Method for transmitting data from an ultrasonic system to a data processing device via a vehicle data bus, ultrasonic sensor, computer system and sensor system - Patents.com

Abstract

The object of the present invention is to further increase the degree of data compression in ultrasonic sensors of a vehicle without compromising the reliability of obstacle recognition and obstacle type. A further object is to further reduce the bus bandwidth requirements for transmitting measurement data from the ultrasonic sensor system to a computer system or to increase the efficiency of data transmission. The present invention relates to a method for transmitting data from an ultrasound system having at least one ultrasound transmitter and ultrasound receiver to a data processing device via a vehicle data bus, wherein a plurality of predetermined signal profile characteristics are extracted from echo signals received by the at least one ultrasound receiver of the ultrasound system, echo signal data representing the plurality of signal profile characteristics extracted from the echo signals is generated, and the echo signal data is transmitted from the ultrasound system to the data processing device via the vehicle data bus.

Description

[Technical field]

[0001] The present invention relates to a method for transmitting data from an ultrasound system having at least one ultrasound transmitter and one ultrasound receiver to data processing via a vehicle data bus (unidirectional or bidirectional single, dual or multi-wire data bus, possibly differential), and in particular to a method for classifying echo signals of an ultrasound system in a vehicle, with a view to data compression and transmission of this compressed data from the ultrasound system to a data processing device. [Background technology]

[0002] Conventionally, ultrasonic systems are already used in vehicles for environmental recognition, where ultrasonic burst signals are emitted by at least one ultrasonic transmitter which are reflected from an obstacle (generally speaking, an object) and then received by at least one ultrasonic receiver of the ultrasonic system. Mostly, so-called ultrasonic transducers are implemented as ultrasonic transmitters / ultrasonic receivers, which in a first phase act as transmitters and in a later phase at the reference distance act as ultrasonic receivers.

[0003] The demand for recognition of obstacles or objects in the vehicle's environment by ultrasonic systems has been constantly increasing in recent years. Whereas in the past it was enough to know that an obstacle was present at a certain distance from the vehicle, today the aim is to use the echo signal profile to reconstruct the type of obstacle present in the vehicle's environment.

[0004] However, this increases the amount of data transmitted between the ultrasound system and the vehicle's data processing device, and the data buses typically used in vehicles have limited maximum data transfer rates, particularly for economic reasons.

[0005] Therefore, measures have already been taken in the past to reduce the amount of data transmitted. This may be achieved, for example, using obstacle object recognition (with corresponding confidence information about the reliability of the obstacle recognition), as described in DE 10 2008 042 640. Other examples of ultrasonic system implementations that can reduce data transmission can be found in DE 10 2005 024 716 A1 and DE 10 2012 207 164 A1.

[0006] It is desired to transmit more data of the actual measurement signals of the ultrasonic sensors to a central computer system, where these data are processed together with data from other ultrasonic sensor systems and / or together with data from other types of sensor systems, for example radar systems, i.e. by sensor fusion, into a so-called surrounding or environment map. It is therefore desired to perform object recognition only in the computer system using this sensor fusion, and not in the ultrasonic sensor systems themselves, to avoid data loss and thus reduce the possibility of erroneous information and the resulting wrong decisions, and thus reduce the risk of accidents. At the same time, however, the transmission bandwidth of the available sensor data bus is limited. Since they have proven their value in the field, a replacement of the same should be avoided. At the same time, it is therefore desired to avoid an increase in the amount of data to be transmitted. In short, the information content of the data and its relevance for the subsequent obstacle recognition (i.e. object recognition) performed in those computer systems must be increased without increasing the data rate too much, or preferably without increasing the data rate at all. Instead, the data rate requirements must be further lowered to allow data rate capacity for transmitting status data and self-test information of the ultrasonic sensor system to the computer system, which is essential within the framework of functional safety. The present invention addresses this problem.

[0007] Various methods for processing ultrasonic sensor signals are already known in the prior art.

[0008] A method for evaluating an echo signal for vehicle environment detection is known, for example, from WO-A-2012 / 016834. This document proposes emitting a measurement signal with a defined encoding and format, searching the received signal for components of the measurement signal in the received signal by correlation with the measurement signal, and determining these components. The correlation level (and not the level of the envelope of the echo signal) is then evaluated using a threshold value.

[0009] From DE-A-4 433 957 it is known to periodically emit ultrasonic pulses for object recognition and to determine the position of obstacles based on their propagation times, in which, during the evaluation, echoes which remain correlated in time over several measurement cycles are amplified, while echoes which remain uncorrelated are suppressed.

[0010] From DE-A-10 2012 015 967 A method is known for decoding a received signal received by an ultrasonic sensor of a vehicle. In this document, the transmission signal of the ultrasonic sensor is encoded and emitted, and for decoding, the received signal is correlated with a reference signal. Prior to the correlation of the received signal with the reference signal, a frequency shift of the received signal relative to the transmission signal is determined, and the received signal is correlated with the transmission signal as a reference signal shifted in frequency by the determined frequency shift. In order to determine the frequency shift of the received signal, it is Fourier transformed, and the frequency shift is determined on the basis of the result of the Fourier transform.

[0011] A method for detecting the environment of a vehicle by ultrasound is known from DE-A-10 2011 085 286, in which ultrasonic pulses are emitted and ultrasonic echoes reflected from objects are detected, the detection range being divided into at least two distance ranges, where the ultrasonic pulses used for detection in the respective distance ranges are emitted independently of one another and are encoded with different frequencies.

[0012] An apparatus and a method for environment detection by means of a signal converter and an evaluation unit are known from WO-A-2014 / 108300, in which a signal received from an environment having a first pulse response length is filtered according to its propagation time at a first time during a measurement cycle and a signal received from an environment having a second, longer pulse response length is filtered according to its propagation time at a second, later time within the same measurement cycle.

[0013] However, the technical teachings of the above patents are all guided by the idea of ​​performing the recognition of the obstacle (object) detected by the ultrasonic sensor already within the ultrasonic sensor itself, and transmitting the object data only after the recognition of the object, but in this way the synergistic effect is lost when multiple ultrasonic sensors are used.

[0014] Various methods for (partial pre-) processing of signals from ultrasonic sensors with subsequent transmission of these signals are known from DE-A-197 07 651 A1, DE-A-10 2010 044 993 A1, DE-A-10 2012 200 017 A1, DE-A-10 2013 015 402 A1 and EP-A-2 455 779 A1. [Prior art documents] [Patent documents]

[0015] [Patent Document 1] DE 10 2008 042 640 A1 [Patent Document 2] DE 10 2005 024 716 A1 [Patent Document 3] DE 10 2012 207 164 A1 [Patent Document 4] International Publication No. 2012 / 016834 [Patent Document 5] German Patent No. 4 433 957 [Patent Document 6] DE 10 2012 015 967 A1 [Patent Document 7] DE 10 2011 085 286 A1 [Patent Document 8] International Publication No. 2014 / 108300 [Patent Document 9] German Patent No. 197 07 651 [Patent Document 10] DE 10 2010 044 993 A1 [Patent Document 11] DE 10 2012 200 017 A1 [Patent Document 12] DE 10 2013 015 402 A1 [Patent Document 13] European Patent No. 2 455 779 Summary of the Invention [Problem to be solved by the invention]

[0016] The object of the present invention is to further increase the degree of data compression in ultrasonic sensors of a vehicle without compromising the reliability of the recognition of obstacles and obstacle types. A further object is to further reduce the bus bandwidth requirements for the transmission of measurement data from the ultrasonic sensor system to a computer system or to increase the efficiency of data transmission. [Means for solving the problem]

[0017] To this end, the present invention provides a method for transmitting data from an ultrasonic system having at least one ultrasonic transmitter and one ultrasonic receiver to a data processing device via a vehicle data bus, the method comprising: A plurality of predetermined signal profile characteristics are extracted from echo signals received by at least one ultrasound receiver of the ultrasound system; A signal profile object is identified within the echo signal based on a group of the extracted signal profile characteristics. Each identified signal profile object is assigned to one of a plurality of predefined signal profile object classes, each of which is identified by an identifier. For each identified signal profile object at least one object parameter describing it is determined, wherein said at least one object parameter is the time of occurrence of said signal profile object relative to a reference time, the temporal extent of said signal profile object, the time lag for the signal profile object before or after said echo signal, and / or the magnitude of said signal profile object, the peak, in particular the maximum peak, the time of the peak of said signal profile object within the temporal extent, in particular the time of the maximum peak, and / or the size of the area of ​​the echo signal section belonging to said signal profile object, in particular the size of the percentage of the area of ​​the echo signal section belonging to said signal profile object that exceeds a threshold or a threshold signal profile, wherein the echo signal section belonging to said signal profile object can be reconstructed from the identifier of the signal profile object and said at least one object parameter determined for the signal profile object. For an identified signal profile object, the ultrasonic system transmits, via a vehicle data bus, the identifier and the at least one object parameter as echo signal section data representative of echo signal sections of the signal profile object for detecting an obstacle and / or a distance of the obstacle relative to at least one ultrasonic receiver or one of a plurality of ultrasonic receivers of the ultrasonic system.

[0018] The basic idea of ​​the invention comprises the detection of potentially relevant structures in a measurement signal and the compression of this measurement signal by transmitting only a small number of data related to the detected potentially relevant structures instead of transmitting the measurement signal itself. The actual detection of objects, e.g. an obstacle to a parking maneuver, probably occurs after reconstruction of the measurement signal as a reconstructed measurement signal in a computer system, where typically the compressed measurement signals of the ultrasonic sensor systems arrive (and possibly are decompressed). The invention therefore particularly relates to the compression of data by detection of structures in a measurement signal.

[0019] Thus, according to the proposal according to the invention, the echo signals are checked for the presence of certain predetermined signal profile characteristics and transmit data representative of these signal profile characteristics, where further conclusions can be drawn from these echo signal data in the data processing device of the vehicle. For example, if the echo signals are reconstructed or similar signal profile characteristics, possibly offset in time, are repeatedly detected, a conclusion can be made about an obstacle, the type of obstacle, or a change in the distance of the vehicle from the obstacle, etc. In this respect, it is important that the task of detecting obstacles is transferred from the ultrasonic system to the data processing device, which reduces the demands on the "intelligence" of the elements of the ultrasonic system, with a possible reduction in the amount of data transmitted from the ultrasonic system to the data processing device, since the actual analysis of the echo signal data regarding what kind of obstacles are present in the vehicle's environment and how this obstacle changes in the vehicle's environment (especially with respect to the distance from the vehicle) is performed in the data processing device.

[0020] According to the invention, it is provided to check the received echo signal for the presence of certain predetermined signal profile characteristics. One or more of these signal profile characteristics define a special signal profile, hereinafter called signal profile object. There are several object classes, and the identified signal profile object is assigned to one of these object classes. Each object class is provided with an identifier. Furthermore, according to the invention, at least one object parameter is determined, which further describes or characterizes the identified signal profile object. Possible object parameters include, for example: The time of occurrence of a signal profile object relative to a reference time, The time range of the signal profile object, A time lag for the signal profile object before or after the echo signal, and / or The magnitude of the signal profile object, its peaks, especially the maximum peak the time of the peak, in particular the time of the largest peak of the signal profile object within that time range, and / or The size of the area (integral) of the echo signal section belonging to the signal profile object, and In particular, the size of the percentage of the area of ​​the echo signal section that belongs to a signal profile object that exceeds a threshold or a threshold signal profile.

[0021] From the identifier of the signal profile object and said at least one object parameter determined for the signal profile object, the echo signal section belonging to the signal profile object can be reconstructed in the data processing device. In this way, the echo signal section underlying the identified signal profile object has to be transmitted to the data processing device via its (digital) sample values, but can now be transmitted in a compressed state with a significantly lesser use of data transmitted to the data processing device. According to the invention, at least the identifier of the signal profile object class and at least one object parameter describing the signal profile object, so-called echo signal section data, are transmitted. If necessary, further data can be transmitted, which will be addressed later.

[0022] In a suitable embodiment of the present invention, the ultrasonic system comprises a plurality of ultrasonic transmitters and a plurality of ultrasonic receivers, and the echo signal section data represents a plurality of signal profile objects respectively identified from a plurality of echo signals received in a predetermined time slot, and the echo signal section data is transmitted via the vehicle data bus to the data processing device for detecting an obstacle and / or the distance of an obstacle to at least one ultrasonic receiver or one of the plurality of ultrasonic receivers of the ultrasonic system.

[0023] As already mentioned above, in addition to the identifiers of the plurality of signal profile objects and the at least one object parameter, further data can be transmitted as echo signal section data. Advantageously, it can be provided that in addition to the echo signal section data, a plurality of confidence values ​​respectively assigned to the plurality of identified signal profile objects are transmitted from the ultrasonic measuring device to the data processing device via the vehicle data bus.

[0024] According to the invention, the signal profile characteristics searched for in the echo signal are suitably local extrema of the echo signal with an occurrence time, which exceed a threshold or threshold signal, said signal profile characteristics are absolute extrema of the echo signal with an occurrence time, which exceed a threshold or threshold signal, said signal profile characteristics are saddle points of the echo signal with an occurrence time, which exceed a threshold or threshold signal, said signal profile characteristics are exceedances of one or a particular threshold or one or a particular threshold signal with an exceedance time, which exceedance occurs as the signal level of the echo signal increases, and / or said signal profile characteristics are undershoots of one or a particular threshold or one or a particular threshold signal with an undershoot time, which undershoot occurs as the signal level of the echo signal increases, or said signal profile characteristics are one or a plurality of predefined combinations of the above mentioned signal profile characteristics occurring successively in time.

[0025] Preferably, the signal profile characteristics or object parameters further include whether the received echo signals are modulated, and if so, how they are modulated, e.g. with monotonically increasing or strictly monotonically increasing frequency (chirp up), e.g. with monotonically decreasing or strictly monotonically decreasing frequency (chirp down), or e.g. with a constant frequency (no chirp). In this context, reference is made to DE-B-10 2017 123 049, DE-B-10 2017 123 051, DE-B-10 2017 123 052 and DE-B-10 2017 123 050, the disclosures of which are incorporated by reference into the subject matter of the present invention.

[0026] In general, other frequency modulation methods or other modulation methods can also be implemented. For example, modulation of multiple ultrasound signals with various types of encoding is useful. Advantageously, Doppler-resistant encoding can be used. Generally speaking, it is a property of the encoding that it can be seen as a predefined wavelet whose time-averaged value can in particular be different from zero.

[0027] In another advantageous embodiment of the invention, it is provided that a plurality of ultrasonic transmitters of the ultrasonic system emit different modulated ultrasonic signals, and that the echo signal section data transmitted by the plurality of ultrasonic receivers also includes an identifier of the modulation of the received echo signal, respectively. It is also provided that, based on the identifier of the modulation, it can be determined in the data processing device from which ultrasonic transmitter the ultrasonic transmission signal was emitted. The ultrasonic transmission signal is received by the ultrasonic receiver as an echo signal or an echo signal component, and the ultrasonic receiver transmits the echo signal section data relating to the echo signal or the echo signal component to the data processing device.

