Method for the object-recognition operation of a lighting device and lighting device

Abstract

The invention relates to a method for object-recognizing operation of a lighting device (1) and a lighting device (1) for recognizing moving objects (2).

Description

[0001] The invention relates to a method for the object-detecting operation of a lighting device. Furthermore, the invention relates to a lighting device for detecting moving objects and to a lighting system comprising at least two or more lighting devices according to the invention.

[0002] Camera-based object recognition systems, for example, have become known from the state of the art. In these systems, at least one camera usually creates a series of images by receiving ambient light. This series is continuously evaluated, with objects being recognized using appropriate image processing algorithms. In particular, vehicle license plates can be read and used for their recognition. While such systems generally function reliably, they require the presence of sufficiently strong ambient light, which is why additional street lighting systems must be used. A further disadvantage of camera-based object recognition systems is the high cost and the dependence on the clearest possible visibility, which can be obscured by soiling of the camera lens or other weather influences such as heavy rain, glare, etc.

[0003] An object of the invention is therefore to provide a possibility for the efficient detection of objects which overcomes the above-mentioned disadvantages.

[0004] In a first aspect of the invention, this object is achieved by a method for the object-detecting operation of a lighting device, in which objects are moved with respect to the lighting device, wherein the lighting device has a number of LED light sources which are suitable both for emitting light in an emitting mode and for converting light incident on the LED light sources into an electrically measurable signal in a receiving mode, wherein the method comprises the following steps: a) Operating a first subgroup of the number of LED light sources in the emission mode, in which light is emitted by the LED light sources of the first subgroup into the surroundings of the lighting device, b) Operating at least a second subgroup of the number of LED light sources in the reception mode, wherein this reception mode occurs at least temporarily within the duration of the emission mode, so that light emitted by the first subgroup into the surroundings, which light is reflected by a moving object located in the surroundings toward the LED light sources of the second subgroup, is at least partially detected and converted into an electrically measurable actual signal, c) Evaluating the temporal profile of the actual signal measured by the second subgroup, wherein the profile of the temporal change in the actual signal is used to infer the passage of the object,by storing temporal progressions of target signals to predefined objects in a data memory, hereinafter referred to as object profiles, and whereby, depending on the result, a comparison of the temporal progression of the actual signal with the object profiles is used to infer the passage of an object stored in the data memory.

[0005] The inventive solution allows object detection to be performed by the lighting devices themselves, meaning that a sufficient amount of light is available for object detection, and the use of additional means such as cameras can be dispensed with. The object-detecting lighting devices mentioned can, for example, replace conventional lighting devices or be used in combination with conventional lighting devices, e.g., distributed along routes.

[0006] The LED light sources can also be light source arrangements that combine several LEDs (dies) on a common package. Each sub-group has at least one LED light source. The fact that the receiving mode of individual LED light sources occurs at least during the time in which other LED light sources are in emitting mode ensures that a sufficient amount of light is available to be reflected by a passing object. Of course, the light emitted by other light sources and reflected by the object towards the receiving LED light sources can also be taken into account. An increased amount of reflected light thus can be used advantageously, since the assignment of measured signals to object profiles can primarily be achieved via the temporal progression, in particular the relative rates of change of the actual signals.Examples that explain in more detail how signals are assigned to objects follow in the description.

[0007] In particular, the lighting device can be designed to be operated in a road tunnel. This application is particularly advantageous due to the high sensitivity of tunnels to traffic volumes, the particular hazards inherent in tunnels, and the inherent need for continuous lighting in tunnels.

[0008] Furthermore, it can be provided that, based on the temporal course of the actual signal detected by a lighting device (or by comparison with a corresponding target signal), conclusions are drawn about the driving speed and direction of travel of an object passing the lighting device during the passage.

[0009] Furthermore, it can be provided that at least two or more lighting devices suitable for object-detecting operation are operated, which are distributed at a distance from one another along the course of a road, so that an object traveling along the course of the road, in particular a motor vehicle, which passes the lighting devices in question one after the other, is detected one after the other by the lighting devices, wherein the lighting devices are connected to one another directly or indirectly by information technology, so that by evaluating the objects detected by the lighting devices and comparing them with the distance between the lighting devices, the average speed traveled by the detected objects is deduced and it is determined on which section of the road the detected objects are located.In particular, a comparison of the average speed with the actual speed can be made in order to draw conclusions about the uniformity of traffic.

[0010] In particular, it can be provided that, during operation of the lighting devices, new object profiles are created by these lighting devices and exchanged between the lighting devices. The lighting device arranged first along a road creates a new object profile upon detecting and assigning a temporal profile of an actual signal to an object that does not resemble any object stored in the data memory. This object profile is communicated to the lighting devices arranged downstream along the road as an expected passing object. In this way, even previously unknown objects can be detected.

