Shape estimation system, object recognition system, communication system, shape estimation method, object recognition method and communication method

Abstract

The present invention stores as recognition data the distance between a vehicle 1 and a detected object 3, which is recorded on the basis of time, from the time at which a transmitting wave 2A is sent until the time at which a receiving wave 2B scattered / reflected by the detected object 3 is received, as well as an amplitude / intensity value of the receiving wave 2B, which corresponds to the distance; samples the amplitude / intensity value of the receiving wave 2B, which corresponds to the distance between the vehicle 1 and the detected object 3, as sampling data for a desired distance; converts the re-sampling data from the spatial domain to the spatial frequency domain.The present invention stores a set of spectra in the spatial frequency domain for a multitude of preset object shapes as a database in conjunction with the object shape and estimates the shape of the detected object 3 based on the spectrum of the detected object 3 and the stored set of spectra. Furthermore, the present invention estimates the shape of the detected object 3 to be the object shape that corresponds to the spectrum under the set of spectra that most closely approximates the spectrum of the detected object 3.

Description

Technical field

[0001] The present invention relates to a shape estimation system, an object recognition system and a communication system for a vehicle, which recognize objects present in the environment of the vehicle and estimate the shapes of the objects, as well as a shape estimation method, an object recognition method and a communication method. State of the art

[0002] For example, PTL 1 discloses a technique for estimating whether objects in the vicinity of a vehicle are people. PTL 1 discloses: “The human detection device according to the embodiment acquires detection information from an ultrasonic sensor that has received a reflected wave of an ultrasonic wave that was transmitted into a predetermined area. The waveform of the reflected wave, according to the detection information, is normalized based on the distance of the ultrasonic wave to the reflecting object and on distance-dependent attenuation characteristic information relating to the ultrasonic wave. In a case where the peak of the normalized waveform of the reflected wave is greater than or equal to a predetermined threshold, it is determined that an object is detected in a predetermined area.”In a case where an object is detected by an object recognition unit, a waveform of the reflected wave from a predetermined area containing the peak is extracted from the normalized waveform of the reflected wave. A multitude of parameters is generated by performing a predetermined frequency analysis on the waveform of the reflected wave from the predetermined area without losing any temporal information. A parameter for distinguishing whether an object is a human is extracted from the multitude of parameters. A distinction as to whether the object is a human is then made based on this extracted parameter. Protective documents, patent application

[0003] PTL 1: JP 2022-142252 A Summary of the invention: Technical problem

[0004] In a vehicle control system, object detection technology is crucial for safely and appropriately steering the vehicle to a desired position. It is also desirable not only to detect the presence or absence of objects in the vehicle's vicinity, but also to be able to estimate the shapes of those detected objects.

[0005] For example, in a vehicle control system, such as is typically used in automatic parking and parking assistance, it is possible to switch a vehicle control procedure by recognizing the shape of an object, e.g., whether the object is a vehicle or a curb / wheel stopper.

[0006] Incidentally, in a shape estimation system for a vehicle control system, it is desirable to be able to estimate the type of the detected object with high accuracy, and it is practical to be able to estimate not only the type of object but also its dimensions.

[0007] In the case of the system / device / technology disclosed in PTL 1 and the like, however, a type estimation offers the possibility of improving accuracy based on a theoretical calculation result and an actual measurement result. Furthermore, no technique for performing a dimensional estimation is disclosed.

[0008] The present invention was conceived with regard to the problems mentioned above, and the purpose of the present invention is to provide a shape estimation system, an object recognition system and a communication system that makes it possible to estimate the types and dimensions of objects in the environment of a self-propelled vehicle with high accuracy, as well as a shape estimation method, an object recognition method and a communication method. Solution to the problem

[0009] The present invention comprises several means for solving the above-mentioned problem, but one example is a shape estimation system for a vehicle, wherein the system comprises: a storage unit which stores as recognition data the distance between the vehicle and a detected object as well as an amplitude / intensity value of a wave corresponding to the distance, wherein the distance is detected on the basis of time, from the time at which a transmitting wave is sent by a transmitting unit which sends the transmitting wave to a detection area to be observed, until the time at which the wave is received by a receiving unit which receives the wave scattered / reflected by the detected object present in the detection area;a re-sampling unit that, based on the detection data, re-samples an amplitude / intensity value of the wave corresponding to the distance between the vehicle and the detected object as sample data for a desired distance; a conversion processing unit that converts the re-sampling data output by the re-sampling unit from a spatial domain to a spatial frequency domain; a storage unit that stores a set of spectra in the spatial frequency domain for a variety of preset object shapes as a database in conjunction with the object shape;and a shape estimator that estimates the shape of the recognized object based on the spectrum of the recognized object output by the conversion processing unit and the set of spectra stored in the memory unit, wherein the shape estimator estimates as the shape of the recognized object the object shape that corresponds to the spectrum among the set of spectra in the memory unit that is closest to the spectrum of the recognized object output by the conversion processing unit. Advantageous effects of the invention