[0028] The advantages of the method according to the invention are particularly evident when the ultrasonic system comprises a plurality of ultrasonic transmitters and a plurality of ultrasonic receivers, whereby echo signal data is transmitted via a vehicle data bus to a data processing device, which data represents a plurality of signal profile characteristics respectively extracted from a plurality of echo signals received in a given time slot. The data processing device here receives a plurality of echo signals or a plurality of signal profile characteristics detected in these echo signals, which data describes these echo signals and are received by a plurality of ultrasonic receivers in a given time slot. For example, if a plurality of ultrasonic receivers arranged adjacent to each other receive similar echo signals, this can be used to classify the obstacle. It has been found that determining the type of obstacle based on a plurality of compressed echo signals from a plurality of ultrasonic receivers is much more efficient than first examining each echo signal from each ultrasonic receiver individually, determining the type of obstacle and finally synchronizing the obtained results on the type of obstacle with each other.

[0029] In the method of the present invention, a feature vector of the echo signal is generated, which contains, so to speak, signal profile characteristics with associated times within the echo signal profile. The feature vector thus describes individual sections of the echo signal and events of the echo signal, while detection of obstacles etc. is not yet performed.

[0030] According to the invention, an envelope signal can also be generated from the echo signal, which envelope signal can be part of the feature vector or parts of which can be part of the feature vector. The echo signal can be collapsed with the associated ultrasound transmission signal, i.e. with the ultrasound signal received as the echo signal after reflection, thereby forming a correlation signal, features of which can be part of the feature vector.

[0031] The echo signal data describing a number of signal profile characteristics may advantageously include parameter data. The parameter data element may preferably be a timestamp indicating when a particular or one characteristic occurred in the echo signal profile. The time reference (i.e. the reference time) of the timestamp is arbitrary but is predefined for the system formed by the ultrasound system and the data processing device. Another parameter may be the amplitude and / or the extension of a section of the echo signal describing the signal profile characteristic, etc. At this point, it should be noted that in the following the term "amplitude" should be understood in a general way and is used herein as e.g. the (current) signal level of the signal and / or the peak value of the signal.

[0032] The compression of data transmitted via the data bus between the ultrasound system and the data processing device, which is achieved according to the invention, on the one hand reduces the data bus load and thus its importance with regard to EMC requirements, and on the other hand provides free data bus capacity during the reception time of multiple echo signals, which can be used for the transmission of control commands from the data processing device as well as for the transmission of status information and other data of the ultrasound system to the data processing device. The advantageously implemented prioritization of the transmitted data ensures that safety-relevant data is transmitted first, so that unnecessary dead times of echo signal data are not caused.

[0033] According to an advantageous embodiment of the present invention, an ultrasonic system comprises a plurality of ultrasonic transmitters and a plurality of ultrasonic receivers, and echo signal data representing a plurality of signal profile characteristics respectively extracted from a plurality of echo signals received in a predetermined time slot can be transmitted to the data processing device via the vehicle data bus.

[0034] According to an advantageous embodiment of the invention, in addition to the echo signal data, also confidence values ​​assigned to each of the extracted signal profile characteristics can be transmitted from the ultrasonic measuring device to the data processing device via the vehicle data bus.

[0035] According to further advantageous embodiments of the invention, the signal profile characteristics are local extrema of the echo signal with an occurrence time, which exceed a threshold or threshold signal, the signal profile characteristics are absolute extrema of the echo signal with an occurrence time, which exceed a threshold or threshold signal, the signal profile characteristics are saddle points of the echo signal with an occurrence time, which exceed a threshold or threshold signal, the signal profile characteristics are exceeding of a threshold or a specific threshold signal with an exceeding time, which exceeding occurs as the signal level of the echo signal increases, and / or the signal profile characteristics are undershoot of a threshold or a specific threshold signal with an undershoot time, which undershoot occurs as the signal level of the echo signal. However, the signal profile characteristics may also be a time series (in a predetermined order) of said signal profile characteristics.

[0036] Furthermore, in accordance with an advantageous embodiment of the present invention, the plurality of signal profile characteristics may further include whether to modulate the received echo signal, and if so, how to modulate it, e.g., with a monotonically increasing or strictly monotonically increasing frequency (chirp up), e.g., with a monotonically decreasing or strictly monotonically decreasing frequency (chirp down), or e.g., with a constant frequency (no chirp).

[0037] According to a further advantageous embodiment of the invention, the ultrasonic transmitters of the ultrasonic system emit different modulated ultrasonic signals, and based on the modulation of these echo signals it can be determined which ultrasonic transmitter emitted the ultrasonic transmission signal received as an echo signal or echo signal component by the ultrasonic receiver. The invention therefore proposes a method for transmitting sensor data from a sensor to a computer system. The method is particularly useful when implemented for the transmission of data of ultrasonic reception signals from an ultrasonic receiver (hereinafter ultrasonic sensor) to a control device (as computer system or data processing device) in the vehicle. According to a variant of the invention, first an ultrasonic burst is generated and typically emitted in free space in the environment of the vehicle. Here, the ultrasonic burst is composed by several sound wave pulses that follow each other at ultrasonic frequencies. This ultrasonic burst is generated by a mechanical oscillator that starts oscillating and then damps out again. The ultrasonic burst thus emitted is then reflected by several objects (for example several obstacles) and is received by the receiver as an ultrasonic signal and converted into a received signal. It is particularly preferred that the ultrasonic transmitter is identical to the ultrasonic receiver, in which case it is called a transducer that operates alternately as ultrasonic transmitter and ultrasonic receiver. However, the principles described below apply to separate multiple receivers and transmitters.

[0038] The ultrasonic sensor has a signal processing unit that analyzes the received signal for a number of signal profile characteristics that may or have been predetermined in order to minimize the amount of data that needs to be transmitted for the description of the echo signal profile. The signal processing unit of the ultrasonic sensor performs data compression of the received signal as is to generate compressed data, i.e. characteristic echo signal data. The information is then transmitted in a compressed form to the computer system. This minimizes the EMC load caused by the data transmission and allows the status data of the ultrasonic sensor to be transmitted to the computer system via a data bus between the ultrasonic sensor and the computer system for the detection of system errors in time gaps.

[0039] It has proven advantageous to prioritize the transmission of data over the data bus. Here, reports of safety-critical malfunctions of the sensors, in this example the ultrasonic sensors, to the computer have the highest priority, since they impair the validity of the measurement data of the ultrasonic sensors. These data are supplied to the computer system by the ultrasonic system. The second highest priority belongs to requests from the computer system to perform safety-related self-tests. Such commands are sent from the computer system to the ultrasonic system. The data of the ultrasonic sensors themselves have the third highest priority, since latency must not be extended. All other data have a lower priority for transmission over the data bus.

[0040] It is particularly advantageous that the method for transmitting sensor data, in particular data of an ultrasonic sensor, from a sensor to a computer system, in particular a vehicle, comprising emission of an ultrasonic burst with a start 57 and an end 56 of the emission of the ultrasonic burst, receiving an ultrasonic signal and forming a received signal at least during a reception time TE from the end 56 of the emission of the ultrasonic burst, and transmitting the compressed data via a data bus, in particular a single-wire data bus, to the computer system, is configured such that the transmission 54 of data from the sensor to the computer system starts with a start command 53 via the data bus from the computer system to the sensor before the end 56 of the emission of the ultrasonic burst, or starts after a start command 53 via the data bus from the computer system to the sensor and before the start 57 of the emission of the ultrasonic burst. After the start command 53, the transmission 54 is performed in a continuous manner at regular intervals until the end of the data transmission 58. This end of the data transmission 58 occurs chronologically after the end of the reception time TE.

[0041] Another variant of the proposed method provides for the formation of a feature vector signal from the received signal as a first step of data compression. Such a feature vector signal may contain multiple analog and digital data signals. It is therefore a more or less complex data / signal structure. In the simplest case, it can be understood as a vector signal composed by multiple partial signals.

[0042] For example, it may be feasible to form first and / or higher time derivatives of the received signal, or single or multiple integrals of the received signal, which are then multiple sub-signals of the feature vector signal.

[0043] It is then also possible to form an envelope signal which is a partial signal of the feature vector signal.

[0044] Furthermore, it may be feasible to fold the received signal with the emitted ultrasound signal, thus forming a correlation signal that may be a partial signal in the feature vector signal, where on the one hand a signal may be used as the emitted ultrasound signal, which has been used to control the driver of the transmitter, or on the other hand a signal measured at the transmitter, for example, may be used, whereby this signal corresponds better to the actually emitted sound waves.

[0045] Finally, it may be feasible to detect the occurrence of predetermined signal profile characteristics using a matched filter and form a matched filter signal for each of several predetermined signal profile characteristics. A matched filter is a filter that optimizes the signal-to-noise ratio (SNR). Predefined signal profile characteristics are to be detected in a noisy signal. In the literature, the terms "correlation filter", "signal matched filter (SMF)", or simply "matched filter" are often found. A matched filter serves to optimally determine (detect) the presence of a known signal waveform amplitude and / or position, predetermined signal profile characteristics, even in the presence of interference (parameter estimation), for example by signals from other ultrasonic transmitters and / or in case of ground echoes (clutter).

[0046] In that case, the plurality of matched filter signals are preferably a plurality of partial signals within the feature vector signal.

[0047] The specific events may be indicated by individual partial signals of the feature vector signal. These events are also signal profile characteristics in the sense of the present invention. Thus, the signal profile characteristics include not only specific signal waveforms, such as rectangular pulses, wavelets or wave trains, but also characteristic points in the profile of the received signal and / or signal profiles derived therefrom, such as an envelope signal, which may be obtained from the received signal by filtering.

[0048] The other signal, which may be a sub-signal of the feature vector signal, may for example be the envelope of the received signal, ie whether the envelope signal crosses a first predefined threshold value.

[0049] The other signal, which may be a sub-signal of the feature vector signal, is detected, for example, by detecting whether the envelope of the received signal, ie the envelope signal, crosses upwardly a predefined second threshold, which may be identical to the first threshold.

[0050] The other signal, which may be a sub-signal of the feature vector signal, is detected, for example, by detecting whether the envelope of the received signal, ie the envelope signal, crosses downwardly a predefined third threshold, which may be identical to the first threshold.

[0051] Another signal, which may be a partial signal of the feature vector signal, is detected, for example, by detecting whether the envelope of the received signal, ie the envelope signal, has a maximum that exceeds a fourth threshold, which may be identical to the above threshold.

[0052] Another signal, which may be a partial signal of the feature vector signal, is for example detected whether the envelope of the received signal, i.e. the envelope signal, has a minimum above a fifth threshold, which may be identical to the threshold. Here, in order to avoid the detection of noise, it is preferably evaluated whether at least one preceding maximum of the envelope has a minimum distance to the minimum. In this respect, other filterings are conceivable. It may also be checked whether the time distance between this minimum and the preceding maximum is greater than a first minimum time distance. If these conditions are respectively fulfilled, a flag or a signal, which itself is preferably a partial signal of the feature vector signal, is respectively set.

[0053] Similarly, the time and amplitude related distances of other signal profile features must be checked in a similar manner to see if they meet certain validity requirements, such as a minimum time distance and / or a minimum amplitude distance, etc. From these checks, further analog, binary or digital partial signals can be derived, which thereby further increase the dimensionality of the feature vector signal.

[0054] If necessary, the feature vector signals can be converted to significant feature vector signals in a significance augmentation stage, but practice shows that for current applications this is not yet necessary.

[0055] According to a variant of the method of the invention, detection and classification of a plurality of signal profile characteristics in the received signal into a plurality of recognized signal profile characteristic classes is then performed on the basis of the feature vector signal or the significant feature vector signal.

[0056] For example, if the amplitude of the output signal of the matched filter, and thus the amplitude of the partial signal of the feature vector signal, possibly exceeds a sixth threshold specific to the matched filter, the signal profile characteristic for which the matched filter is configured may be considered to be recognized. Here, other parameters are preferably taken into account as well. For example, if an ultrasonic burst with an increasing frequency during the burst (called chirp-up) is transmitted, an echo with this modulation function is predicted. If the signal waveform of the envelope, for example the triangular signal waveform of the envelope, coincides locally in time with the expected signal waveform, but the modulation function does not, this is not an echo from the transmitter, but a noise signal that may result from other ultrasonic transmitters or overreach. In that respect, the system can distinguish between multiple inherent echoes and multiple external echoes, whereby the same signal waveform is associated with two different signal profile characteristics, namely multiple inherent echoes and multiple external echoes. Here, the transmission of multiple inherent echoes is preferably prioritized over the transmission of multiple external echoes, since the former is generally safety-related and the latter is generally not safety-related.

[0057] Typically, at least one signal profile characteristic parameter is assigned to each recognized signal profile characteristic or determined for this signal profile characteristic. Preferably, this is a time stamp indicating when the characteristic occurs in the echo signal. The time stamp is also referred to as, for example, the time start of the signal profile characteristic of the received signal or the time end or time position of the time focus of this signal profile characteristic, etc. Other signal profile characteristic parameters such as amplitude, extension, etc. can be considered as well. In a variant of the proposed method, at least one of the assigned signal profile characteristic parameters is transmitted together with at least one recognized class of the signal profile characteristics, which is a time value indicating a time position in time from the emission of the preceding ultrasonic burst suitable for drawing a conclusion therefrom. Preferably, this is later used to determine the distance of an object (e.g. an obstacle) in the vehicle's environment depending on the transmitted time value, thus determined and transmitted.

[0058] Finally, a prioritized transmission of the recognized signal profile characteristic classes is preferably performed together with the assigned signal profile characteristic parameters, respectively. The transmission may also be performed in more complex data structures. For example, it is conceivable to first transmit the time of the recognized safety-related signal profile characteristics (e.g. identified obstacles) and then transmit the recognized signal profile characteristic classes of the safety-related signal objects. In this way, the latency is further reduced.

[0059] According to a variant of the method of the invention, the chirp value is determined as an assigned signal profile characteristic parameter, which signal profile characteristic parameter indicates whether the recognized signal profile characteristic is an echo of an ultrasonic transmit burst having multiple chirp up functions, multiple chirp down functions, or multiple no-chirp functions. "Chirp up" means that the frequency goes up within the received signal profile characteristic of the received signal. "Chirp down" means that the frequency goes down within the received signal profile characteristic of the received signal. "No-chirp" means that the frequency remains substantially the same within the received signal profile characteristic of the received signal.

[0060] According to a variant of the method of the invention, it is possible to generate a confidence signal (confidence value) for example by forming a correlation between the received signal or a signal derived therefrom and a reference signal, for example an ultrasonic transmission signal or other expected wavelet, the confidence signal then usually being a partial signal of the feature vector signal.

[0061] In another variant of the method, a phase signal is also formed on this basis, which signal indicates a phase shift of a reference signal, e.g. an ultrasonic transmission signal and / or another reference signal, for example the received signal or a signal formed therefrom (e.g. a reliable signal).

[0062] Similarly, according to a further variant of the proposed method, by forming a correlation between the phase signal or a signal derived therefrom on the one hand and a reference signal on the other hand, it is possible to form a phase confidence signal and use it like a partial signal of the feature vector signal.

[0063] It is then feasible to perform a comparison of the phase confidence signal with one or more thresholds in the evaluation of the feature vector signal to generate a discrete phase confidence signal which itself can again become part of the feature vector signal.