[0011] Furthermore, it can be provided that the average speed and / or position of an object is automatically recorded by sequentially detecting an object at different sections of a road, and on this basis, conclusions are drawn about the traffic situation, in particular about accidents or traffic jams. Preferably, a large number of objects, particularly preferably all objects passing the lighting devices, are detected in this way. In particular, this can be used to record distance data or perform traffic analyses. For example, the composition of the traffic can be analyzed. These additional parameters can also be efficiently collected using machine learning methods and applied statistical methods.

[0012] In particular, it can be provided that the data storage has stored a number of object profiles of motor vehicles, whereby this number is greater than ten, the number can also be determined dynamically and variably and utilize the storage volume of a data storage accordingly.

[0013] Furthermore, it can be provided that the total number of LED light sources consists exclusively of LED light sources from the first subgroup and the second subgroup. This means that all LED light sources are used, thus ensuring particularly high utilization.

[0014] In particular, it can be provided that the receiving operation of the second subgroup takes place exclusively during the transmission operation of the first subgroup. This ensures that sufficient light is always emitted into the surrounding area.

[0015] When this application refers to "light," it refers to light visible to the human eye.

[0016] Furthermore, it can be provided that the assignment of the LED light sources to the subgroups varies over time, specifically in such a way that, in order to distribute the thermal load on the LED light sources during transmission mode, they are alternately switched to reception mode. The duration of transmission mode can, for example, be evenly distributed among the LED light sources. It can be provided that at least one LED light source is assigned to this mode during reception mode, but two or more LED light sources can also be active in reception mode simultaneously.

[0017] In particular, it can be provided that the LED light sources are controlled in such a way that, from the perspective of the human eye, the lighting device emits light continuously. This means that the change in the operating modes of the individual LED light sources occurs in such a way that this is not perceptible to the human eye. The light emission of the lighting device creates the impression that it occurs continuously and evenly, without the change of individual LED light sources between emission and reception mode being visible in the overall emission. This can, for example,This can be achieved by a correspondingly high clock rate for changing operating states, for example, a clock rate above 60 Hz, in particular above 100 Hz, particularly preferably above 200 Hz, and / or by ensuring that when an LED light source switches to receive mode, another LED light source takes over the corresponding emission by switching to emission mode. The frequency is therefore advantageously selected so that the clock rate of the light emission is above the perception threshold of the human eye.

[0018] Furthermore, it can be provided that, during the transmission period of the first subgroup according to step a), all LED light sources are temporarily in transmission mode, and only individual LED light sources are temporarily switched to reception mode according to step b). As already mentioned, reception mode and / or transmission mode can be synchronized. For example, reception mode can occur for 50% of the transmission period. By selecting the appropriate timing conditions, light emission can be optimized.

[0019] In particular, it can be provided that the LED light sources are arranged in switching lines, with at least two LED light sources connected in parallel to each other per switching line. For example, three LED light sources per line can be connected in parallel, and four lines can be provided per lighting device, with three lines being operated in transmission mode and one line in reception mode, for example.

[0020] Furthermore, each lighting device can be provided with a temperature sensor for estimating the operating temperature of the LED light sources, with measures for controlling the current through the LED light sources being provided depending on the operating temperature. For example, a current reduction can be provided if overheating is imminent. Likewise, a current increase can be provided if there is no danger of overheating yet, but the luminous efficacy of the LED light sources decreases due to higher temperatures and is to be compensated by increasing the current.

[0021] In particular, it can be provided that each LED light source is connected to a controllable cooling unit, in particular a heat sink with a controllable fan, wherein the fan is controlled depending on the detected temperature.

[0022] Furthermore, it can be provided that each lighting device has a microcontroller unit, wherein the LED light sources are controlled via a PWM output of the microcontroller unit, in such a way that a microcontroller can switch between the emission mode and the reception mode by PWM-dependent control of two transistors, wherein a transimpedance amplifier is connected to the LED light sources for evaluating the state of the LED light sources in the reception mode.

