and the second protective field (38B) are each a partial area of the monitoring area (18) configured by means of geometric specifications in the first control and evaluation unit (28) or imported via an interface (30), wherein the first protective field (38A) comprises a plurality of first partial protective fields and the second protective field (38B) comprises a plurality of second partial protective fields, and wherein the first partial protective fields and the second partial protective fields are arranged alternately in succession along a line.

2. The monitoring device according to claim 1, wherein the safety controller (36) is configured to evaluate a time sequence of the safe output signals at the first safe output (32) and the second safe output (34) in order to determine a position of the object (20) along the first line.

3. The monitoring device according to claim 1 or 2, wherein the first partial protection fields form a first code track and the second partial protection fields form a second code track.

4. The monitoring device according to any of the preceding claims, wherein the first partial protective fields and the second partial protective fields partially overlap one another in the direction of the first line, in particular by half, and / or wherein a free region without a partial protective field is arranged along the first line between the respective partial protective fields.

5. The monitoring device according to any of the preceding claims, wherein the safety controller (36) is configured to evaluate the time sequence of the safe output signals using an incremental encoder method for evaluating the scanning signals of two code tracks of a measuring standard which are offset from one another.

6. The monitoring device according to any of the preceding claims, wherein the safety controller (36) is configured to determine a direction of movement of the object (20) from the time sequence of the safe output signals, and wherein in particular the safety controller (36) comprises a counter that is counted up and down in accordance with the direction of movement when entering a partial protective field and exiting a partial protective field.

7. The monitoring device according to any of the preceding claims, wherein the safety controller (36) is configured to process a safe output signal at one of the safe outputs (32, 34) as a new safe output signal only if a safe output signal is present at the other safe output (34, 32) in the meantime.

8. The monitoring device according to any of the preceding claims, wherein the partial protective fields have the same shape and / or size.

9. The monitoring device according to any of the preceding claims, wherein the safety controller (36) is configured to shift the protective fields (38A, 38B) along the first line as soon as a safe output signal is present at one of the safe outputs (32, 34), wherein in particular the shifting takes place by switching to a staggered arrangement of the protective fields (38A, 38B).

10. The monitoring device according to claim 9, wherein the safety controller (36) is configured to shift the protective fields (38A-B) in one direction along the first line in the event of a safe output signal at the first safe output (32) and in the other direction along the first line in the event of a safe output signal at the second safe output (34).

11. The monitoring device according to any of the preceding claims, wherein the first safe optoelectronic sensor (10, 10a) is a 3D camera, in particular a time-of-flight camera.

12. The monitoring device according to any of the preceding claims, wherein the safety controller (36) is configured to trigger a safeguarding of a machine monitored by the monitoring device if the object (20) is in a dangerous position and / or in a dangerous movement.

13. The monitoring device according to any of the preceding claims, comprising a second optoelectronic sensor (10b) with a second light receiver, at least one third safe output and a fourth safe output and a second control and evaluation unit configured for protective field monitoring of at least one third protective field with a plurality of third partial protective fields and a fourth protective field with a plurality of fourth partial protective fields with output of a safe output signal at the third safe output or the fourth safe output in the event of an object intrusion, wherein the third partial protective fields and fourth partial protective fields are arranged in alternating succession along a second line transverse to the first line and wherein, in particular, the safety controller (36) is configured to evaluate a time sequence of the safe output signals at the third safe output and the fourth safe output in order to determine a position of the object (20) along the second line.

14. The monitoring device according to any of claims 1 to 12, wherein the first control and evaluation unit (28) is configured to switch the protective field monitoring to a third protective field with a plurality of third partial protective fields and a fourth protective field with a plurality of fourth partial protective fields, wherein the third partial protective fields and fourth partial protective fields are arranged in alternating succession along a second line transversely to the first line, and wherein in particular the safety controller (36) is configured to evaluate a time sequence of the safe output signals in order to determine a position of the object along the second line.

15. A method for safe object tracking of an object (20) in a monitoring area (18) by means of at least one first safe optoelectronic sensor (10, 10a) that monitors, in a first control and evaluation unit (28), at least a first protective field (38A) and a second protective field (38B) in the monitoring area (18) for object intrusions using a received signal from a light receiver (26) and outputs a safe output signal at a first safe output (32) in the event of an object intrusion into the first protective field (38A) and outputs a safe output signal at a second safe output (34) in the event of an object intrusion into the second protective field (38B), wherein the safe output signals are evaluated, wherein the terms "safe" and "safety" mean that measures are taken to control faults up to a specified safety level and to comply with regulations of a safety standard for machine safety or electro-sensitive protective equipment, and wherein the first protective field (38A) and the second protective field (38B) are each a partial area of the monitoring area (18) configured by means of geometric specifications in the first control and evaluation unit (28) or imported via an interface (30), wherein the first protective field (38A) has a plurality of first partial protective fields and the second protective field (38B) has a plurality of second partial protective fields, and wherein the first partial protective fields and the second partial protective fields are arranged alternately in succession along a line.