[0064] In a variant of the proposed method, the evaluation of the feature vector signal and / or the significant feature vector signal may be performed by forming one or more distance values ​​between the feature vector signal and one or more signal profile characteristic prototype values ​​for a plurality of detectable signal profile characteristic classes. Such distance values ​​may be Boolean, binary, discrete, digital or analog values. Preferably, all distance values ​​are linked together with a non-linear function. Thus, if a chirp-up echo of triangular form is expected, a received chirp-down echo of triangular form may be rejected. This rejection is a non-linear process.

[0065] Conversely, the triangle can be defined differently in the received signal. This first considers the amplitude of the triangle in the received signal. If the amplitude of the received signal is sufficient, the matched filter associated with this triangular signal provides a signal that exceeds, for example, a predefined seventh threshold. In this case, this signal profile characteristic class (triangular signal) can be assigned to this exceedance as a recognized signal profile characteristic. In this case, the distance value between the feature vector signal and the prototype (here the seventh threshold) undershoots one or more predefined binary, digital, or analog distance values ​​(here zero crossings).

[0066] In other variations of the method, at least one signal profile characteristic class is associated with and detected by wavelets estimated by multiple estimators (e.g. multiple matched filters) and / or multiple estimation methods (e.g. multiple estimation programs executed in a digital signal processor). The term wavelet refers to a function that can be used as the basis for a continuous or discrete wavelet transform. The word is a new creation formed from the French "ondelet" meaning "little wave" and is translated into English partly literally ("onde" to "wave") and partly phonetically ("-lette" to "let"). The term "wavelet" was coined in the 1980s in geophysics (Jean Maury, Alex Grossman) as a function generalizing the short-time Fourier transform, but since the late 1980s it has now become common knowledge. In the 1990s, a real wavelet boom was created, triggered by the discovery of compact and stable (arbitrarily differentiable) orthogonal wavelets by Ingrid Daubechy (1988) and the development of the fast wavelet transform (FWT) using multiscale analysis (multiresolution analysis - MRA) by Stéphane Marat and Yves Meillet (1989).

[0067] Unlike the sine and cosine functions of the Fourier transform, the most frequently used wavelets have locality not only in the frequency spectrum but also in the time domain. Here, "locality" should be understood in the sense of slight differences. The probability density is the normalized square of the absolute value of the function of interest or its Fourier transform. Here, the product of both variances is always greater than a constant, similar to the Heisenberg uncertainty principle. Due to this restriction, the Paley-Wiener theorem (Raymond Paley, Norbert Wiener) was developed in the field of function calculus, which is the precursor of the discrete wavelet transform, and the Calderon-Zygmunt theorem (Alberto Calderon, Antoni Zygmunt) corresponds to the continuous wavelet transform.

[0068] The integral of a wavelet function is always zero, so that it usually takes the form of an outwardly petering (getting smaller) wave (a "small wave" is an ondelet, a wavelet). However, wavelets with integrals different from zero are also allowed in the sense of the present invention. The rectangular and triangular wavelets described below may be mentioned as examples in this context.

[0069] Important examples of wavelets with zero integral are the Haar wavelet (Alfred Haar 1909), the Daubechy wavelet named after Ingrid Daubechy (c. 1990), the coiffure wavelet constructed by her, and the more theoretically important Meillet wavelet (Yves Meillet, c. 1988).

[0070] Wavelets exist in spaces of any dimension, with 1D wavelet-based tensor products being most frequently used. Due to the fractal nature of the two-scale equations of MRA, most wavelets have complex shapes, and most wavelets do not have closed forms. This is particularly important because the feature vector signals mentioned above are multidimensional, and therefore multidimensional wavelets can be used for signal object detection.

[0071] Therefore, a particular variant of the proposed method is the use of multidimensional wavelets with more than two dimensions for the detection of signal objects. In particular, it is proposed to use matched filters corresponding to the detection of such wavelets with more than two dimensions in order to possibly complement the feature vector signal with a number of further partial signals suitable for detection.

[0072] A wavelet that is particularly useful for the analysis of envelope signals is, for example, the triangular wavelet, which is characterized by a start time of the triangular wavelet, a linear increase in the wavelet amplitude in time up to the maximum amplitude of the triangular wavelet, an increase occurring in time after the start time of the triangular wavelet, and a linear decrease in the wavelet amplitude in time up to the end of the triangular wavelet, an event occurring in time after the maximum of the triangular wavelet.

[0073] Another particularly suitable wavelet is the rectangular wavelet, which also encompasses the trapezoidal wavelet in the sense of the present invention. The rectangular wavelet is characterized by an increase in the wavelet amplitude of the rectangular wavelet with a first time gradient of the rectangular wavelet until the start time of the rectangular wavelet, followed by a first plateau time of the rectangular wavelet. Following the first plateau time of the rectangular wavelet, there is a wavelet amplitude with a second gradient of the wavelet amplitude until the second plateau time of the rectangular wavelet. The second plateau time of the rectangular wavelet is followed by a decrease with a third time gradient until the time end of the rectangular wavelet. Here, the slope of the second time gradient is less than 10% of the slope of the first time gradient and less than 10% of the slope of the third time gradient.

[0074] Instead of the wavelets mentioned above, other two-dimensional wavelets may also be used, such as for example half-sine wavelets that also have integrals different from zero.

[0075] When using wavelets, it is proposed that the time shift of the respective wavelet of the recognized signal profile characteristic is determined as a signal profile characteristic parameter, for example by correlation and / or exceedance time of the output level of a matched filter suitable for detecting the respective wavelet, via a predefined threshold for this signal profile characteristic or for this wavelet, respectively. Preferably, the envelope of the received signal, and / or the phase signal, and / or the confidence signal, etc. are evaluated.

[0076] Another possible signal profile characteristic parameter that can be determined is the temporal compression or expansion of the respective wavelets of the recognized signal profile characteristic. Similarly, the amplitude of the wavelets of the recognized signal profile characteristic can be determined.

[0077] During the development of the proposal of the disclosed method, it was found to be advantageous that data of the recognized signal profile characteristics of very fast arriving echoes are first transmitted from the sensor to the computer system, and data of the recognized signal profile characteristics are then transmitted. Preferably, at least the recognized signal profile characteristic class and a time stamp are transmitted each time, this time stamp indicating when the signal profile characteristic arrives at the sensor. In the framework of the detection process, a score can be assigned to the different signal profile characteristics considered as sections of the received signal. This score indicates the probability due to the presence of this signal profile characteristic according to the estimation algorithm used. In the simplest case, such a score is binary. However, preferably, it is complex, real or integer. In case of the signal profile characteristics having high score values, it may also be feasible to transmit data of the recognized signal profile characteristics with lower scores. To enable correct processing by the computer system, not only the date of the recognized signal profile characteristics and the time stamp of the respective signal profile characteristic, but also the determined score value must be transmitted in this case. Thus, in addition to the recognized signal profile characteristics, a list of hypotheses with the assigned score values ​​is transmitted in this case to the computer system.

[0078] Preferably, the recognized signal profile characteristic classes and the assigned data, e.g. timestamps and scores of each of the recognized signal profile characteristic classes, i.e. the signal profile characteristic parameters, are transmitted according to the FIFO principle, which ensures that the data of the nearest object reflections are always transmitted first, thus ensuring that safety-critical cases of vehicle collisions with obstacles are prioritized and processed according to probability.

[0079] In addition to the transmission of the measurement data, a transmission of an error state of the sensor may also be performed. This may take place during the reception time TE if the sensor determines via the self-test device that a defect exists and that the previously transmitted data is potentially defective. It is thus ensured that the computer system obtains information about changes in the evaluation of the measurement data at the earliest possible time and can reject them or process them in some other way. This is particularly important for emergency braking systems. The reason is that emergency braking is a safety-critical operation and must only be initiated if the underlying data has a corresponding confidence value. In comparison therewith, the transmission of the measurement data, for example the date of the recognized signal profile characteristic class and / or the single assigned signal profile characteristic parameter, is therefore postponed and thus has a lower priority. Of course, an interruption of the transmission is also conceivable if an error occurs in the sensor. However, in some cases an error may appear possible, but its presence may not be confirmed. In such cases, the continuation of the transmission is therefore recommended. The transmission of safety-critical errors of the sensor is therefore performed with a higher priority.

[0080] In addition to the above-mentioned two wavelets with integrals of zero and signal sections with integrals other than zero, which are additionally called wavelets here, certain times of the profile of the received signal are considered as signal profile characteristics in the sense of the present invention. This invention can be used for data compression and can be transmitted instead of sample values ​​of the received signal. Subsets of this set of possible signal profile characteristics are referred to below as signal profile points. Signal profile points are therefore a special form of signal profile characteristics in the sense of the present invention.

[0081] The first possible signal profile point, and therefore signal profile characteristic, is the crossing of the amplitude of the threshold signal SW with the amplitude of the envelope signal 1 in the ascending direction.

[0082] A second possible signal profile point, and therefore signal profile characteristic, is the crossing of the amplitude of the threshold signal SW with the amplitude of the envelope signal 1 in the downward direction.

[0083] A third possible signal profile point, and therefore signal profile characteristic, is the maximum amplitude of the envelope signal 1 above the amplitude of the threshold signal SW.

[0084] A fourth possible signal profile point, and therefore signal profile characteristic, is the minimum amplitude of the envelope signal 1 above the amplitude of the threshold signal SW.

[0085] For these four exemplary types of signal profile points, as well as other types of signal profile points, it may conceivably be feasible to use multiple threshold signals specific to the type of signal profile point.

[0086] The time series of the signal profile characteristics is usually not arbitrary. For example, if a triangular wave is expected in the envelope signal 1 with sufficient amplitude, then in addition to the corresponding minimum level at the output of a matched filter suitable for detecting such a triangular wave, what is expected in time correlation with the exceeding of said minimum level at the output of said matched filter is: 1. The occurrence of a first possible signal profile point involving an upward crossing of the amplitude of the threshold signal SW with the amplitude of the envelope signal 1, followed by 2. The occurrence of a third possible signal profile point having a maximum value of the amplitude of the envelope signal 1 that exceeds the amplitude of the threshold signal SW, followed by 3. Generation of a second possible signal profile point having a downward crossing of the amplitude of the threshold signal SW and the amplitude of the envelope signal 1. It is.

[0087] Additionally, this exceeding of the minimum level at the output of the matched filter is another example of a fifth possible signal profile point and thus a further possible signal profile characteristic.

[0088] The resulting grouping and time series sequence of the recognized signal profile characteristics may itself be recognized, for example by a Viterbi decoder, as a predefined, expected grouping or time series sequence of signal profile characteristics and thus may itself be a signal profile characteristic. Thus, such a predefined grouping and / or time series sequence of other signal profile characteristics is a sixth possible signal profile point and thus a signal profile characteristic.

[0089] If such a grouping of a plurality of signal profile characteristics or a time-series sequence of a plurality of signal profile characteristic classes is recognized, then a transmission of this recognized summary signal profile characteristic class and at least one assigned signal profile characteristic parameter is performed, preferably instead of the transmission of the individual plurality of signal profiles, since in this way significant data bus capacity is saved. It is possible that both are transmitted. Here, the date of the signal profile characteristic class of the signal profile characteristic is transmitted, which is a predefined time-series sequence and / or a grouping of a plurality of other signal profile characteristics. In order to achieve compression, it is advantageous if at least one signal profile characteristic class in at least one of the other plurality of signal profile characteristics is not transmitted.

[0090] In particular, a grouping of the time series of multiple signal profile features is given if the time distance of these signal profile features does not exceed a predefined distance. In the above example, the propagation time of the signal in the matched filter should be taken into account. Usually, the matched filter is slower than the comparator. Therefore, the change in the output signal of the matched filter should have a fixed time correlation with the time occurrence of the associated multiple signal profile points.

[0091] It is particularly preferred to carry out the data transmission in the vehicle via a bidirectional single-wire data bus. Here, the return path is preferably ensured by the vehicle body. Preferably, the sensor data is transmitted to the computer system in a current-modulated state. The data for controlling the sensor is preferably transmitted by the computer system to the sensor in a voltage-modulated state. According to the invention, it has been found that the use of a PSI5 data bus and / or a DSI3 data bus is particularly suitable for the data transmission. Furthermore, it has been found to be particularly advantageous to carry out the transmission of data to the computer system with a transmission rate of more than 200 kBit / s and to carry out the transmission of data from the computer system to the at least one sensor with a transmission rate of more than 10 kBit / s, preferably more than 20 kBit / s. Furthermore, it has been found that the transmission of data from the sensor to the computer system should be modulated on a data bus whose current is less than 50 mA, preferably less than 5 mA, more preferably less than 2.5 mA. These buses should be adapted to these operating values ​​in a corresponding manner. The basic principle, however, remains the same.

[0092] To carry out the above-mentioned method, a computer system is needed having a data interface to the above-mentioned data bus, preferably the above-mentioned single-wire data bus, which supports the decompression of the data thus compressed. Generally, however, the computer system does not perform a full decompression, but only evaluates, for example, the time stamp and the recognized signal profile characteristic types. The sensor required to carry out one of the above-mentioned methods comprises at least one transmitter and at least one receiver for generating a received signal, which may be implemented in combination as one or more converters. Furthermore, it comprises at least a number of devices for processing and compressing the received signal, and a data interface for transmitting the data to the computer system via a data bus, preferably a single-wire data bus. In the case of compression, the compression device preferably comprises at least one of the following sub-devices, which are a number of matched filters, a number of comparators, a threshold signal generating device for generating one or more threshold signals SW, a number of differentiators for forming a number of derivatives, an integrator for forming a number of integral signals, a number of other filters, an envelope former for generating an envelope signal from the received signal, a correlation filter for comparing the received signal or a number of signals derived therefrom with a reference signal.

[0093] The invention will now be described in more detail with reference to several embodiments and to several drawings. In particular, the figures show: [Brief description of the drawings]

[0094] [Figure 1] FIG. 1 shows the basic process of signal compression and transmission. [Diagram 2] FIG. 2 shows in more detail the basic process of signal compression and transmission. [Diagram 3]FIG. 3a is a conventional ultrasound echo signal and its conventional evaluation, FIG. 3b is a conventional ultrasound echo signal with transmitted amplitude and its conventional evaluation, FIG. 3c is an ultrasound echo signal with chirp direction included, and FIG. 3d is multiple signal objects (multiple triangular signals) recognized in the signal of FIG. 3c with rejection of multiple signal parts that are not recognized. [Figure 4] FIG. 4a is a conventional transmission that is not claimed, FIG. 4b is a transmission of analyzed data after complete reception of an ultrasound echo that is not claimed, and FIG. 4c is a transmission of compressed data that is claimed by the present invention, where in this example multiple symbols for multiple basic signal objects that are substantially uncompressed are transmitted corresponding to the prior art. [Diagram 5] FIG. 5 illustrates a transmission of compressed data as claimed by the present invention, in which symbols of elementary signal objects are compressed to form symbols of signal objects. [Figure 6] FIG. 6 illustrates the transmission of compressed data as claimed by the present invention, in which symbols of a number of basic signal objects are compressed to form symbols of a number of signal objects, and in which not only the envelope signal but also the confidence signal is evaluated. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0095] As explained above, the technical teachings of the prior art are all guided by the idea of ​​performing the recognition of obstacles (objects) detected by the ultrasonic sensor already present in the ultrasonic sensor itself, and transmitting object data only after the recognition of the object.However, it has been found that within the framework of the present invention, it is not feasible to transmit not only the echo data of the ultrasonic sensor itself, but also all of the data, since the synergistic effect is lost when multiple ultrasonic sensors are used.