[0023] In a further aspect, the invention relates to a lighting device for detecting moving objects, wherein the lighting device has a number of LED light sources which are configured both to emit light in an emitting mode and to convert light incident on the LED light sources into an electrically measurable signal in a receiving mode, wherein the LED light sources are divided into two subgroups, namely a first subgroup for an emitting mode, in which light is emitted by the LED light sources of the first subgroup into the environment of the lighting device, and a second subgroup for a receiving mode, wherein this receiving mode takes place at least temporarily within the duration of the emitting mode, so that light emitted by the first subgroup into the environment,which is reflected by a moving object in the environment towards the LED light sources of the second subgroup, is at least partially detected and can be converted into an electrically measurable actual signal, wherein the lighting device is configured to vary the assignment of the LED light sources to the subgroups over time, wherein the lighting device further comprises an evaluation unit configured to detect the temporal progression of the actual signal measured by the second subgroup and to infer the passage of an object by storing temporal progressions of target signals for predefined objects in a data memory of the evaluation unit, hereinafter referred to as object profiles, and inferring the passage of a stored object depending on the result of a comparison of the temporal progression of the actual signal with the temporal progressions of the object profiles.

[0024] In addition, the invention relates to a lighting system comprising at least two or more lighting devices according to the invention, wherein these lighting devices are arranged at a distance from one another along a road course and are connected to one another directly or indirectly by information technology for the evaluation of the detected objects.

[0025] The invention is explained in more detail below with reference to exemplary and non-limiting embodiments, which are illustrated in the figures. Figure 1 a schematic representation of a lighting device according to the invention for applying a method according to the invention, Figure 2 a scenario of use of lighting devices according to the invention in a tunnel, which together form a lighting system, Figures 3 to 5 exemplary reflection measurements of various vehicles, Figure 6 an assignment of a curve to the vehicle VW Golf 7 Variant,

[0026] In the following figures, unless otherwise stated, the same reference symbols denote the same features.

[0027] Fig. 1 shows a schematic representation of an inventive lighting device 1 for applying an inventive method. The lighting device 1 is configured to detect moving objects by having a number of LED light sources 3. These are Fig. 1 shown in two circuits shown side by side, which can essentially be constructed identically. The LED light sources 3 are configured both for emitting light in a emitting mode B1 (see left circuit) and for converting light incident on the LED light sources 3 into an electrically measurable signal Sist in a receiving mode B2 (see right circuit).

[0028] The LED light sources 3 are divided into two subgroups, namely a first subgroup 3_1 (left) for a radiation mode B1, in which light is emitted by the LED light sources 3 of the first subgroup 3_1 into the environment 1a of the lighting device 1, and a second subgroup 3_2 (right) for a reception mode B2. The reception mode B2 occurs at least temporarily within the duration of the radiation mode B1, so that light emitted by the first subgroup 3_1 into the environment 1a, which light is reflected by a moving object 2 located in the environment 1a toward the LED light sources 3 of the second subgroup 3_2, can be at least partially detected and converted into an electrically measurable actual signal Sact.The lighting device 1 is configured to vary the assignment of the LED light sources 3 to the subgroups 3_1 and 3_2 over time, wherein the lighting device 1 further comprises an evaluation unit 7a which is configured to record the temporal profile of the actual signal Sist measured by the second subgroup and to infer the passage of an object 2. This is achieved by storing temporal profiles of target signals for predefined objects 2, hereinafter referred to as object profiles Ssoll_x, in a data memory 4 of the evaluation unit 7a, and inferring the passage of a stored object 2 depending on the result of a comparison of the temporal profile of the actual signal Sist with the temporal profiles of the object profiles Ssoll_x.

[0029] The lighting device 1 enables a method according to the invention for the object-detecting operation of a lighting device 1, which objects 2 are moved with respect to the lighting device 1, wherein the lighting device 1 has the number of LED light sources 3, which are suitable both for emitting light in an emitting mode B1 and for converting light incident on the LED light sources 3 into an electrically measurable signal Sout in a receiving mode B2, wherein the method comprises the following steps: a) Operation of the first subgroup 3_1 of the number of LED light sources 3 in the emission mode B1, in which light is emitted by the LED light sources 3 of the first subgroup 3_1 into the environment 1a of the lighting device 1, b) Operation of at least the second subgroup 3_2 of the number of LED light sources 3 in the reception mode B2, wherein this reception mode B2 takes place at least temporarily within the duration of the emission mode B1, so that light emitted by the first subgroup 3_1 into the environment 1a, which light is reflected by a moving object 2 located in the environment 1a towards the LED light sources 3 of the second subgroup 3_2, is at least partially detected and converted into an electrically measurable actual signal Sist, c) Evaluation of the temporal profile of the actual signal Sist measured by the second subgroup 3_2, wherein the profile of the temporal change of the actual signal Sist is used to infer the passing of the object 2 is used,in that temporal profiles of target signals to predefined objects 2_x are stored in a data memory 4, hereinafter referred to as object profiles Ssoll_x, and wherein, depending on the result, a comparison of the temporal profile of the actual signal Sist with the object profiles Ssoll_x is used to infer the passing of an object 2 stored in the data memory 4.

[0030] In particular, it can be provided that the lighting device 1 is operated in a road tunnel 5.