[0010] The present invention makes it possible to estimate the types and dimensions of objects located in the vicinity of a private vehicle with high accuracy. Problems, configurations, and effects not described above are clarified by the following description of embodiments. Brief description of the drawings [ Fig. 1] Fig. Figure 1 shows a block diagram illustrating an example of an overall configuration of a vehicle equipped with a shape estimation system according to a first embodiment of the present invention. [ Fig. 2] Fig. Figure 2 shows a schematic diagram illustrating an example of the spatial arrangement of a vehicle, transmit / receive waves and an object at the time of object detection. [ Fig. 3] Fig. Figure 3 shows a diagram illustrating examples of waveforms of a recognition signal in a vehicle at the time of object recognition and of signals processed in the shape estimation system. [ Fig. 4] Fig. Figure 4 shows an explanatory diagram illustrating one embodiment of object type estimation performed by the shape estimation system. [ Fig. 5] Fig. Figure 5 shows an explanatory diagram illustrating one embodiment of an object dimension estimation performed by the shape estimation system. [ Fig. 6] Fig. Figure 6 shows an example of the processing sequence by the shape estimation system according to the first embodiment of the present invention. [ Fig. 7A] Fig. Figure 7A shows a diagram illustrating an example of a waveform when the technical validity of the shape estimation system according to the first embodiment of the present invention is verified by simulation. [ Fig. 7B] Fig. Figure 7B shows a diagram illustrating an example of a waveform when the technical validity of the shape estimation system according to the first embodiment of the present invention is verified by simulation. [ Fig. 7C] Fig. Figure 7C shows a diagram illustrating an example of a waveform when the technical validity of the shape estimation system according to the first embodiment of the present invention is verified by simulation. [ Fig. 7D] Fig. Figure 7D shows a diagram illustrating an example of a waveform when the technical validity of the shape estimation system according to the first embodiment of the present invention is verified by simulation. [ Fig. 8] Fig. Figure 8 shows a diagram illustrating an example of an estimation result when the technical validity of the shape estimation system according to the first embodiment of the present invention is verified by simulation. [ Fig. 9] Fig. Figure 9 shows a diagram illustrating an example of a relationship between an error rate and a noise quantity when the technical validity of the shape estimation system according to the first embodiment of the present invention is verified by simulation. [ Fig. 10] Fig. Figure 10 shows a block diagram illustrating an example of the overall configuration of a vehicle equipped with a shape estimation system and a communication system according to a second embodiment of the present invention. Description of the embodiments

[0011] The following describes embodiments of a shape estimation system, an object recognition system, and a communication system, as well as a shape estimation method, an object recognition method, and a communication method of the present invention with reference to the drawings. It should be noted that the present invention is not limited to the embodiments described below and various modifications can be made within the scope of the technical concept. In the drawings used in this description, identical or corresponding components are designated with identical or similar reference numerals, and a repeated description of these components may be omitted. <Erste Ausführungsform>

[0012] A first embodiment of a shape estimation system, an object recognition system and a communication system, as well as a shape estimation method, an object recognition method and a communication method of the present invention, is described with reference to the Fig. 1 to 9 described.

[0013] First, an overall configuration of a vehicle equipped with the shape estimation system is described with reference to Fig. 1 described. Fig. Figure 1 shows a block diagram illustrating an example of the overall configuration of a vehicle equipped with the shape estimation system according to the first embodiment of the present invention.

[0014] As in Fig. As illustrated in Figure 1, a vehicle 1 comprises an object recognition system 100, a vehicle control unit 130, and an HMI (human-machine interface) 140. Although the vehicle 1 contains many functional blocks, only those functional blocks relating to the present invention are shown.

[0015] The object recognition system 100 comprises a transmitter / receiver unit 110 and a shape estimation system 120 and is able to recognize objects present in the vicinity of the vehicle and also to estimate the shapes of the objects.