[0096] Furthermore, in the central computer system, data of preferably several sensors can be advantageously evaluated. However, for this purpose, the compression of the data for transmission via a data bus with low bus bandwidth must be performed in a different way than in the prior art. This allows synergistic effects to be exploited. It is conceivable, for example, that a vehicle has several ultrasonic sensors. It is feasible if these two sensors transmit with different encodings in order to distinguish between the two sensors. Unlike the prior art, here both sensors detect the ultrasonic echoes emitted by both of the two ultrasonic sensors and transmit these in a suitably compressed way to the central computer system, where the ultrasonic reception signals are reconstructed and combined. Recognition of obstacles (several objects in the environment) is only performed after reconstruction (decompression). In addition, this allows further combination of the data of the ultrasonic sensor with data of other multiple sensor systems, such as radar.

[0097] The invention provides a method for transmitting sensor data from a sensor to a computer system. The method is particularly suitable for transmitting data of ultrasonic reception signals from an ultrasonic sensor to a control device as a computer system in a vehicle. The method is described with reference to FIG. 1. According to the proposed method, first an ultrasonic burst is generated and typically emitted in free space in the environment of the vehicle (step α in FIG. 1). Here, the ultrasonic burst is composed by a number of sound pulses following each other at ultrasonic frequencies. This ultrasonic burst is generated by a mechanical oscillator that slowly starts oscillating and then damps the oscillation. Thus, the ultrasonic burst emitted by the exemplary ultrasonic transducer is then reflected by an object in the environment of the vehicle and is received as an ultrasonic signal by an ultrasonic receiver or by the ultrasonic transducer itself and converted into an electrical reception signal (step β in FIG. 1). It is particularly preferred that the ultrasonic transmitter is identical to the ultrasonic receiver, hereinafter referred to as the transducer, which is an electro-acoustic component that is alternately operated as an ultrasonic transmitter or an ultrasonic receiver and thus as an ultrasonic sensor. However, the principles described below are also applicable to individual receivers and transmitters. The proposed ultrasonic sensor comprises a signal processing unit that analyses and compresses the electrical reception signal (hereinafter referred to as the "received signal") corresponding to the ultrasonic reception signal (step γ in FIG. 1), thereby minimizing the required data transmission (amount of data transmitted) and providing free space for e.g. status messages and further control commands from the control computer to the signal processing unit or to the ultrasonic sensor system. The compressed electrical reception signal is then transmitted to the computer system (step δ in FIG. 1).

[0098] The related method therefore serves in particular for transmitting sensor data of an ultrasonic sensor from the sensor to a computer system, in particular in a vehicle. The transmission is preceded by the emission of an ultrasonic burst (step α in FIG. 1), the reception of the ultrasonic signal and the formation of an electrical reception signal (step β in FIG. 1). Subsequently, a data compression of the reception signal is performed (step γ in FIG. 1) by detecting preferably at least two or more predetermined reception signal characteristics in the reception signal for the generation of compressed data (step γ in FIG. 1). By sampling (step γa in FIG. 2, where step γa is divided into 5 substeps), the electrical reception signal is preferably converted into a sampled reception signal formed by a time-discrete stream of sample values. Typically, a sampling time of one can be assigned to each sample value as a timestamp of this sample value. The compression can be performed, for example, by a wavelet transform (step γb in FIG. 2). For this purpose, the received ultrasonic signal in the form of a sampled reception signal can be compared with a predefined basic signal form (further called a plurality of signal profile characteristics as mentioned above) stored in a library by forming a correlation integral (see also Wikipedia for this term) between the predefined basic signal form and the sampled reception signal. A time series sequence of base signal forms in the received signal forms signal objects each assigned to one of a number of signal object classes. By forming a correlation integral, a number of associated spectral values ​​for each of the signal object classes are determined for each of these signal object classes. Because this occurs continuously, the number of spectral values ​​themselves represent a stream of instantaneous spectral values ​​that are discrete in time, where a timestamp can be assigned to each spectral value. An alternative mathematically equivalent method is to use a matched filter for each given signal object class (base signal form).Since typically multiple signal object classes are used, and therefore may be subjected to different time spreads (see "Wavelet Analysis"), a time-discrete stream of multi-dimensional vectors of spectral values ​​of different signal object classes and their respective different time spreads is obtained in this way, again with a time stamp being assigned to each of these multi-dimensional vectors. Each of these multi-dimensional vectors is a so-called feature vector. The stream is thus a time-discrete stream of feature vectors. Again a time stamp is assigned to each of these feature vectors (step γb in FIG. 2).

[0099] Thus, by successive time shifts, the time dimension is also obtained. As a result, the feature vector of the spectral values ​​can be complemented by past values, values ​​dependent on them, such as time integrals or derivatives, or filter values ​​of one or more of these values. This can further increase the dimension of these feature vectors in the feature vector data stream. Thus, in order to keep the subsequent effort low, a restriction of some signal object classes is feasible during the extraction of the feature vectors from the sampled received signals of the ultrasonic sensor. Thus, for example, a matched filter can be used to continuously monitor the occurrence of these signal object classes in the received signal.

[0100] Particularly simple examples of signal object classes include the names isosceles triangle and double peak, among others.Signal object classes are generally formed by predefined spectral coefficient vectors, i.e. predefined feature vector values.

[0101] For the determination of the relevance of the spectral coefficients of the feature vector of the ultrasound echo signal, the determination of the distance values ​​of these characteristics, the elements of the vector of the instantaneous spectral coefficients (feature vector) are a combination of these characteristics (prototypes) in the form of a signal object class, this signal object class being symbolized by a predefined feature vector (prototype or prototype vector) from a library of at least predefined signal object class vectors (step γd in FIG. 2). Preferably, before the correlation, the spectral coefficients of the feature vector are normalized by the prototypes (step γc in FIG. 2). The distance determined by this distance determination can be formed, for example, by the sum of all the differences between the spectral coefficients of the predefined feature vector (prototype or prototype vector) of the respective prototype and the corresponding normalized spectral coefficients of the current feature vector of the ultrasound echo signal. The Euclidean distance is formed by the square root of the sum of the squares of all the differences between the spectral coefficients of the predefined feature vector (prototype or prototype vector) of the respective prototype and the corresponding normalized spectral coefficients of the current feature vector of the ultrasound echo signal, respectively. However, this distance formation is generally too complicated. Other methods of distance formation are conceivable. Then, each given feature vector (prototype or prototype vector) may have a symbol and possibly parameters, for example a distance value and / or amplitude, assigned to them before normalization. If the distance thus determined is below a first threshold and is the minimum distance from the current feature vector value to one of the given feature vector values ​​(prototypes or prototype vector values), the symbol continues to be used as the recognized prototype. Thus, a pair is formed consisting of the recognized prototype and the timestamp of the current feature vector. Then, preferably, in the case of the determined symbol that best symbolizes the recognized prototype, for example its distance and time of occurrence (timestamp), the transmission of data to the computer system is performed only if the value of this distance is below the first threshold and the recognized prototype is the prototype to be transmitted (step δ in FIG. 2).It is quite possible that several prototypes are stored that should not be recognized, e.g. noise, i.e. lack of reflection, etc. These data should not be transmitted, since they are not relevant for the detection of obstacles. Thus, a prototype is recognized (step γe in FIG. 2) when the determined distance value (prototype or prototype vector value) between the current feature vector value and the given feature vector value falls below this first threshold.

[0102] Therefore, it is no longer the ultrasound echo signal itself that is transmitted, but rather only a series of symbols of recognized typical time signal profiles in the echo signal and time stamps belonging to these signal profiles in a certain time (step δ in FIG. 2). Then, for each recognized signal object, only one symbol for the recognized signal form prototype, its parameters (e.g. envelope amplitude and / or temporal extension) and the time reference point of occurrence of this signal form prototype (time stamp) are preferably transmitted respectively as recognized signal object. The transmission of individual sample values ​​or the transmission of the time when a threshold is exceeded by the envelope of the sampled received signal, etc., is omitted. This selection of relevant prototypes thus results in a large amount of data compression and a reduction in the required bus bandwidth.

[0103] Thus, the quantitative detection of the presence of a combination of characteristics is performed forming an estimate, in this example for example the inverse distance between a representative of a signal object class in the form of a predetermined feature vector (prototype or prototype vector) and the subsequent transmission of compressed data to the computer system, if this estimate (for example the inverse distance) exceeds a second value or if the inverse estimate is below a first threshold. Thus, the signal processing unit of the ultrasonic sensor performs a data compression of the received signal to generate compressed data. The ultrasonic sensor then transmits the thus compressed data, preferably only the encodings (symbols) of the thus recognized prototypes, their amplitudes and / or temporal extensions and occurrence times (timestamps) to the computer system. Thereby, the EMC load caused by the data transmission via the data bus between the ultrasonic sensor and the computer system is minimized and during the time interval other data, for example status data of the ultrasonic sensor, can be transmitted to the computer system via the data bus between the ultrasonic sensor and the computer system for system error detection. Thereby, the latency is improved.

[0104] Within the framework of the present invention, it has been found that data should be transmitted with priority over the data bus. Reports to the computer system about safety-critical malfunctions of the sensor, i.e. the ultrasonic sensor in the present example, have the highest priority, since they impair with high probability the validity of the measurement data of the ultrasonic sensor. These data are supplied to the computer system by the ultrasonic system. The second highest priority belongs to requests from the computer system to perform safety-related self-tests. Such commands are transmitted from the computer system to the ultrasonic system. The data of the ultrasonic sensor itself has the third highest priority, since latency must not be extended. All other data have a (even) lower priority for transmission over the data bus.

[0105] It is particularly advantageous if the method for transmitting sensor data, in particular of an ultrasonic sensor, from a sensor to a computer system, in particular in a vehicle, comprises: Emitting an ultrasonic burst with a start (57) and end (56) of the emission of the ultrasonic burst; receiving an ultrasonic signal, forming a received signal for at least a receive time (TE) from the end (56) of the emission of the ultrasonic burst; and The transmission of compressed data via a data bus, in particular a single-wire data bus, to a computer system, whereby the transmission of data from the sensor to the computer system (54) begins with a start command (53) from the computer system to the sensor via the data bus, either before the end of the emission of an ultrasonic burst (56) or after a start command (53) from the computer system to the sensor via the data bus, which begins before the start of the emission of an ultrasonic burst (57), where after the start command (53) the transmission (54) continues periodically until the end of the data transmission (58), which occurs chronologically after the end of the reception time (TE).

[0106] Therefore, another variant of the proposed method provides for the formation of a feature vector signal (a stream of feature vectors with n feature vector values ​​and n as the dimension of the feature vector) from the received signal as a first step of data compression. Such a feature vector signal may contain several analog and digital data signals. Thus, the feature vector signal is a more or less complex data / signal structure. In the simplest case, it can be understood as a vector signal composed by several partial signals.

[0107] For example, it may be feasible to form first and / or higher time derivatives of the received signal, or single or multiple integrals of the received signal, which are multiple sub-signals within the feature vector signal.

[0108] It is also possible to form an envelope signal of the received signal, which is a subsignal of the feature vector signal.

[0109] Furthermore, it may be feasible to fold the received signal with the emitted ultrasound signal, thereby forming a correlation signal which may be a partial signal within the feature vector signal, where on the one hand a signal may be used as the emitted ultrasound signal that has been used to control the driver of the transmitter, or on the other hand a signal may be used which is for example measured at the transmitter and thus corresponds better to the sound waves actually emitted.

[0110] Finally, it may be feasible to detect occurrences of predetermined signal objects using matched filters and form matched filter signals for each of several predetermined signal objects. A matched filter is a filter that optimizes the signal-to-noise ratio (SNR). In noisy signals, predefined signal objects are detected. In the literature, several terms are often found: correlation filter, signal matched filter (SMF), or simply matched filter. A matched filter serves to optimally determine (detect) the presence of a known signal waveform, the amplitude and / or the position of a predetermined signal object even in the presence of interference, e.g., by signals from other ultrasonic transmitters and / or ground clutter (parameter estimation).

[0111] The matched filter signals are then preferably a plurality of sub-signals within the feature vector signal.

[0112] Particular events may be indicated in individual partial signals of the feature vector signal, which are elementary signal objects in the sense of the present invention. Elementary signal objects therefore do not comprise signal waveforms, e.g. rectangular pulses or other types of wavelets or wave trains, but distinctive points in the profile of the received signal and / or in the profile of signals derived therefrom, e.g. derived envelope signals that may be obtained from the received signal by filtering.

[0113] Another signal, which may be a partial signal of the feature vector signal, detects, for example, the envelope of the received signal, i.e. whether the envelope signal crosses a predefined third threshold, thus indicating the presence of a basic signal object in the received signal and thus the feature vector signal.

[0114] Another signal, which may be a sub-signal of the feature vector signal, may for example detect whether the envelope of the received signal, i.e. the envelope signal, crosses upwardly a predefined fourth threshold, which may be identical to the third threshold, thus indicating the presence of a basic signal object in the received signal and thus the feature vector signal.

[0115] Another signal, which may be a partial signal of the feature vector signal, detects for example whether the envelope of the received signal, i.e. the envelope signal, crosses downwards a predetermined fifth threshold, which may be identical to the third or fourth threshold, thus indicating the presence of a basic signal object in the received signal and thus the feature vector signal.

[0116] Another signal, which may be a partial signal of the feature vector signal, detects for example whether the envelope of the received signal, i.e. the envelope signal, has a maximum above a sixth threshold, which may be identical to the third to fifth thresholds mentioned above, and thus indicates the presence of a basic signal object in the received signal and thus the feature vector signal.

[0117] Another signal, which may be a partial signal of the feature vector signal, detects for example whether the envelope of the received signal, i.e. the envelope signal, has a minimum above a seventh threshold, which may be identical to the third to sixth thresholds mentioned above, and thus indicates the presence of a basic signal object in the received signal and thus the feature vector signal.

[0118] Here, it is preferably evaluated whether at least one preceding maximum of the envelope has a minimum distance to the minimum in order to avoid the detection of noise. Other filterings can be considered at this point. It can also be checked whether the time distance between this minimum and the preceding maximum is greater than a first minimum time distance. The fulfillment of each of these conditions respectively sets a flag or a signal which is preferably itself a partial signal of the feature vector signal.

[0119] Similarly, it must be checked in a similar manner whether the time and amplitude related distances of other signal objects meet certain validity requirements, such as observing a minimum time distance and / or a minimum amplitude distance. From these checks, further sub-signals, analog, binary or digital, can be derived, which further increases the dimensionality of the feature vector signal.

[0120] If necessary, the feature vector signals can be transformed into significant feature vector signals in a significance augmentation step, for example by a linear map or a higher order matrix polynomial, but practice shows that this is not yet necessary, at least in some current applications.

[0121] According to the method of the present invention, detection and classification of signal objects in the received signal into recognized signal object classes is then performed based on the feature vector signal or the significant feature vector signal.