[0031] Fig. 2shows an application scenario of lighting devices 1 according to the invention in a tunnel 5, wherein the lighting devices 1 together form a lighting system 8. In this case, at least two or more, in the present case e.g.exactly three lighting devices 1 suitable for object-detecting operation are operated, which are distributed at a distance from one another along a road course 5a, so that an object 2 traveling along the road course 5a, in particular a motor vehicle, which passes the relevant lighting devices 1 one after the other, is detected one after the other by the lighting devices 1, wherein the lighting devices 1 are connected to one another directly or indirectly by information technology, so that by evaluating the objects 2 detected by the lighting devices 1 and comparing them with the distance between the lighting devices 1, the average speed traveled by the detected objects 2 is deduced and it is determined on which section of the road course 5a the detected objects 2 are located.

[0032] With a view to Fig. 1It should be mentioned that during operation of the lighting devices 1, new object profiles Ssoll_x can be created by these lighting devices 1 and exchanged with one another by the lighting devices 1, wherein the lighting device 1 arranged first along a road course 5a creates a new object profile Ssoll_x upon detection and assignment of a temporal profile of an actual signal Sist to an object 2 which does not resemble any object 2 stored in the data memory 4, and communicates this object profile Ssoll_x as an expected passing object 2 to the lighting devices 1 downstream along the road course 5a.In particular, it can be provided that the average speed and / or positions of the object 2 are automatically detected by successive detection of an object 2 at different sections of a road course 5a, based on which the traffic situation, in particular accidents or traffic jams, is deduced, wherein preferably a plurality of objects 2, particularly preferably all objects 2 passing the lighting devices 1, are detected in this way.

[0033] Furthermore, it can be provided that the data memory 4 has stored a number of object profiles Ssoll_x of motor vehicles, wherein this number is greater than ten.

[0034] In particular, it can be provided that the total number of LED light sources 3 consists exclusively of LED light sources of the first subgroup 3_1 and the second subgroup 3_2.

[0035] Furthermore, it can be provided that the receiving operation B2 of the second subgroup 3_2 takes place exclusively during the duration of the transmitting operation B1 of the first subgroup 3_1.

[0036] In particular, it can be provided that the allocation of the LED light sources 3 into the subgroups 3_1, 3_2 varies over time, in such a way that in order to distribute the thermal load on the LED light sources 3 arising in the emission mode B1, these are alternately switched to the reception mode B2.

[0037] Furthermore, it can be provided that the LED light sources 3 are controlled in such a way that, from the perspective of a human eye, uninterrupted light emission occurs through the lighting device 1. In particular, it can be provided that, within the duration of the emission mode B1 of the first subgroup according to step a), all LED light sources 3 are temporarily in emission mode B1 and only individual LED light sources 3 are temporarily switched to reception mode B2 according to step b).

[0038] In Fig. 1 It can be seen that the LED light sources 3 can be arranged in (e.g., four) switching strands 3s, wherein at least two (in this case, exactly three) LED light sources 3 are connected in parallel to each other in each switching strand 3s. The number and arrangement of the LED light sources can be adapted depending on the respective application profile of the lighting device 1 and the properties of the LED light sources.

[0039] Furthermore, it can be provided that each lighting device 1 has a temperature sensor 6 for estimating the operating temperature of the LED light sources 3, wherein measures for controlling the current through the LED light sources 3 are provided depending on the operating temperature.

[0040] In addition, it can be provided that each LED light source 3 is connected to a controllable cooling unit 7, in particular a heat sink with a controllable fan, wherein the fan is controlled depending on the detected temperature.

[0041] Furthermore, it can be provided that each lighting device 1 has a microcontroller unit 7, wherein the LED light sources 3 are controlled via a PWM output of the microcontroller unit 7, in such a way that a microcontroller can switch between the emission mode B1 and the reception mode B2 by PWM-dependent control of two transistors Q1, Q2, wherein a transimpedance amplifier TIA is connected to the LED light sources 3 for evaluating the state of the LED light sources 3 in the reception mode B2.

[0042] Figure 2 shows a lighting system 8 comprising at least two or more lighting devices 1 according to the invention, wherein these lighting devices 1 are arranged at a distance from one another along a road course 5a and are connected to one another directly or indirectly by information technology for the evaluation of the detected objects 2.

[0043] The invention is described below in additional terms, including specific embodiments, which are to be interpreted merely as examples and not as limiting.

[0044] The invention enables detection, classification and real-time tracking of moving vehicles in tunnel environments using LED light sources as transceivers.