[0016] The transmit / receive unit 110 comprises a transmitting unit 111, which transmits a transmitting wave 2A (see Fig. 2) sends to a detection area to be observed, a receiving unit 112 which sends a receiving wave 2B (see Fig. 2) receives, which is scattered / reflected by a detected object 3 present in the detection area, and a distance calculation unit 113, which determines the distance between the vehicle 1 and the detected object 3 on the basis of the time from the time at which the transmitting wave 2A is sent by the transmitting unit 111 until the time at which the receiving unit 112 receives the receiving wave 2B, and the transmitting / receiving unit 110 projects waves in a desired area from the transmitting unit 111 to detect objects in the vicinity of the vehicle.

[0017] Here, the transmitting wave 2A, which is sent by the transmitting unit 111, and the receiving wave 2B, which is received by the receiving unit 112, are, for example, ultrasound waves, optical waves, a laser (which is considered particularly coherent among optical waves), radio waves, or the like. In the present invention, future descriptions will be given by way of example using ultrasound waves.

[0018] In a case where an object is present within the detection range, the receiving unit 112 receives a scattered / reflected wave from the object. The distance calculation unit 113 then receives an output signal from the receiving unit 112 and calculates the distance between the vehicle 1 and the object based on the time from when the wave is transmitted by the transmitting unit until when it is received by the receiving unit. This time can be considered the time required for the wave to travel back and forth between the vehicle 1 and the object.

[0019] The shape estimation system 120 comprises a storage unit 121, a storage unit 122, a rescanning unit 123, a conversion processing unit 124 and a comparison calculation unit 125.

[0020] The storage unit 121 is a storage medium that acts as a database, storing a set of spectra in a spatial frequency range for a multitude of predefined object shapes. This set of spectra, in conjunction with the object shapes, also stores information about the set of spectra in the spatial frequency range of scattered / reflected waves corresponding to the multitude of objects. This set of spectra can be generated, for example, based on a simulation or on actual measurement data.

[0021] The storage unit 122 is a recording medium that stores as recognition data the distance between the vehicle 1 and the detected object 3 and an amplitude / intensity value of the received wave 2B corresponding to the distance, the distance being recorded on the basis of time, from the time at which the transmitting wave 2A is sent by the transmitting unit 111, which sends the transmitting wave 2A to the detection area to be observed, until the time at which the received wave 2B, which is scattered / reflected by the detected object 3 present in the detection area, is received by the receiving unit 112, which receives the received wave 2B.

[0022] The re-sampling unit 123 is a component that re-samples the amplitude / intensity value of the received wave 2B, corresponding to the distance between the vehicle 1 and the detected object 3, as sample data for a desired distance based on the detection data recorded in the storage unit 122. This is necessary because, when the driver or the system controls the vehicle, it is assumed that the vehicle will move forward or backward repeatedly depending on the situation, and there is not necessarily data available that would allow the shape estimation system 120 to easily estimate the shape of the object. Therefore, processing is required to obtain the desired data.

[0023] As an example of resample processing, it is conceivable to perform operational processing to calculate amplitude / intensity value data when the distance between the vehicle and the object changes at regular intervals.

[0024] The conversion processing unit 124 is a component that converts the re-sampling data output by the re-sample unit 123 from the spatial domain to the spatial frequency domain. For example, the amplitude / intensity value newly sampled by the re-sample unit 123 is input, and the conversion processing unit 124 converts the data in the spatial domain, for example, into data in the spatial frequency domain. It should be noted that the processing for conversion to the spatial frequency domain has been described in this description as an example; however, the present invention is not limited to this conversion processing, and it is considered possible to use various conversion methods.

[0025] The comparison calculation unit 125 is a part comprising a type estimation unit 126 and a dimension estimation unit 127, and which estimates the shape of the recognized object 3 based on the spectrum of the recognized object 3 output by the conversion processing unit 124 and the set of spectra stored in the storage unit 121. In the present embodiment, the comparison calculation unit 125 estimates as the shape of the recognized object 3 the object shape that corresponds to the spectrum among the set of spectra in the storage unit 121 that most closely matches the spectrum of the recognized object 3 output by the conversion processing unit 124.