[0122] For example, if the amplitude of the output signal of the matched filter, and thus the amplitude of the partial signal of the feature vector signal, exceeds a matched filter-specific (e.g., 8th) threshold, the signal object for detection can be considered as recognized, and this detection configures the matched filter. Here, preferably, other parameters are taken into account as well. For example, if an ultrasonic burst with an increasing frequency is transmitted during the burst (called chirp-up), an echo with this modulation function is expected. If the signal waveform of the envelope, for example the triangular signal waveform of the envelope, locally matches the signal waveform expected in time, but does not match the modulation function, this is not an echo from the transmitter, but a noise signal that may result from other ultrasonic transmitters or overreach. In that respect, the system can distinguish between multiple specific echoes and multiple external echoes, whereby one or the same signal waveform is associated with two different signal objects, namely multiple specific echoes and multiple external echoes. Here, the transmission of multiple specific echoes from the sensor to the computer system via the data bus is preferably prioritized over the transmission of multiple external echoes, because the former is generally safety-related and the latter is generally not safety-related.

[0123] Typically, at least one signal object parameter is assigned to each recognized signal object or determined for this signal object. Preferably, this is a time stamp indicating when the object occurred in the echo signal. The time stamp can be called, for example, the time start of the signal object in the received signal, the time end or time duration of the signal object, or the time position of the time focus of the signal object, etc. Other signal object parameters such as amplitude, extension, etc. are conceivable as well. In a variant of the proposed method, at least one of the assigned signal object parameters is transmitted together with the symbol of at least one signal object class to which at least one recognized class of the signal objects belongs. The signal object parameter is preferably a time value as a time stamp, indicating a time position suitable for drawing conclusions therefrom with respect to the time from the emission of the preceding ultrasonic burst. Preferably, this is thus determined and transmitted and is used to later determine the distance of the object from there depending on the transmitted time value.

[0124] Finally, a prioritized transmission of the recognized signal object classes in the form of assigned symbols with a time stamp is preferably performed together with each assigned signal object parameter. The transmission can also be performed in more complex data structures (records). For example, it is conceivable to first transmit the times of the recognized safety-related signal objects (e.g. identified obstacles) and then the recognized signal object classes of the safety-related signal objects. In this way the latency is further reduced.

[0125] The proposed method, at least in a variant, comprises the determination of a chirp value as an assigned signal object parameter indicating whether the recognized signal object is an echo of an ultrasonic transmit burst having multiple chirp up functions, multiple chirp down functions or multiple no-chirp functions. "Chirp up" means that the frequency increases in the received signal object of the received signal. "Chirp down" means that the frequency decreases in the received signal object of the received signal. "No-chirp" means that the frequency remains substantially the same in the received signal object of the received signal.

[0126] In a variant of this method, a confidence signal (confidence value) is also generated by forming a correlation, for example forming a time-continuous or time-discrete correlation signal between the received signal, or a signal derived from the received signal instead of the received signal, on the one hand, and a reference signal, for example an ultrasonic transmission signal or another expected wavelet, on the other hand. The confidence signal is then usually a partial signal of the feature vector signal, i.e. a component of the feature vector, which is constituted by a series of vector sample values ​​(feature vector values).

[0127] In a variant of the method, a phase signal is also formed on this basis, which signal indicates for example a phase shift of the received signal or a signal formed therefrom (for example a confidence signal), with respect to a reference signal, for example an ultrasonic transmission signal and / or another type of reference signal. The phase signal is then usually also a partial signal of a feature vector signal, i.e. a component of a feature vector constituted by a sequence of a number of vector sample values.

[0128] In a similar manner, according to a further variant of the proposed method, it is possible to form the phase confidence signal by forming a correlation between the phase signal or a signal derived therefrom on the one hand and a reference signal, and it is also possible to use the same as a partial signal of the feature vector signal. The phase confidence signal is then usually also a partial signal of the feature vector signal, i.e. a component of the feature vector constituted by a sequence of vector sample values.

[0129] Upon evaluating the feature vector signal, it may be feasible to perform a comparison of the phase confidence signal with one or more thresholds to generate a discretized phase confidence signal which itself may again become part of the feature vector signal.

[0130] In a variant of the proposed method, the evaluation of the feature vector signal and / or the significant feature vector signal may be performed by forming one or more distance values ​​between the feature vector signal and one or more signal object prototype values ​​for a number of recognizable signal object classes.

[0131] Such distance values ​​can be Boolean, binary, discrete, digital or analog values. Preferably, all distance values ​​are linked together with a non-linear function. Thus, if chirp-up echoes of triangular type are expected, received chirp-down echoes of triangular type can be rejected. In the sense of the present invention, this rejection is a "non-linear" process.

[0132] Conversely, the triangle can be defined differently in the received signal. This is done by first considering the amplitude of the triangle in the received signal. If the amplitude of the received signal is sufficient, the matched filter associated with this triangle signal provides a signal that exceeds, for example, a predefined 9th threshold. In this case, a recognized signal object can be assigned to this signal object class (for a plurality of triangle signals) at the point of exceedance. In this case, the distance value (here, the 9th threshold) between the feature vector signal and the prototype undershoots one or more predefined binary, digital or analog distance values ​​(here, 0 is the crossing).

[0133] Within the framework of the invention, it has been found to be advantageous to first transmit data of recognized signal objects of echoes that arrive very quickly from the sensor to the computer system, and only transmit subsequent data of the recognized signal objects later. Preferably, at least the recognized signal object class and a time stamp are transmitted each time, the time stamp indicating when the signal object arrives at the sensor. In the framework of the detection process, a score can be assigned to the different signal objects considered as sections of the received signal, the score indicating the probability due to the presence of this signal object according to the estimation algorithm used. In the simplest case, such a score is binary. However, it is preferably complex, real or integer. It may be, for example, a determined distance. In the case of signal objects with high score values, it may also be feasible to transmit data of recognized signal objects with lower scores. In order to enable correct processing by the computer system, the determined score value as well as the date (symbol) of the recognized signal object and the time stamp of the respective signal object must be transmitted in this case. Instead of transmitting only the date (symbol) of the recognized signal object and the timestamp for the signal object corresponding to this symbol, the date (symbol) of the signal object with the second smallest distance and the timestamp for the signal object corresponding to this second most probable signal object can be additionally transmitted. Thus, in this case, a list of hypotheses including two recognized signal objects, their time positions and additionally assigned multiple score values ​​is transmitted to the computer system. Similarly, a list of hypotheses including more than two symbols of more than two recognized signal objects, their time positions and additionally assigned multiple score values ​​can also be transmitted to the computer system.

[0134] Preferably, the data on the recognized signal object classes and the assigned data, for example the timestamps and scores of each of the recognized signal object classes, i.e. the signal object parameters, are transmitted according to the FIFO principle, so that the data of the reflections of the nearest objects are always transmitted first, in this way ensuring that safety-critical events of vehicle collisions with obstacles are processed in a prioritized manner according to probability.

[0135] In addition to the transmission of the measurement data, the transmission of error states of the sensor can also be performed. This can also take place during the reception time TE if the sensor, via the self-test device, determines that a defect exists and that the previously transmitted data is potentially defective. It is thus ensured that the computer system obtains information about changes in the evaluation of the measurement data as early as possible and can reject them or process them in some other way. This is particularly important for emergency braking systems, since emergency braking is a safety-critical operation that should only be initiated if the underlying data has a corresponding confidence value, and also for other driver assistance systems. In comparison, the transmission of the measurement data, for example the date of the recognized signal object class and / or the assigned single signal object parameters, is therefore postponed and thus has a lower priority. Of course, it is also conceivable that the transmission is interrupted if an error occurs in the sensor. However, even if an error appears to occur, its presence is not confirmed. In such cases, the continuation of the transmission is therefore recommended. The transmission of safety-critical errors of the sensor is therefore performed with a higher priority.

[0136] In addition to the above-mentioned wavelets with an integral value of zero and signal sections with integral values ​​different from zero, additionally denoted herein as wavelets, certain positions / phases in the profile of the received signal are considered as elementary received signal objects in the sense of the present invention, which may be used for data compression and transmitted instead of sample values ​​of the received signal. This subset of the set of possible elementary signal objects is referred to below as signal profile points. Signal profile points are therefore a special form of elementary signal objects in the sense of the present invention.

[0137] The first possible signal profile point, and therefore the basic signal object, is the intersection in the ascending direction of the profile of the threshold signal (SW) with the envelope signal (1).

[0138] A second possible signal profile point, and therefore the fundamental signal object, is the intersection in the downward direction of the profile of the threshold signal (SW) with the envelope signal (1).

[0139] The third possible signal profile point, and therefore the fundamental signal object, is a local or absolute maximum in the profile of the envelope signal (1) that exceeds the amplitude of the thirteenth threshold signal (SW).

[0140] A fourth possible signal profile point, and therefore a fundamental signal object, is a local or absolute minimum of the profile of the envelope signal (1) that exceeds a threshold signal (SW).

[0141] For these four exemplary types of signal profile points, as well as other types of signal profile points, it may possibly be feasible to use a threshold signal (SW) that is typical for the underlying signal object.

[0142] Usually, the time sequence of the elementary signal objects is not arbitrary. The invention takes advantage of this, since it is preferable to transmit not elementary signal objects of a simpler nature, but recognized patterns of the time sequence of these elementary signal objects, which represent the actual signal objects. For example, if a triangular wavelet with sufficient amplitude is predicted in the envelope signal (1), what can be predicted in time correlation with the exceedance of said minimum level at the output of a matched filter suitable for the detection of such a triangular wavelet, as well as the corresponding minimum level at the output of said matched filter, is: 1. The occurrence of a first possible signal profile point in a threshold signal (SW) that is crossed upward by the amplitude of an envelope signal (1), followed by 2. The occurrence of a second possible signal profile point at a maximum of the envelope signal (1) that exceeds the amplitude of one or a particular threshold signal (SW), followed by 3. The occurrence of a third possible signal profile point at one or a particular threshold signal (SW) that is crossed by the amplitude of the envelope signal (1) in a descending direction.

[0143] Thus, in this example, an exemplary signal object of a triangular wavelet is constituted by a predefined sequence of three basic signal objects, from which the signal object is recognized and assigned to a signal object class, where this information is transmitted as a symbol of the signal object class and parameters describing the recognized signal object, such as, inter alia, a time of occurrence, i.e., a time stamp. Additionally, this exceedance of the minimum level at the output of the matched filter is a fifth possible signal profile point and thus yet another example of a possible signal profile characteristic.

[0144] The resulting grouping and time sequence of the recognized basic signal objects may itself be recognized by the Viterbi decoder as a predefined expected grouping or time sequence of basic signal objects and may itself be a basic signal object. Thus, such a predefined grouping and / or time sequence of other basic signal objects is a sixth possible signal profile point and thus a basic signal object.

[0145] If such a grouping of signal profile characteristics or a time-series sequence of signal object classes is recognized, then a transmission of the recognized summary signal object class and of the symbols of at least one assigned signal object parameter is preferably carried out subsequently instead of the transmission of the individual basic signal objects, since significant data bus capacity is saved in this manner. It is possible that both are transmitted. Here, the date (symbol) of the signal object class of the signal object is transmitted, which is a predefined time-series sequence and / or grouping of the other basic signal objects. To achieve compression, it is advantageous if at least one signal object class (symbol) of at least one of the other basic signal objects is not transmitted.

[0146] A time-series grouping of a number of elementary signal objects is given, in particular if the distance in time of these elementary signal objects does not exceed a predefined distance. In the above example, the propagation time of the signal in the matched filter must be taken into account. Matched filters are usually slower than comparators. The changes in the output signal of the matched filter must therefore have a fixed time correlation with the occurrence in time of the associated signal profile points.

[0147] According to a variant of the invention, a method is provided for transmitting sensor data from a sensor, in particular an ultrasonic sensor, in a vehicle, to a computer system, starting after the emission of an ultrasonic burst and the reception of an ultrasonic signal forming a time-discrete received signal constituted by a sequence of a plurality of sample values, a time-dated (timestamp) being assigned to each sample value. The method starts with the use of at least one suitable filter (e.g. matched filter) from a sequence of sample values ​​of the received signal with the determination of at least two parameter signals regarding the presence of a respective elementary signal object assigned to each parameter signal. The resulting parameter signals (feature vector signals) are also constructed as a time-discrete sequence of respective parameter signal values ​​(feature vector values) associated with a date (timestamp). Thus, preferably, exactly one temporal date (timestamp) is assigned to each parameter signal value (feature vector value). These parameter signals are hereinafter collectively referred to as feature vector signal. Thus, the feature vector signal is constructed as a time-discrete sequence of feature vector signal values, each having n parameter signal values, which are constituted by a plurality of parameter signal values ​​and a further plurality of parameter signal values, each having the same temporal date (timestamp), where n is the dimensionality of the respective plurality of feature signal values, which are preferably identical from one feature vector value to the next. Each feature vector signal value thus formed has a respective temporal date (timestamp) assigned to it. The resulting time profile of the feature vector signal in the n-dimensional phase space is then evaluated and the recognized signal object is concluded while determining an evaluation value (e.g. distance). As explained above, a signal object is formed by a temporal sequence of elementary signal objects. Usually, a predefined symbol is assigned to the signal object. In metaphorical terms, it is checked whether on a path through the n-dimensional space, a point pointed to by the n-dimensional feature vector signal in the n-dimensional phase space approaches, in a predefined time sequence, a predefined point of this n-dimensional phase space to a distance smaller than a predefined maximum distance. The feature vector signal has a temporal profile. An evaluation value (e.g. distance) is then calculated, which may indicate, for example, the probability that a certain sequence exists. This evaluation value is then again assigned a temporal date (timestamp).This evaluation value is compared to a threshold vector forming a Boolean result which may have a first and a second value. If this Boolean result of the timestamp has a first value, the symbol of the signal object and the timestamp assigned to it are transmitted from the sensor to the computer system. If necessary, further parameters can be transmitted depending on the recognized signal object.

[0148] It is particularly preferred that the data transmission in the vehicle is carried out via a serial bidirectional single-wire data bus. An electrical return path is preferably ensured by the vehicle body. Preferably, the sensor data is transmitted to the computer system in a current modulated state. The data for controlling the sensor is transmitted by the computer system to the sensor, preferably in a voltage modulated state. According to the invention, it has been found that the use of a PSI5 data bus and / or a DSI3 data bus is particularly suitable for the data transmission. Furthermore, it has been found to be particularly advantageous to carry out the transmission of data to the computer system with a transmission rate of more than 200 kBit / s and to carry out the transmission of data from the computer system to the at least one sensor with a transmission rate of more than 10 kBit / s, preferably more than 20 kBit / s. Furthermore, it has been found that the transmission of data from the sensor to the computer system should be modulated on a data bus whose current is less than 50 mA, preferably less than 5 mA, more preferably less than 2.5 mA. These buses must be adapted to these operating values ​​in a corresponding manner. However, the basic principle remains the same. To carry out the above-mentioned method, a computer system is required with a data interface to the above-mentioned data bus, preferably to the above-mentioned single-wire data bus, which supports the decompression of the data thus compressed. However, in general the computer system does not perform a full decompression, but may for example only evaluate the timestamp and the recognized signal object type. A sensor required to perform one of the above methods comprises at least one transmitter and at least one receiver for generating a received signal, which may be implemented in combination as one or more converters. Furthermore, the sensor comprises at least a number of devices for processing and compressing the received signal, as well as a data interface for transmitting data to the computer system via a data bus, preferably the above-mentioned single-wire data bus. For compression, the compression device preferably comprises at least one of the following units: Matched filters, comparator, A multiple threshold signal generator for generating one or more threshold signals (SW) Differentiators for forming derivatives, an integrator for forming a plurality of integrated signals; Several other filters, an envelope former for generating an envelope signal from the received signal; A plurality of correlation filters for comparing the received signal or signals derived therefrom with a plurality of reference signals.