[0045] The invention offers a method for detecting, classifying, and tracking moving vehicles in real time in a tunnel environment using commercially available white LEDs. These LEDs are used as both illumination and a sensor element to exploit changes in the reflected visible spectrum (light). A demonstrator developed for this setup includes, for example, 12 white LEDs (Lumileds LMZ7-QW40) that are continuously alternately connected as transmitters and receivers. To increase the sensitivity of the respective LEDs, unlike in previous experiments, 3 LEDs were grouped in parallel to form a total of 4 receivers in order to sum the resulting photocurrent. Depending on the control, these LED groups are switched to transmit (light emitting) or receive (light capturing) mode.

[0046] The functioning of the circuits according to Fig. 1will be discussed in more detail below: In the transmission phase, the output of the LED driver circuit shown is activated, transistor Q1 conducts, Q2 is off, and thus has a high resistance. This creates a current flow through the LED, and photons are emitted. The buck driver circuit responsible for the constant current consists of the MAX20050ATC / V+ buck converter IC in combination with passive components (capacitors, resistors, coil, diode), which can deliver an LED current of up to 2A.

[0047] In the receiver phase, Q1 is off and Q2 is on, thus switching the LED into a quasi-short-circuit state. When photons are incident, the resulting photocurrent is converted into a proportional voltage by the transimpedance amplifier and then read into an ADC pin of microcontroller 7. To keep switching losses, and thus the switching time of the power MOSFETs, as low as possible, gate drivers are connected upstream of the gate inputs. Before the PWM signal from microcontroller 7 reaches the gate driver, it is converted to the appropriate 5V level by a line driver.

[0048] Since temperature can have a significant influence on the behavior of the LEDs during both transmit and receive modes, a temperature sensor 6 was placed on the PCB. This sensor is continuously read by the microcontroller 7 via a serial I2C data bus to better evaluate the behavior of the LED light sources 3. Additionally, a heat sink is provided for each LED light source.

[0049] The demonstrator presented here is powered by a 12V DC switching power supply. A linear voltage regulator circuit based on the TLV76701DRVR LDO regulator developed by Texas Instruments is implemented to supply the 3.3V supply for the sensors and transimpedance amplifier stages. In this LDO IC, the output voltage is set by a voltage divider and is thus variably definable. In this case, the ratio of the resistors was designed to guarantee a constant output voltage of 3.3V.

[0050] The hardware is controlled by the Nucleo-WB55 microcontroller board from STMicroelectronics, which features a 32-bit ARM Cortex-M4 processor core and an ARM Cortex-M0+ radio co-processor clocked at 64 MHz and equipped with sufficient peripherals (timer, ADC, DMA, etc.). On the software side, the hardware is controlled in two phases. In the so-called "illumination phase" (transmission phase), the four LED groups provide illumination. The dimming value, or PWM duty cycle, can be freely selected. In the "measurement phase" (reception phase), three of the four LED groups emit light, while one LED group is switched to quasi-short-circuit mode as described above, subsequently converting the received light reflections into a proportional voltage.These two phases can be switched cyclically so that a different LED group takes over the reflection measurement during each cycle.

[0051] Experiments were conducted using a demonstrator to demonstrate the feasibility of the invention and from which object profiles Ssoll_x were derived, as discussed below. Those skilled in the art are able to select suitable algorithms that, by comparing the similarity of measured actual signals with the object profiles Ssoll_x, infer a match with the object profile Ssoll_x and thus with a specific vehicle, or allow such an inference. Previously known typical deviations of the actual signals from the object profiles Ssoll_x can also be taken into account, for example, by taking into account that at higher vehicle speeds the signal curve is compressed and at lower speeds the signal curve expands over time (in the sense of signal stretching). Ratios of high to low values ​​can also be determined.In particular, proportions of the measured intensity values ​​can be recognized, which are largely retained even when the speed changes. Fig. 2 shows, for example, the process of normalization and classification of three recorded actual signals (ie detected reflections).

[0052] Based on an implementation concept, different vehicles are detected and classified to enable real-time tracking. To demonstrate this with initial measurements, the following scenarios were simulated in an underground parking garage: Car traffic (forward and backward)

[0053] The demonstrator (i.e. a lighting device 1 according to Fig. 1) was mounted on the ceiling of an underground parking garage and aligned in line of sight to the "surface of interest" at a distance of 3m. Subsequently, the intensity of the reflection was measured and evaluated over time by the presented demonstrator in various scenarios. The voltage values ​​of the respective reflection measurements for each LED group were averaged for evaluation.