[0026] In particular, the type estimator 126 estimates the type of the detected object 3 as the shape estimate of the detected object 3. Here, for example, the type estimator 126 is able to perform machine learning of the features of each type by using as training data data from the set of spectra prepared as a database in the storage unit 121 and corresponding object shape information, and estimating the type from the measured spectrum of the detected object 3 based on the result of the machine learning. Alternatively, the type estimator 126 can, for each object type, calculate the average value of the difference between the set of spectra prepared as a database in the storage unit 121 and the measured spectrum, and estimate the type for which the average value of this difference is substantially minimal as the type of object to be observed.

[0027] In particular, the dimension estimation unit 127 estimates the dimensions of the detected object 3 as the shape estimate of the detected object 3. For example, a method using scatterometry can be considered to perform dimension estimation by comparing a pre-existing database with actual measurement data. However, in normal scatterometry, the approximate shape of an object is known, or the measurement technique is the same each time, and therefore dimension estimation is relatively straightforward. In the case of a vehicle, however, the approach method, such as the speed at which the object approaches or the number of repetitions of the forward / backward movement, is different each time, and therefore it is not easy to apply this technique.

[0028] Furthermore, various scenarios are conceivable, such as when the observation target is a wall, a vehicle, a post, or a curb, and therefore it is not straightforward to apply normal scatterometry. Therefore, as described above, it is conceivable that the rescanning unit 123 performs rescan processing to obtain a data format suitable for shape estimation, regardless of the specific case when using the vehicle approach method.

[0029] Furthermore, in the comparison calculation unit 125, for example, the type estimation unit 126 narrows down the types in advance, and then the dimension estimation unit 127 performs a dimension estimation, whereby the dimension estimation of an object to be observed can be carried out by applying scatterometry.

[0030] As an example of processing by the dimension estimation unit 127, a procedure can be considered that involves extracting a set of spectra prepared as a database in the storage unit 121 only for the type estimated by the type estimation unit 126, calculating the difference between the actually extracted set of spectra and the measured spectrum, and estimating dimensions for which the difference is substantially minimal as the dimensions of the object to be observed. Using the above procedure, it is also possible to apply this technique to a vehicle where there is a high degree of operational freedom, the object to be observed is also unknown, and it is extremely difficult to apply scatterometry; the technique makes it possible to perform an estimation of the shape of the object to be observed.

[0031] For example, the vehicle control unit 130 inputs shape information of the object to be observed from the comparison calculation unit 125 and changes the procedure for controlling the vehicle according to the type and dimensions of the object.

[0032] For example, the HMI 140 inputs shape information of the object to be observed from the comparison calculation unit 125 and notifies the driver about the object information.

[0033] Fig. Figure 2 shows an example of the spatial arrangement of a vehicle, transmit / receive waves, and an object at the time of object detection. The transmit wave 2A emitted by vehicle 1 is scattered / reflected by the detected object 3 and received by vehicle 1 as a receive wave 2B. At the time of object detection, it is conceivable, for example, that vehicle 1 is approaching the detected object 3 from the positive direction in the drawing in the direction of the origin O, and it is possible to estimate the shape of a target object that the vehicle is approaching at this time.

[0034] Fig. Figure 3 shows examples of waveforms of a recognition signal in a vehicle at the time of object recognition and of signals processed in the shape estimation system.

[0035] (a) in Fig. Figure 3 illustrates an example of a change in the distance between the vehicle and the object over time. In the example shown in the drawing, the vehicle first approaches the object, then moves away from it, and then approaches it again, which is considered a case where the vehicle is being driven to a target parking space in a parking lot.

[0036] (am Fig. Figure 3 illustrates an example of the temporal change in the detection level of a wave scattered / reflected by an object. The level tends to increase as the wave approaches the object. Note that this example illustrates a 14-bit detection range, and the detection level is saturated at 16384.

[0037] (c) in Fig. Figure 3 illustrates an example where the distance between the vehicle and the object is plotted on the horizontal axis and the detection level of the wave scattered / reflected by the object is plotted on the vertical axis. It can be seen that specific noise is generated near a distance of 1080 mm. Furthermore, overall random noise is generated, and this noise affects the shape estimation, necessitating countermeasures. Since the plotted data is also not within a desired distance interval, it is desirable to perform the resample processing described above.

[0038] Therefore, the re-scanning unit 123 is able to calculate the scan data from among the data contained in the recognition data by calculating the average value of the amplitude / intensity values ​​of the received wave 2B that falls within a predetermined distance range from the desired distance.