[0149] In a particularly simple form, the proposed method for transmitting sensor data, in particular of ultrasonic sensors, from a sensor to a computer system, in particular in a vehicle, is carried out as follows.

[0150] The preceding steps are, for example, the emission of an ultrasonic burst and the reception of an ultrasonic signal, i.e. typically a reflection, and the formation of a time-discrete received signal constituted by a time sequence of a plurality of sample values. Here, a time date (time stamp) is assigned to each sample value. Usually, this indicates the time of sampling. According to the invention, on the basis of this data stream, a first characteristic of a first parameter signal is determined from a sequence of a plurality of sample values ​​of the received signal using a first filter. Preferably, the parameter signal is again formed as a time-discrete sequence of a plurality of parameter signals. Again, exactly one time date (time stamp) is assigned to each parameter signal value. This date preferably corresponds to the most recent time date of the sample values ​​used to form this respective parameter signal value. At the same time, preferably, at least one further parameter signal and / or a characteristic assigned to this further parameter signal is determined from a sequence of a plurality of sample values ​​of the received signal using another filter assigned to this further parameter signal, and the respective further parameter signal values ​​are again formed as a time-discrete sequence of a further plurality of parameter signal values. Again, the same temporal date (timestamp) assigned to the corresponding parameter signal value is also assigned to each further parameter signal value.

[0151] Hereinafter, the first parameter signal and the further plurality of parameter signals are jointly referred to as a parameter vector signal or a feature vector signal. This feature vector signal is thus a time-discrete sequence of feature vector signal values ​​formed by the parameter signal values ​​and the further plurality of parameter signal values ​​each having the same temporal date (timestamp). This respective temporal date (timestamp) may thus be associated with each feature vector signal value thus formed, i.e. each parameter signal value.

[0152] Preferably, a quasi-continuous comparison of the feature vector signal values ​​of the temporal dates (timestamps) with a threshold vector, preferably a prototype vector, is then performed to obtain a Boolean result which may have a first and a second value. For example, it is conceivable to compare the amount of the current feature vector signal value, for example representing a first component of the feature vector signal value, with a threshold representing the first component of the threshold vector, and if the amount of the feature vector signal value is smaller than this threshold, set the Boolean result to a first value, otherwise set the Boolean result to a second value. If the Boolean result has a first value, it is further conceivable to compare the amount of the feature vector signal, for example representing a further component of this feature vector signal, with a further threshold representing a further component of the threshold, and if the amount of the further feature vector signal is smaller than this threshold, leave the Boolean result at the first value, otherwise set the Boolean result to a second value. In this manner, all further feature vector signal values ​​can be ascertained. Of course, other classifiers are conceivable. A comparison with a number of different threshold vectors is likewise possible. These threshold vectors thus represent a number of prototypes of a given number of signal waveforms. They arise from the library mentioned above. Preferably again, each threshold vector has a symbol assigned to it.

[0153] As a final step in this case, the symbol and possibly the feature vector signal value and a timestamp assigned to this symbol or feature vector signal value are transmitted from the sensor to the computer system if the Boolean result for this timestamp has a first value.

[0154] Therefore, all other dates are not transmitted (anymore). Furthermore, interference is avoided by multi-dimensional evaluation.

[0155] Therefore, based on this, a sensor system is proposed, comprising at least one computer system configured to execute one of the above-mentioned methods and at least two sensors configured to execute one of the above-mentioned methods. These two sensors can communicate with the computer system via signal object recognition, transmit multiple external echoes in a compact manner and provide this additional information to the computer system. Thus, a sensor system is typically provided to ensure that data transmission between the at least one sensor and the computer system is or can be performed according to the above-mentioned methods. At least two sensors of the sensor system typically transmit one ultrasonic reception signal each, i.e. at least two ultrasonic reception signals are compressed using one of the above-mentioned proposed methods and transmitted to the computer system. Here, in the computer system, the at least two ultrasonic reception signals are reconstructed into multiple ultrasonic reception signals. The computer system then uses the reconstructed multiple ultrasonic reception signals to perform object recognition of multiple objects in the environment of the multiple sensors. Unlike the prior art, the sensors do not perform object recognition. They only provide data on the recognized multiple signal objects and the encoding of their parameters, and therefore transmit the reception signal profile in a compressed manner.

[0156] In addition, the computer system preferably performs recognition of objects, i.e. multiple obstacles in the multiple sensor environment, using the reconstructed ultrasonic received signal and possibly additional signals of further multiple sensors, in particular the signal of a radar sensor.

[0157] As a final step, based on the recognized objects, the computer system preferably creates an environment map of the sensors or the device having the sensors to which the sensors belong or which is part of it.

[0158] The compressed transmission of signal profile data between the sensor and the computer system via the data bus as provided by the invention reduces, on the one hand, the load of the data bus and therefore its importance with regard to EMC requirements, and on the other hand creates free data bus capacity for the transmission of control commands from the computer system to the sensor and for the transmission of status information and other data from the sensor to the computer system. The priority with which the compressed data of the received signal profile and other data such as status information and error messages are transmitted ensures that safety-relevant dates are transmitted first and thus does not cause unnecessary dead times of the sensor.

[0159] Fig. 3a shows a conventional temporal profile of an ultrasonic echo signal (1) (see the wider solid line) and its conventional evaluation in any number of units. Starting with the emission of a transmit burst (SB) (see the leftmost signal profile section and reference SB in Fig. 3d), a threshold signal (SW) (see the dashed line) is transmitted. Each time the envelope signal of the ultrasonic echo signal (1) exceeds the threshold signal (SW), the output (2) (see the thinner solid line) is set to logic 1. This is a temporal analog interface with a digital output level. Further evaluation is then performed in the control device of the sensor. An indication of errors or control of the sensor is not possible via the analog interface corresponding to the prior art.

[0160] Fig. 3b shows a conventional temporal profile (1) of an ultrasonic echo signal and its conventional evaluation in any number of units. Starting with the emission of a transmission burst (SB), a threshold signal (SW) is transmitted. However, each time the envelope signal (1) of the ultrasonic echo signal exceeds the threshold signal (SW), the output (2) is set here to a level corresponding to the magnitude of the detected reflection (see the wider dashed-dotted line). This is a temporal analog interface with a digital output level. Further evaluations are then carried out on the sensor. An indication of errors or a control of the sensor is not possible via the analog interface corresponding to the prior art.

[0161] FIG. 3c is an illustration of an ultrasound echo signal, where the chirp directions (eg, A for chirp up, B for chirp down) are indicated by hatched lines from upper left to lower right and lower left to upper right, respectively.

[0162] In Fig. 3d the principle of signal transmission of symbols is explained. Instead of the signal of Fig. 3c, only two types of (triangular) signal objects are transmitted as an example. Specifically, these are a first triangular object (A) (shown in Fig. 3d for both chirp up and chirp down cases) and a second triangular object (B) (shown for the chirp down case). At the same time, the time, the peak values ​​and the base width of the triangular object are transmitted. If the signal is reconstructed based on these data, a signal corresponding to Fig. 3d is obtained. Signal parts that do not correspond to the triangular signals have been deleted from this signal. Unrecognized signal parts are therefore rejected, which leads to a large amount of data compression.

[0163] FIG. 4a shows a conventional analog transmission of the intersection of an envelope signal (1) and a threshold signal (SW) of an ultrasonic echo signal, which is not claimed by the present invention.

[0164] FIG. 4b illustrates the transmission of analyzed data after complete reception of the ultrasound echo, which is not claimed by the present invention.

[0165] FIG. 4c illustrates the transmission of compressed data as claimed by the present invention, where in this example symbols of elementary signal objects are transmitted with little to no compression.

[0166] Fig. 5 shows the claimed transmission of compressed data, where in this example symbols for a number of elementary signal objects are compressed to form symbols of a number of signal objects. First, a triangular object (59) is transmitted with a time sequence of threshold exceedances and maxima of threshold undershoots (see time sequence of signal profile points 5, 6, 7 of the signal profile of an ultrasound echo signal in the top diagram of Fig. 5). Then, a double peak with a saddle point (60) above the threshold signal is recognized, where exceedances of a threshold (SW) by the envelope signal (1), maxima of the envelope signal (1), minima above the threshold signal (SW) and undershoots of the threshold signal (SW) (see signal profile points 8, 9, 10, 11, 12 in the top diagram) are characteristic. After recognition, a symbol of this double peak with saddle point is transmitted. A time stamp is then transmitted. Preferably, further parameters of the double peak with saddle point are also transmitted, for example with maxima and minima positions or with a scaling factor. Then, a triangular signal (61) (i.e., a basic signal object) is recognized again as an exceedance of the threshold signal (SW) by the envelope signal (1), followed again by an undershoot of the threshold signal (SW) by the envelope signal (1) (see the signal profile points 13, 14, 15 in the figure above). This is again followed by the recognition of a double peak (62), but where the envelope signal minimum is below the threshold signal (SW) (see the signal profile points 16, 17, 18, 19, 20, 21 in the figure above). This double peak can thus be treated, for example, as a separate signal object. Finally, a triangular signal is recognized from the signal profile points 22, 23, 24 in the figure above. It can be easily seen that the processing of this signal reduces a large amount of data.

[0167] Figure 6 shows the claimed compressed data transmission corresponding to Figure 3, where in this example not only the envelope signal but also the confidence signal is evaluated. In the top and center diagrams of Figure 6, the thick dashed line indicates the threshold signal. It can be seen that the received signal is evaluated only if it exceeds the threshold signal. The dotted signal profile line in the center diagram of Figure 6 indicates that the signal object is chirp-up or chirp-down modulated (see also Figures 3c and 3d, where chirp-up and chirp-down are distinguished by different diagonal hatching). An upward dotted signal profile line indicates that the signal object has been recognized as chirp-up modulated, and a downward dotted signal profile line indicates that the signal object has been recognized as chirp-down modulated.

[0168] For the purposes above and below, it should be noted that the terms are defined as follows: A signal object is also called a signal profile object. The signal object class is also called the signal profile object class. A symbol is an identifier for a signal profile object class. The signal object parameters are synonymous with object parameters. The fundamental signal waveform refers to the signal profile characteristics.

[0169] By definition, a signal object is formed by two or more elementary signal objects, i.e. a signal profile object is formed by one or more signal profile characteristics. A signal object belongs to one of several signal object classes. A signal object may be described by one or more signal object parameters, i.e. in terms of position, magnitude, distortion, elongation.

[0170] A basic signal object is also called a basic signal profile function, or signal profile feature.

[0171] The parameters additionally describe the shape of the signal object.

[0172] The parameter signal is made up of a number of parameter signal values.

[0173] The feature vector signal is composed of a number of parameter signals.

[0174] The value of the feature vector signal is made up of a number of parameter signal values.

[0175] The plurality of parameter signals forms a feature vector signal, the feature vector signal value, also referred to as the parameter vector signal value, comprising the plurality of parameter signal values.

[0176] Individual embodiments of the invention are presented below, it being noted that the individual groups of features presented below can be combined with each other as required (reference numbers refer to Figures 1 and 2 and are to be understood as being merely exemplary, i.e. not limiting).

[0177] 1. A method for transmitting sensor data, particularly of an ultrasonic sensor, from a sensor, particularly in a vehicle, to a computer system, comprising: emitting an ultrasonic burst; receiving an ultrasonic sensor and forming a received signal; performing data compression of the received signal to generate compressed data; transmitting said compressed data to said computer system; A method for providing the above.

[0178] 2. A method for transmitting sensor data, particularly of an ultrasonic sensor, from a sensor, particularly in a vehicle, to a computer system, comprising: emitting an ultrasonic burst at a start 57 and an end 56 of the emission of the ultrasonic burst; receiving an ultrasonic signal and forming a received signal at least for a receive time TE from the end 56 of the emission of the ultrasonic burst; transmitting said compressed data to said computer system via a data bus, in particular a single-wire data bus; Equipped with Transmitting 54 data from said sensor to said computer system; After a start command 53 from the computer system to the sensor via the data bus and before the end 56 of the emission of an ultrasonic burst, or starting after a start command 53 from the computer system to the sensor via the data bus and before the start 57 of the emission of the ultrasonic burst; After the start command 53, said transmission 54 continues periodically until the end of said data transmission 58, How this end of said data transmission 58 occurs chronologically after the end of the reception time TE.

[0179] 3. A method for transmitting sensor data, particularly in a vehicle, from a sensor, particularly an ultrasonic sensor, to a computer system, comprising: emitting an ultrasonic burst; receiving an ultrasonic sensor and forming a received signal; forming a feature vector signal from the received signal; recognising and classifying signal profile characteristics of a plurality of recognised signal profile characteristic classes within the received signal, wherein at least one assigned signal profile characteristic parameter is assigned to or determined for each recognised signal profile characteristic; A method comprising the step of performing prioritized transmission of at least one recognized signal profile characteristic class and at least one assigned signal profile characteristic parameter.

[0180] 4. The method described in paragraph 3, wherein at least one of the assigned plurality of signal profile characteristic parameters transmitted together with at least one recognized signal profile characteristic class is a time value indicating a time position suitable for drawing a conclusion regarding the time from the emission of a preceding ultrasonic burst.

[0181] 5. The method of claim 4, comprising the additional step of determining the distance of an obstacle object depending on the time value.

[0182] 6. The method of claim 3, further comprising a step of determining a chirp value as the assigned signal profile characteristic parameter, the signal profile characteristic parameter indicating whether the recognized signal profile characteristic is an echo of an ultrasonic transmit burst having multiple functions of chirp up, chirp down or no chirp.

[0183] 7. A method according to claim 3, comprising the step of generating a confidence signal by forming a correlation between the received signal or a signal derived therefrom and a reference signal.

[0184] 8. The method of claim 3, comprising the step of generating a phase signal.

[0185] 9. A method according to claim 8, comprising the step of generating a phase confidence signal by forming a correlation between the phase signal or a signal derived therefrom and a reference signal.

[0186] 10. The method of claim 9, comprising comparing the phase confidence signal to one or more thresholds to generate a discrete phase confidence signal.

[0187] 11. forming at least one binary, digital, or analog distance value between the feature vector signal and one or more signal profile characteristic prototype values ​​for a plurality of recognizable signal profile characteristic classes; assigning a recognizable signal profile characteristic class as a recognized signal profile characteristic if the distance value undershoots one or more predetermined binary, digital or analog distance values. Item 4. The method according to item 3, comprising:

[0188] 12. The method of claim 3, wherein at least one signal profile characteristic class is a wavelet.

[0189] 13. The method of claim 3, wherein the at least one wavelet is a triangular wavelet.

[0190] 14. The method of claim 12, wherein the at least one wavelet is a rectangular wavelet.