[0054] In this scenario, car traffic was simulated by different vehicles (VW Golf 7 Variant - silver, Seat Ibiza - black and BMW 3 Series - grey), speeds (5, 10 and 20 km / h) and directions of travel (forward and reverse). The results in Table 1, Table 2 and Table 3 show the temporal progression of the measured reflections at different speeds. Depending on the shape and paint finish of the car, the measurements exhibit different progressions. Especially with the silver and grey paint finishes of the VW and the BMW, the shape of the car could be clearly reconstructed through the reflection progression. It should be noted that, as in Figure 6(left) shows that the roof of the BMW is covered with a black film. As expected, the intensity of the reflections is highest on the hood and roof of the respective cars, whereas only a low intensity of reflection was observed on the windscreen, rear window, and roof. For these measurements, the reflection measurements of the two middle LED groups were used. The reflection measurements for the aforementioned vehicles are shown in Figures 3 to 5 shown. Fig. 3 shows reflection measurements for the mentioned golf cart, Fig. 4 concerns the BMW car mentioned and Fig. 5 concerns the mentioned Seat car.

[0055] The Figures 3 to 5 Recorded reflections can now be classified, and parameters such as forward or backward direction and speed can be derived. To illustrate the possibilities in detail, the recorded reflections will now be discussed as examples.

[0056] With regard to the presentation of a Fig. 6 In the forward motion of the VW Golf 7 Variant shown, it is clear that the curve path can be directly assigned to the actual dimensions and shape of the vehicle, or corresponds to it. The recorded curve path can also be used to approximate the vehicle's speed, given known vehicle parameters (e.g., roof length).

[0057] To illustrate this, a calculation is made using the Figure 6 The curve shown was used. Data: Length of the roof approximately 2.2 m. Duration of the stable reflection from the roof approximately 2600 ms = 2.6 seconds. Using the formula speed = distance / time, the following results: V 0 = 2,2 m Länge Dach / 2,6 s = 0,84 m / s ∼ 3 km / h

[0058] In comparison, at 20 km / h: V 0 = 2,2 m Länge Dach / 0,3 s = 7,3 m / s ∼ 26 km / h

[0059] This shows that the speed of the vehicle can be well approximated.

[0060] In comparison to the VW Golf 7 Variant, the BMW 3 has the following assignment of the curve to the vehicle, see Figure 7 This vehicle already represents a scenario in which the vehicle has different colors and materials. Nevertheless, clear features for classification can be derived, such as the windshield and rear window.

[0061] The final example, a small car of the brand Seat Ibiza, see Figure 7 , has two special features: Firstly, it is a small car, and secondly, the vehicle has a completely black paint finish. Figure 8 shows, analogous to the previous examples, the assignment of the curve to the vehicle.

[0062] These examples allow two important conclusions regarding evaluation and classification. First, as can be seen from the recorded reflections of the black vehicle roof, even black vehicles can be detected. Second, not only can the reflection of the vehicle be detected, but also analysis is possible in areas where little light is reflected back to the receiver due to the color and shape of the vehicle. In such cases, the difference from the "usual" reflection of the road surface is ultimately recorded. In the case of such difficult-to-distinguish vehicles, an object profile can therefore simply be created that serves as a collective profile for all vehicles whose passing is detected, but whose assignment to a specific vehicle model is not possible.In addition, the respective specific vehicle reflection can be stored and transmitted to subsequent lighting devices 1 in order to check whether a similar vehicle reflection profile passes the respective subsequent lighting devices 1 and thus to track the vehicle.

[0063] Based on the reflections of each vehicle, unique characteristics (local maxima / minima, e.g., gap between tractor and trailer, etc.) can now be used for normalizing the temporal progression (interpolation) and the subsequent application of supervised learning (machine learning) methods. Our normalization of the temporal progression now allows or enables a speed-independent classification of the aforementioned vehicles.

[0064] With a view to Fig. 2The possibility of real-time vehicle tracking should be emphasized again. With such a method, traffic flow, the average speed of individual vehicles, the composition of the traffic (trucks, cars, etc.), the number of vehicles, the time of entry and exit, traffic jam detection, etc. can be mapped. The accuracy of the system can be increased modularly with the help of additional lighting elements.