[0191] 15. The method of claim 12, wherein the at least one wavelet is a half-sine wavelet.

[0192] 16. One of the plurality of signal object parameters is time shift of the wavelet of the recognized signal profile characteristics, a temporal compression or expansion of the wavelet of the recognized signal profile characteristic; or the amplitude of said wavelet of said recognized signal profile characteristic; Item 13. The method according to item 12, wherein

[0193] 17. The method according to item 3, wherein the transmission of the at least one recognized signal profile characteristic class and the at least one assigned signal profile characteristic parameter is performed according to the FIFO principle.

[0194] 18. The transmission of the error state of the above sensor is transmitting said at least one recognised signal profile characteristic class; and / or Transmission of assigned signal profile characteristic parameters from 1 above The method of claim 3, wherein the method is executed with a higher priority than

[0195] 19. The method of claim 3, wherein the signal profile characteristic is the amplitude of the envelope signal 1 crossing the amplitude of the threshold signal SW in an ascending direction.

[0196] 20. The method of claim 3, wherein the signal profile characteristic is the amplitude of the envelope signal 1 crossing the amplitude of the threshold signal SW in a descending direction.

[0197] 21. The method of claim 3, wherein the signal profile characteristic is a maximum of the amplitude of the envelope signal 1 over the amplitude of the threshold signal SW.

[0198] 22. The method of claim 3, wherein the signal profile characteristic is the minimum of the amplitude of the envelope signal 1 that exceeds the amplitude of the threshold signal SW.

[0199] 23. The method of claim 3, wherein the signal profile characteristic is a predetermined temporal sequence and / or temporal grouping of a plurality of other signal profile characteristics.

[0200] 24. The method according to clause 23, wherein the transmission of the at least one recognised signal profile characteristic class and the at least one assigned signal profile characteristic parameter is a transmission of the signal profile characteristic class of a signal profile characteristic, which is a predefined sequence of a plurality of other signal profile characteristics, and at least one signal profile characteristic class of at least one of these other signal profile characteristics is not transmitted.

[0201] 25. The method of claim 1 or 3, wherein the data transmission is performed via a bidirectional single-wire data bus, wherein the sensor transmits the data to the computer system in a current modulation manner and the computer system transmits the data to the sensor in a voltage modulation manner.

[0202] 26. The method according to claim 25, characterized in that a PSI5 data bus and / or a DSI3 data bus is used for data transmission.

[0203] 27. The method according to claim 25, wherein the transmission of the data to the computer system is performed at a transmission speed of more than 200 kBit / s and the transmission of the data from the computer system to at least one sensor is performed at a transmission speed of more than 10 kBit / s, preferably more than 20 kBit / s.

[0204] 28. The method according to claim 25, wherein for the data transmission from the sensor to the computer system, a transmission current is modulated on the data bus, the current intensity of the transmission current being less than 50 mA, preferably less than 5 mA.

[0205] 29. A sensor, in particular an ultrasonic sensor, suitable for carrying out the method according to one or more of paragraphs 1 to 28.

[0206] 30. A computer system adapted to carry out the method according to one or more of clauses 1 to 28.

[0207] 31. A method for transmitting sensor data, particularly in an ultrasonic sensor, from a sensor to a computer system, particularly in a vehicle, comprising: emitting an ultrasonic burst α; receiving an ultrasonic sensor and forming a received signal β; performing data compression of the received signal to generate compressed data γ; transmitting said compressed data to a computer system δ; A method comprising the steps of transmitting sensor data together or the above.

[0208] 32. A method for transmitting sensor data, particularly in an ultrasonic sensor, from a sensor to a computer system, particularly in a vehicle, comprising: emitting an ultrasonic burst at a start 57 and an end 56 of the emission of the ultrasonic burst α; receiving an ultrasonic signal and forming a received signal at least during a receive time TE from the end 56 of the emission of the ultrasonic burst β; transmitting the compressed data to the computer systems γ, δ via a data bus, in particular a single-wire data bus; or comprising the steps of transmitting sensor data together; Transmitting 54 data from said sensor to said computer system; After a start command 53 from the computer system to the sensor via the data bus, but before the end 56 of the emission of the ultrasonic burst; or Beginning after a start command 53 via a data bus from the computer system to the sensor and before the start 57 of the emission of the ultrasonic burst; After the start command 53, the transmission 54 continues periodically until the end of the data transmission 58, The manner in which the end of this data transmission 58 occurs chronologically after the end of the reception time TE.

[0209] 33. A method for transmitting sensor data from a sensor, particularly in an ultrasonic sensor, to a computer system, particularly in a vehicle, comprising or with the following steps: Emitting an ultrasonic burst. Receiving an ultrasonic sensor and forming a received signal. forming a feature vector signal from said received signal; Recognizing and classifying a plurality of signal profile characteristics within a plurality of recognized signal object classes within the received signal. wherein to each signal object thus recognized and classified, at least one assigned signal object parameter and one symbol are assigned corresponding to the signal object class assigned to the signal object; Wherein, for each signal object thus recognized and classified, at least one assigned signal object parameter and a symbol of this signal object are determined. Transmitting at least a symbol of the recognized signal object class and at least one assigned signal object parameter of the recognized signal object class.

[0210] 34. The method according to clause 33, wherein the transmission of at least a symbol of a recognised signal object class and at least one assigned signal object parameter of this recognised signal object class is performed with priority.

[0211] 35. The method according to one or more of clauses 33 to 34, wherein at least one assigned signal object parameter transmitted together with at least one recognized signal object class is a time value indicating a time position suitable for drawing a conclusion with respect to the time from the emission of a preceding ultrasonic burst.

[0212] 36. The method according to one or more of clauses 34 or 35, further comprising a step of determining the determined distance of the object depending on the time value.

[0213] 37. The method further comprises determining a chirp value as the assigned signal object parameter; 37. The method of one or more of clauses 33 to 36, wherein the assigned signal object parameter indicates whether the signal object is an echo of an ultrasonic transmit burst having multiple functions of chirp up, chirp down or no chirp.

[0214] 38. The method according to one or more of clauses 33 to 37, further comprising the step of generating a confidence signal by forming a correlation between the received signal or a signal derived therefrom and a reference signal.

[0215] 39. The method of one or more of clauses 33 to 38, further comprising the step of generating a phase signal.

[0216] 40. The method of claim 39, further comprising the step of generating a phase confidence signal by forming a correlation between the phase signal or a signal derived therefrom and a reference signal.

[0217] 41. The method of claim 40, further comprising comparing the phase confidence signal to one or more thresholds to generate a discrete phase confidence signal.

[0218] 42. forming at least one binary, digital, or analog distance value between said feature vector signal and one or more signal object prototype values ​​of a plurality of recognizable signal object classes; assigning a recognizable signal object class as a recognizable signal object if the magnitude of the distance value undershoots one or more predetermined binary, digital, or analog distance values. 42. The method according to one or more of clauses 33 to 41, further comprising:

[0219] 43. The method of one or more of paragraphs 33 to 42, wherein at least one signal object class is a wavelet.

[0220] 44. The method of claim 43, wherein at least one wavelet is a triangular wavelet.

[0221] 45. The method according to one or more of paragraphs 43 to 44, wherein at least one wavelet is a rectangular wavelet.

[0222] 46. ​​The method of one or more of paragraphs 43 to 46, wherein at least one wavelet is a half-sine wavelet.

[0223] 47. One of the signal object parameters is a time shift of the wavelet of said recognized signal object, a wavelet temporal compression or expansion of said recognized signal object; or The amplitude of the wavelet of the above recognized signal object 47. The method according to any one or more of paragraphs 43 to 46, wherein

[0224] 48. The method according to one or more of clauses 33 to 47, wherein the transmission of at least one recognized signal object class and at least one assigned signal object parameter is performed according to the FIFO principle, which means that recognized signal object classes having an earlier timestamp are transmitted earlier than recognized signal object classes having a later timestamp.

[0225] 49. The transmission of an error state of the above sensor is and / or Sending signal object parameters assigned to 1 49. The method according to claim 33, wherein the method is executed with higher priority than

[0226] 50. The method of one or more of clauses 33 to 49, wherein the signal object comprises a combination of two, three or four or more basic signal objects.

[0227] 51. The method according to claim 50, wherein the basic signal object is the intersection point at the intersection of the magnitude of the envelope signal 1 and the magnitude of the threshold signal SW.

[0228] 52. The method according to one or more of paragraphs 50 to 51, wherein the basic signal object is the intersection point at the intersection in the ascending direction of the magnitude of the envelope signal 1 and the magnitude of the threshold signal SW.

[0229] 53. The method according to one or more of clauses 50 to 52, wherein the basic signal object is the point of intersection of the magnitude of the envelope signal 1 and the magnitude of the threshold signal SW when they cross in the downward direction.

[0230] 54. The method according to one or more of clauses 50 to 53, wherein the basic signal object is the maximum of the amount of the amplitude of the envelope signal 1 that exceeds the amount of the threshold signal SW.

[0231] 55. The method according to one or more of clauses 50 to 54, wherein the basic signal object is the minimum of the amount of the amplitude of the envelope signal 1 that exceeds the amount of the threshold signal SW.

[0232] 56. The method of one or more of clauses 50 to 55, wherein the basic signal object is a predefined time sequence and / or time grouping of a number of other signal objects.

[0233] 57. The method according to one or more of clauses 33 to 56, wherein the transmission of the symbol of at least a recognised signal object class and at least one assigned signal object parameter of this recognised signal object class is a transmission of a signal object class of a signal object that is a predefined time sequence of a plurality of other signal objects, and at least one signal object class of at least one of these other signal objects is not transmitted.

[0234] 58. The data transmission is performed via a bidirectional single-wire data bus, wherein: the sensor transmits the data to the computer system in a current modulation manner; Item 58. The method according to one or more of items 31 to 57, wherein the computer system transmits the data to the sensor in a voltage modulation manner.

[0235] 59. The method according to claim 58, characterized in that a PSI5 data bus and / or a DSI3 data bus is used for data transmission.

[0236] 60. A method according to one or more of clauses 58 or 59, wherein the transmission of data to the computer system is performed at a transmission rate of more than 200 kBit / s and the transmission of data from the computer system to at least one of the sensors is performed at a transmission rate of more than 10 kBit / s, preferably more than 20 kBit / s.

[0237] 61. A method according to one or more of clauses 58, 59 or 60, wherein for transmitting the date from the sensor to the computer system, a transmission current is modulated on a data bus, and the current intensity of the transmission current is less than 50 mA, preferably less than 5 mA.

[0238] 62. A method for transmitting sensor data from a sensor, particularly in an ultrasonic sensor, to a computer system, particularly in a vehicle, comprising or in conjunction with the following steps: Emitting an ultrasonic burst. Receiving an ultrasonic signal and forming a time-discrete received signal formed by a sequence of a plurality of sample values. Here, a temporal date (timestamp) is assigned to each sample value. Determining a first parameter signal of a first characteristic from the sequence of sample values ​​of the received signal using a first filter. wherein the parameter signal is configured as a time-discrete sequence of a plurality of parameter signal values; Here, exactly one temporal date (timestamp) is assigned to each parameter signal value. Determining at least one further parameter signal of a characteristic assigned to said further parameter signal from a sequence of a plurality of sample values ​​of said received signal using a further filter assigned to said further parameter signal. wherein the further plurality of parameter signals are configured as a time-discrete sequence of a further plurality of parameter signal values, respectively; Here, the same time stamp assigned to the corresponding parameter signal value is also assigned to each further parameter signal value, respectively. Hereinafter, the first parameter signal and the further plurality of parameter signals will be collectively referred to as a feature vector signal. Here, the feature vector signal is constructed as a time-discrete sequence of parameter vector signal values, hereinafter also referred to as feature vector signal values, formed by a parameter signal value and further parameter signal values ​​each having the same temporal date (timestamp). Here, this respective time stamp is assigned to each feature vector signal value thus formed. Comparing the temporal date feature vector signal values ​​to a threshold vector to form a Boolean result that can take first and second values. Transmitting a plurality of feature vector signal values ​​and a temporal date (timestamp) assigned to the feature vector signal values ​​from the sensor to the computer system if a Boolean result for the temporal date (timestamp) has a first value.

[0239] 61. A method for transmitting sensor data from a sensor, particularly in an ultrasonic sensor, to a computer system, particularly in a vehicle, comprising or in conjunction with the following steps: Emitting an ultrasonic burst. Receiving an ultrasonic signal and forming a time-discrete received signal formed by a sequence of a plurality of sample values. Here, a temporal date (timestamp) is assigned to each sample value. A step of determining at least two parameter signals using at least one filter from the sequence of a plurality of sample values ​​of the received signal, each referring to the presence of one respective elementary signal object assigned to the respective parameter signal. wherein the parameter signals are configured as a time-discrete sequence of a respective plurality of parameter signal values; Here, exactly one temporal date is assigned to each parameter signal value. Hereinafter, the first parameter signal and the further plurality of parameter signals are collectively referred to as a feature vector signal. Here, said feature vector signal is constructed as a time-discrete sequence of feature vector signal values ​​formed by a plurality of parameter signal values ​​and further parameter signal values ​​each having the same temporal date (time stamp). Here, this respective time stamp is assigned to each feature vector signal value thus formed. Evaluating a temporal profile of said feature vector signal and determining an evaluation value (distance) in the temporal profile while concluding with respect to a signal object formed by a temporal sequence of a plurality of basic signal objects and to which a symbol is assigned. Comparing the timestamp evaluation value to a threshold vector to form a Boolean result that can take first and second values. Transmitting a symbol of the signal object and a temporal date (timestamp) assigned to the symbol from the sensor to the computer system if the Boolean result for the temporal date has a first value.

[0240] 64. A sensor, in particular an ultrasonic sensor, adapted or provided for carrying out the method according to one or more of paragraphs 32 to 63.

[0241] 65. A computer system adapted or provided for carrying out the method according to one or more of paragraphs 32 to 63.

[0242] 66. At least one computer system according to paragraph 63, At least two sensors according to item 65; Equipped with A sensor system, the sensor system being provided to ensure that data transmission between the sensor and the computer system is performed or can be performed according to the method described in one or more of paragraphs 32 to 63.

[0243] 67. In the sensor, each one of the ultrasonic reception signals, i.e. at least two ultrasonic reception signals, is compressed and transmitted to the computer system using a method corresponding to the method described in one or more of paragraphs 32 to 63; Item 45. The sensor system of item 44, wherein in the computer system, at least two ultrasonic receive signals are reconstructed into a plurality of reconstructed ultrasonic receive signals.

[0244] 68. The sensor system of claim 67, wherein the computer system uses the reconstructed ultrasonic received signal to perform object recognition of objects in the environment of the sensor.

[0245] 69. A sensor system as described in clause 68, wherein the computer system performs object recognition of objects in the environment of the sensor using the reconstructed ultrasonic received signals and additional sensor signals of further sensors, in particular signals from multiple radar sensors.

[0246] 70. A sensor system according to one or more of paragraphs 68 or 69, wherein the computer system creates an environmental map of the sensor or of the device of which the sensor is a part, based on the recognized objects. [Explanation of symbols]

[0247] Emission of alpha ultrasonic bursts β Receiving ultrasonic bursts reflected by an object and converting them into electrical received signals γ Compression of the electrical receive signal and formation of a sampled electrical receive signal. Preferably, a timestamp can be assigned to each sample value of said electrical receive signal. Determination of a plurality of spectral values, e.g. by a plurality of matched filters for a plurality of signal object classes, which together form a feature vector, preferably performed continuously such that a stream of feature vector values ​​is obtained. Preferably, a timestamp value can be reassigned to each feature vector value. γc is an arbitrary bar in the form of a predetermined plurality of feature vector values ​​of a prototype library, where normalization of the feature vector spectral coefficients of each feature vector of the timestamp values ​​is preferably performed prior to correlation with the plurality of signal object classes. γd - Determination of the distance between the current feature vector value and a plurality of values ​​of a plurality of signal object classes in the form of a plurality of predetermined feature vector values ​​of a prototype library Selection of the most similar signal object class, in the form of predefined feature vector values ​​of the γe (prototype) library, preferably with the smallest distance to the current feature vector, and accepting as recognized signal object the symbols of this signal object class together with their time stamp values ​​as compressed data. If necessary, further data can be accepted as compressed data, in particular several signal object parameters, such as for example their amplitude. These compressed data form the compression of said received signal. δ Transmission of the compressed electrical received signal to a computer system 1. Envelope of the received ultrasonic signal 2. Output signal (transmitted information) of IO interface according to conventional technology 3. Information transmitted by the LIN interface according to the prior art 4. The first intersection point in the downward direction between the envelope 1 of the received signal and the threshold signal SW 5. The first intersection point in the ascending direction between envelope 1 and threshold signal SW 6 First maximum of envelope 1 above threshold signal SW 7. The second intersection point between the envelope 1 and the threshold signal SW in the downward direction 8. The second intersection point in the ascending direction between envelope 1 and threshold signal SW 9 Second maximum of envelope 1 above threshold signal SW 10 First minimum of envelope 1 above threshold signal SW 11 Third maximum of envelope 1 above threshold signal SW 12 The third intersection point in the downward direction between envelope 1 and threshold signal SW 13 The third intersection point in the ascending direction between envelope 1 and threshold signal SW 14 Fourth maximum of envelope 1 above threshold signal SW 15 The fourth intersection point in the downward direction between envelope 1 and threshold signal SW 16 The fourth intersection point in the ascending direction between envelope 1 and threshold signal SW 17 Fifth maximum of envelope 1 above threshold signal SW 18 Fifth intersection point in the downward direction between envelope 1 and threshold signal SW 19 The fifth intersection point in the ascending direction between envelope 1 and threshold signal SW 20 6th maximum of envelope 1 above threshold signal SW 21 The sixth intersection point in the downward direction between the envelope 1 and the threshold signal SW 22 The sixth intersection point in the ascending direction between envelope 1 and threshold signal SW 23 7th maximum of envelope 1 above threshold signal SW 24 Seventh intersection point in the downward direction between envelope 1 and threshold signal SW 25 Envelope during ultrasonic bursts 26. Transmission of the data of the first intersection 4 of the envelope 1 with the threshold signal SW, preferably in the downward direction via a bidirectional data bus. 27 Transmission of the data of the first intersection point 5 of the envelope 1 with the threshold signal SW, preferably in the ascending direction via a bidirectional data bus. 28 Transmission of the first up to 6 data of envelope 1 exceeding the threshold signal SW, preferably via a bidirectional data bus 29. Transmission of the data of the second intersection 7 of the envelope 1 with the threshold signal SW, preferably in the downward direction via a bidirectional data bus. 30 Transmission of the data of the second intersection point 8 of the envelope 1 with the threshold signal SW, preferably in the ascending direction via a bidirectional data bus 31 Transmission, preferably via a bidirectional data bus, of the data of the first maximum 9 of the envelope 1 exceeding the threshold signal SW and of the data of the exemplary first minimum 10 of the envelope 1 exceeding the threshold signal SW. 32 Transmission, preferably via a bidirectional data bus, of the data of the third maximum of envelope 1 exceeding the threshold signal SW 34. Transmission of the data of the third intersection 12 of the envelope 1 with the threshold signal SW, preferably in the downward direction via a bidirectional data bus. 35 Transmission of the data of the third intersection 13 of the envelope 1 with the threshold signal SW, preferably in the ascending direction via a bidirectional data bus. 36 Transmission of the fourth up to 14th data of envelope 1 exceeding the threshold signal SW, preferably via a bidirectional data bus 37 Transmission of the data of the fourth crossing point 15 of the envelope 1 with the threshold signal SW, preferably in the downward direction via a bidirectional data bus 38 Transmission of the data of the fourth intersection 16 of the envelope 1 with the threshold signal SW, preferably in the ascending direction via a bidirectional data bus 39 Transmission of the fifth up to the seventeenth data of envelope 1 exceeding the threshold signal SW, preferably via a bidirectional data bus 40 Transmission of the data of the fifth intersection 18 of the envelope 1 with the threshold signal SW, preferably in the downward direction via a bidirectional data bus 41 Transmission of the data of the fifth intersection 19 of the envelope 1 with the threshold signal SW, preferably in the ascending direction via a bidirectional data bus 42 Transmission of the data of the sixth intersection 21 of the envelope 1 with the threshold signal SW, preferably in the downward direction via a bidirectional data bus 43 Transmission of received echo data on the LIN bus after reception is completed using conventional technology 44 Transmitting data on a LIN bus using conventional technology before emitting an ultrasonic burst 45 Transmitting data through the IO interface prior to emitting an ultrasonic burst 46 Effect of ultrasonic transmission bursts on the output signal of an IO interface according to the prior art 47 Signals of the first echo 5, 6, 7 at the IO interface according to the prior art 48 Second echo 8, 9, 10, 11, 12 signals at IO interface according to prior art 49 Signals of the third and fourth echoes 13, 14, 15 at an IO interface according to the prior art 50 Signals of the 5th echo 16, 17, 18 at the IO interface according to the prior art 51 Signals of the sixth echo 19, 20, 21 at the IO interface according to the prior art 52 Signals of the sixth echo 22, 23, 24 at the IO interface according to the prior art 53 Start command via data bus from computer system to sensor 54 Automatic regular data transmission between the sensor and the computer system, preferably according to the DSI3 standard; Diagnostic bit after 55 measurement cycles 56 End of transmission of ultrasonic burst (end of transmission burst). Preferably, the end of the ultrasonic burst coincides with point 4. 57 Start of ultrasonic burst emission (start of transmit burst) 58 End of data transmission 59 Triangle Objects 60 Saddle Points 61 Triangle Signal 62 peaks a) Transmission information for transmitting received ultrasound echoes by a prior art IO interface A First Triangle Object au "arbitrary unit", i.e. a freely chosen unit B Transmission information for transmitting received ultrasonic echoes by a LIN interface of the prior art c) Transmission information for the transmission of received ultrasound echoes by the proposed method and the proposed device with an envelope for comparison. d) Transmission information for transmitting received ultrasonic echoes by the proposed method and the proposed device without an envelope e. Schematic signal waveform of receiving echo information transmitted by the IO interface of the prior art En is the amplitude of the envelope of the received ultrasonic signal. f Schematic signal waveform of the transmission of received echo information by the conventional LIN interface g. Schematic signal waveform for transmitting received echo information via bidirectional data interface SB Transmit Burst SW Threshold t time T E Reception time. The reception time usually starts at the end 56 of the emission of the ultrasonic burst. It is also possible to start reception already earlier. However, this may cause problems that require additional measures.

Claims

1. 1. A method for transmitting data representative of a received ultrasonic signal reflected from an obstacle from an ultrasonic sensor to a computer system, comprising: forming a feature vector signal from the electrical receive signals representative of the ultrasonic receive signals, the feature vector signal being non-obstacle detection; recognizing a profile for detecting an obstacle from the electrical reception signal as a signal object, and classifying the recognized signal object into a signal object class for classifying signal objects; Equipped with Classification of said signal objects into said signal object classes, or by assigning to each recognised signal object signal object parameters describing or characterising said signal object and a symbol which is an identifier of the signal object class to which this signal object belongs; or For each recognized signal object, at least one signal object parameter assigned to said signal object and a symbol for this signal object are determined. transmitting at least a symbol of the signal object class to which the recognized signal object belongs and at least one signal object parameter of the signal object to the computer system as data of the recognized signal object. The method further comprises:

2. 2. The method according to claim 1, characterized in that the transmission of the symbol of the signal object class and of the at least one signal object parameter assigned to the recognized signal object is performed with priority.

3. 3. The method according to claim 1 or 2, characterized in that at least one signal object parameter assigned to the recognized signal object is a time value representing the time period since the emission of a preceding ultrasonic burst.

4. The method of claim 3 , further comprising the step of determining a spatial distance of an obstacle represented by the recognized signal object based on the time value.

5. determining a chirp value as a signal object parameter assigned to the recognised signal object; 5. A method according to any one of claims 1 to 4, which indicates whether the recognised signal object is an echo of an ultrasonic transmit burst having a chirp up function, a chirp down function or a no chirp function.

6. 6. The method of claim 1, further comprising the step of generating a confidence signal by forming a correlation between the electrical received signal or a signal derived therefrom and a reference signal.

7. The method of claim 6 , further comprising the step of generating a phase signal indicative of a phase shift of the electrical received signal or the trust signal.

8. The method of claim 7, further comprising the step of generating a phase confidence signal by forming a correlation between the phase signal or a signal derived therefrom and the reference signal.

9. The method of claim 8 , further comprising comparing the phase confidence signal to one or more thresholds to generate a discrete phase confidence signal.

10. forming at least one binary, digital, or analog distance value between said feature vector signal and one or more signal object prototype values ​​indicative of a predetermined threshold for a given said signal object class; assigning the signal object class to the recognized signal object if the magnitude of the distance value undershoots one or more predetermined binary, digital, or analog distance values. The method of any one of claims 6 to 9, further comprising:

11. 11. A method according to any one of claims 1 to 10, characterized in that at least one signal object class is a wavelet.

12. 12. The method of claim 11, wherein the wavelet is a triangular wavelet.

13. 12. The method of claim 11, wherein the wavelet is a rectangular wavelet.

14. 12. The method of claim 11, wherein the wavelet is a half-sine wavelet.

15. 15. A method according to any one of claims 11 to 14, characterized in that one of the signal object parameters is a time shift of the wavelet of the recognised signal object, a time compression or expansion of the wavelet of the recognised signal object, or an amplitude of the wavelet of the recognised signal object.

16. 16. The method according to claim 1, characterized in that the transmission of the symbols of the signal object class and of the at least one assigned signal object parameter is performed according to the FIFO principle, which means that recognized signal object classes with an earlier time stamp are transmitted earlier than recognized signal object classes with a later time stamp.

17. 17. The method according to any one of claims 1 to 16, characterized in that the transmission of an error state of the ultrasonic sensor is performed with a higher priority than the transmission of data of the recognized signal object.

18. 18. The method according to any one of claims 1 to 17, characterized in that the signal objects comprise combinations of at least two, three, four or more basic signal objects indicating specific events and / or their temporal sequences.

19. 20. The method according to claim 18, wherein the basic signal object is an intersection of the ultrasonic reception signal and a threshold signal SW in a graph showing a profile of the ultrasonic reception signal.

20. The method according to claim 18, characterized in that the basic signal object is an intersection point between an ascending graph line of the ultrasonic reception signal and a threshold signal SW in a graph showing a profile of the ultrasonic reception signal.

21. The method according to claim 18, characterized in that the basic signal object is an intersection point between a descending graph line of the ultrasonic reception signal and a threshold signal SW in a graph showing a profile of the ultrasonic reception signal.

22. 19. The method of claim 18, wherein the basic signal object is a profile of the ultrasonic reception signal or a local or absolute maximum of the profile of the ultrasonic reception signal, which maximum exceeds a threshold signal SW or exceeds the magnitude of the threshold signal SW.

23. 19. The method of claim 18, wherein the basic signal object is a profile of the ultrasonic reception signal or a local or absolute minimum of the profile of the ultrasonic reception signal, which minimum exceeds a threshold signal SW or exceeds the magnitude of a threshold signal SW.

24. 24. A method according to any one of claims 18 to 23, characterized in that the basic signal object is a predefined temporal sequence and / or temporal grouping of a number of other signal objects.

25. said transmission of at least a symbol of said recognized signal object class and at least one assigned signal object parameter of said recognized signal object class being a predefined time sequence of a signal object; 25. The method according to any one of claims 1 to 24, characterized in that the signal object class of the other signal object is not transmitted.

26. 26. The method according to claim 1, wherein the data transmission is performed via a bidirectional single-wire data bus, the ultrasonic sensor transmitting the data to the computer system in a current modulation manner, and the computer system transmitting the data to the ultrasonic sensor in a voltage modulation manner.

27. 27. Method according to claim 26, characterized by the use of a PSI5 data bus and / or a DSI3 data bus for data transmission.

28. 28. The method according to claim 26 or 27, characterized in that the transmission of the data to the computer system is performed at a transmission rate of more than 200 kBit / s and the transmission of data from the computer system to the at least one ultrasonic sensor is performed at a transmission rate of more than 10 kBit / s.

29. 30. The method of claim 28, wherein said transmission rate from said computer system to said at least one ultrasonic sensor is greater than 20 kBit / s.

30. 29. The method according to any one of claims 26 to 28, characterized in that for the transmission of data from the ultrasonic sensor to the computer system, a transmission current is modulated on the data bus, the current strength of the transmission current being less than 50 mA.

31. 31. The method of claim 30, wherein the current strength of the transmit current is less than 5 mA.

32. 32. The method of any one of claims 1 to 31, carried out in a vehicle.

33. 31. An ultrasonic sensor for carrying out the method for transmitting data according to any one of claims 1 to 30, comprising: a signal processing unit for analyzing and compressing the received ultrasonic signal to minimize the amount of data before transmission; The analysis and compression of the ultrasonic reception signal by the signal processing unit: forming a feature vector signal from the electrical receive signal representative of the ultrasonic receive signal, or forming a time-discrete receive signal from a sequence of a plurality of sample values; 2. An ultrasonic sensor comprising:

34. An ultrasonic sensor according to claim 33; A computer that receives the data transmitted by the ultrasonic sensor and performs analysis to recognize an obstacle and / or the distance of the obstacle; A sensor system comprising:

35. At least one of the above computers; At least two of the ultrasonic sensors; 35. The sensor system of claim 34, comprising:

36. Transmitting the ultrasonic reception signals received by each of the at least two ultrasonic sensors to the computer system as compressed signals; 36. The sensor system of claim 35, wherein said computer system decompresses each transmitted compressed signal into a plurality of ultrasonic receive signals.

37. 37. The sensor system of claim 36, wherein the computer system is further configured to perform recognition of a plurality of obstacles in an environment of the ultrasonic sensor using a plurality of ultrasonic reception signals extracted from reception data of the ultrasonic sensor.

38. 40. The sensor system of claim 37, wherein the computer system is configured to perform recognition of obstacles in an environment of the plurality of ultrasonic sensors using a plurality of ultrasonic reception signals of a further plurality of ultrasonic sensors.

39. 40. The sensor system of claim 38, wherein the computer system is configured to create an environment map of the ultrasonic sensors, or a device of which they are a part or which includes the ultrasonic sensors, based on the recognized obstacles.

Images