[0065] The following possibilities are particularly advantageous aspects of the invention: .) Recording of the temporal progression of reflections or shading (absorption) of road reflections caused by vehicles passing through (see curve of the Seat Ibiza) using a light that preferably operates groups of parallel LEDs alternately as a sensor or as a light source (LED as a transceiver). .) Detection of the presence of vehicles using an LED as a transceiver, for example by exceeding or falling below a threshold value. This threshold value is continuously calculated on the basis of or in comparison to the received reflections from the environment (e.g. road surface). .) Classification of vehicles using an LED as a transceiver. The type of vehicle is determined by extracting parameters such as the duration of the reflections, shape, intensity and number of local minima and maxima, and the relationship between sections of the received reflection curve, e.g. car, lorry, etc.) Approximation of the vehicle speed using an LED as a transceiver. The vehicle speed can be approximated based on the time course of the reflections. .) Distance measurement or detection of a distance between vehicles being exceeded using an LED as a transceiver. The distance between individual vehicles can be determined based on presence detection. I.e. presence detection of vehicle A (duration x), detection that no vehicle is present (duration y), presence detection of vehicle B (duration z). It follows that if duration y is too short, the distance between the vehicles can be detected to be too small. .) Normalization of the time courses of the reflection values ​​recorded using an LED as a transceiver. By normalizing based on the times at which the respective threshold values ​​are exceeded or undershot, a signature is generated that is independent of the vehicle speed. .) Real-time tracking of vehicles using multiple LEDs as transceiver lighting fixtures.

[0066] By recording the signature (or object profile, see previous point) at the beginning of the monitored section (e.g., tunnel entrance) and continuously comparing the signatures recorded on additional LEDs acting as transceiver lighting fixtures, the signature (or vehicle) is tracked in real time along the monitored section, and the resulting parameters (e.g., entry and exit) are derived (see Figure 11). Furthermore, based on typical movement patterns, i.e., temporal or physical relationships, it can be detected that a vehicle has not reached a certain location within the expected time, e.g., the vehicle has entered a breakdown bay. .) Detection of vehicle standstill or detection of objects on the road using an LED as a transceiver. If the specified threshold is exceeded or undershot, but no corresponding change in reflection is recorded over time, a static change can be detected, i.e., there is an object on the road or a vehicle has come to a standstill.

[0067] The invention is not limited to the embodiments shown, but is defined by the entire scope of the claims. Individual aspects of the invention or the embodiments may also be taken up and combined with one another. Any reference symbols in the claims are exemplary and serve only to facilitate the readability of the claims, without limiting them.

Claims

1. Method for the object-recognizing operation of a lighting device (1), which objects (2) are moved with respect to the lighting device (1), wherein the lighting device (1) has a number of LED light sources (3) which are suitable both for emitting light in an emitting mode (B1) and for converting light incident on the LED light sources (3) into an electrically measurable signal (Sout) in a receiving mode (B2), wherein the method comprises the following steps: a) operating a first subgroup (3_1) of the number of LED light sources (3) in the emitting mode (B1), in which light is emitted by the LED light sources (3) of the first subgroup (3_1) into the environment (1a) of the lighting device (1), b) operating at least a second subgroup (3_2) of the number of LED light sources (3) in the receiving mode (B2), wherein this receiving mode (B2) is at least temporarily within the time duration of the Radiation operation (B1) takes place,such that light emitted by the first subgroup (3_1) into the environment (1a), which is reflected by a moving object (2) located in the environment (1a) towards the LED light sources (3) of the second subgroup (3_2), is at least partially detected and converted into an electrically measurable actual signal (Sist), c) evaluation of the temporal course of the actual signal (Sist) measured by the second subgroup (3_2), wherein the course of the temporal change of the actual signal (Sist) is used to infer the passing of the object (2) by storing in a data memory (4) temporal courses of target signals for predefined objects (2_x), hereinafter referred to as object profiles (Ssoll_x), and wherein, depending on the result, a comparison of the temporal course of the actual signal (Sist) with the object profiles (Ssoll_x) on the passing of an object (2) stored in the data memory (4) is inferred., 2. Method according to claim 1, wherein the lighting device (1) is operated in a road tunnel (5).

3. Method according to claim 1 or 2, wherein, on the basis of the time course of the actual signal (Sist) detected by a lighting device (1), conclusions are drawn about the driving speed and direction of travel of an object (2) passing the lighting device (1) during the passing.

4. Method according to one of the preceding claims, wherein at least two or more lighting devices (1) suitable for object-detecting operation are operated, which are distributed at a distance from one another along a road course (5a), so that an object (2) traveling along the road course (5a), in particular a motor vehicle, which passes the relevant lighting devices (1) one after the other, is detected one after the other by the lighting devices (1), wherein the lighting devices (1) are directly or indirectly connected to one another by information technology, so that by evaluating the objects (2) detected by the lighting devices (1) and comparing them with the distance between the lighting devices (1), the average speed traveled by the detected objects (2) is deduced and it is determined on which section of the road course (5a) the detected objects (2) are located.

5. Method according to claim 4, wherein during operation of the lighting devices (1) new object profiles (Ssoll_x) are created by these lighting devices (1) and exchanged with one another by the lighting devices (1), wherein the lighting device (1) arranged first along a road course (5a) creates a new object profile (Ssoll_x) upon detection and assignment of a temporal profile of an actual signal (Sist) to an object (2) which does not resemble any object (2) stored in the data memory (4) and communicates this object profile (Ssoll_x) as an expected passing object (2) to the lighting devices (1) arranged downstream along the road course (5a).

6. Method according to claim 4 or 5, wherein the average speed and / or positions of the object (2) are automatically detected by successive detection of an object (2) at different sections of a road (5a), based on which the traffic situation, in particular accidents or traffic jams, is deduced, wherein preferably a plurality of objects (2), particularly preferably all objects (2) passing the lighting devices (1), are detected in this way.

7. Method according to one of the preceding claims, wherein the data memory (4) has stored a number of object profiles (Ssoll_x) of motor vehicles, wherein this number is greater than ten, or the number can also be set dynamically and variably.

8. Method according to one of the preceding claims, wherein the total number of LED light sources (3) consists exclusively of LED light sources of the first subgroup (3_1) and the second subgroup (3_2).

9. Method according to one of the preceding claims, wherein the receiving operation (B2) of the second subgroup (3_2) takes place exclusively during the duration of the transmitting operation (B1) of the first subgroup (3_1).

10. Method according to one of the preceding claims, wherein the allocation of the LED light sources (3) into the subgroups (3_1, 3_2) varies over time, in such a way that in order to distribute the thermal load of the LED light sources (3) arising in the emission mode (B1), these are alternately switched to the reception mode (B2).

11. Method according to one of the preceding claims, wherein the LED light sources (3) are controlled in such a way that, from the perspective of a human eye, uninterrupted light emission occurs through the lighting device (1).

12. The method according to claim 11, wherein within the duration of the emission mode (B1) of the first subgroup according to step a), all LED light sources (3) are temporarily in the emission mode (B1) and only individual LED light sources (3) are temporarily switched to the reception mode (B2) according to step b).

13. Method according to one of the preceding claims, wherein the LED light sources (3) are arranged in switching strands (3s), wherein at least two LED light sources (3) are connected in parallel to one another in each switching strand (3s).

14. Method according to one of the preceding claims, wherein each lighting device (1) has a temperature sensor (6) for estimating the operating temperature of the LED light sources (3), wherein measures for controlling the current through the LED light sources (3) are provided as a function of the operating temperature.

15. The method according to claim 14, wherein each LED light source (3) is connected to a cooling unit (7), which is in particular a heat sink with a controllable fan, wherein the fan is controlled as a function of the detected temperature.

16. Method according to one of the preceding claims, wherein each lighting device (1) has a microcontroller unit (7), wherein the LED light sources (3) are controlled via a PWM output of the microcontroller unit (7), in such a way that a microcontroller can switch between the emission mode (B1) and the reception mode (B2) by PWM-dependent control of two transistors (Q1, Q2), wherein a transimpedance amplifier (TIA) is connected to the LED light sources (3) for evaluating the state of the LED light sources (3) in the reception mode (B2).

17. Lighting device (1) for detecting moving objects, wherein the lighting device (1) has a number of LED light sources (3) which are configured both to emit light in an emitting mode (B1) and to convert light incident on the LED light sources (3) into an electrically measurable signal (Sist) in a receiving mode (B2), wherein the LED light sources (3) are divided into two subgroups, namely a first subgroup (3_1) for an emitting mode (B1), in which light is emitted by the LED light sources (3) of the first subgroup (3_1) into the environment (1a) of the lighting device (1), and a second subgroup (3_2) for a receiving mode (B2), wherein this receiving mode (B2) takes place at least temporarily within the duration of the emitting mode (B1), so that light emitted by the first subgroup (3_1) into the environment (1a),which is reflected by a moving object (2) located in the environment (1a) towards the LED light sources (3) of the second subgroup (3_2), is at least partially detected and can be converted into an electrically measurable actual signal (Sactual), wherein the lighting device (1) is designed to vary the assignment of the LED light sources (3) to the subgroups (3_1, 3_2) over time, wherein the lighting device (1) further comprises an evaluation unit (7a) which is designed to detect the temporal progression of the actual signal (Sactual) measured by the second subgroup () and to infer the passing of an object (2) by storing temporal progressions of target signals for predefined objects (2) in a data memory (4) of the evaluation unit (7a), hereinafter referred to as object profiles (Ssoll_x),and depending on the result, a comparison of the temporal course of the actual signal (Sist) with the temporal courses of the object profiles (Ssoll_x) is used to infer the passing of a stored object (2).

18. Lighting system (8) comprising at least two or more lighting devices (1) according to claim 17, wherein these lighting devices (1) are arranged at a distance from one another along a road course (5a) and are directly or indirectly connected to one another by information technology for the evaluation of the detected objects (2).

Images