[0039] For example, (d) illustrates in Fig. 3 An example of the relationship between the distance and the detection level after resample processing. During resample processing, for example, as described above, the detection level value is subjected to resample processing to obtain sample data at equal intervals. When calculating the detection level value at a desired distance, for example, a method may be considered that involves extracting detection level values ​​related to a predetermined distance range from the desired distance and calculating the average value. As a result, the effects of random noise and specific noise can be reduced, as in (d) of Fig. 3 illustrated.

[0040] (a Fig. Figure 3 illustrates an example of the relationship between distance and detection level after preprocessing in transformation processing. For instance, in a case where discrete Fourier transform is used as transformation processing, periodic boundary conditions are assumed. Therefore, it is desirable for the detection level value at minimum distance and the detection level value at maximum distance to be essentially the same. To meet this requirement, a window function or similar is generally applied, but this intentionally distorts the original detection waveform, resulting in a reduction in the accuracy of the object's shape estimation.

[0041] Therefore, after combining the re-sampling data output by the re-sampling unit 123 and the data obtained by inverting the sign of the spacing of the re-sampling data, the conversion processing unit 124 can perform a conversion processing from a spatial area to a spatial frequency area.

[0042] For example, as in (e) of Fig. Figure 3 illustrates a possible approach to preprocessing data to fit periodic boundary conditions by combining data obtained by inverting the sign of the distance. Here, the data obtained by inverting the sign of the distance correspond to data captured in the range where the distance in Fig. 2 is negative, and can, for example, be considered as captured data that are captured when the same detected object 3 is detected by the transmitter / receiver unit provided at the rear of the vehicle 1. That is to say, the data in (e) of Fig. 3 can be considered as data for which it is assumed that data with the same detection level value as data captured in a positive distance area can also be captured in a negative distance area.

[0043] (f) in Fig. Figure 3 illustrates an example of a spectrum when the data of the in (e) of Fig. 3 illustrated detection levels of the discrete Fourier transform.

[0044] Fig. Figure 4 shows an embodiment of object type estimation performed by the shape estimation system. In performing type estimation, the shape estimation system estimates, for example, the type of object to be observed by comparing the measured spectrum with a set of spectra that has been prepared in advance as a database.

[0045] A database might contain, for example, a set of spectra in case the dimensions, such as height and thickness, are changed for each type of wall, curb, post, and the like. As described above, the shape estimation system, for example, learns spectral characteristics for each type using information from the set of spectra in this database and uses this training data to estimate the type of object being observed from the measured spectrum.

[0046] Alternatively, the shape estimation system is capable of calculating, for each type, the average value of the difference between the set of spectra stored in the database in storage unit 121 and the spectrum of the detected object 3, which represents the actual measurement data, and estimating the type for which the average value of this difference is minimal as the type of detected object 3. It should be noted that the minimum value may not be obtained exactly and may therefore be a substantially minimal value. Here, the substantially minimal type can be estimated as the type of object to be observed. It should be noted that the set of spectra stored as a database can be created by simulation or using actual measurement data.

[0047] Fig. Figure 5 shows an embodiment of an object dimension estimation performed by the shape estimation system.

[0048] If the dimension estimation is performed by the dimension estimation unit 127 at the time of the shape estimation of the detected object 3, the dimension estimation unit 127 is able to estimate the dimensions of the detected object 3 after referring only to data under the set of spectra relating to the identified type of the detected object 3, since it is assumed that the type of the object to be observed has already been estimated by the type estimation unit 126.

[0049] For example, the dimension estimation unit 127 estimates the dimensions at which the difference between the measured spectrum and the set of spectra stored as a database is substantially minimal, as the dimensions of the detected object to be observed. 3 Alternatively, the dimensions can be estimated using machine learning. It should be noted that the set of spectra stored as a database can be created by simulation or using actual measurement data.

[0050] Fig. Figure 6 shows an example of the processing sequence by the shape estimation system according to the first embodiment of the present invention.

[0051] First, S601 acquires detection data regarding an object to be observed. As described above, the distance between the vehicle and the object, and the amplitude / intensity value of the scattered / reflected wave from the object, are considered detection data.

[0052] The recognition data captured in S602 is then subjected to re-sampling processing. For example, the amplitude / intensity value is re-sampled to ensure that the data is captured at regular intervals.

[0053] The newly sampled amplitude / intensity value is then converted in S603. For example, a spectrum in the local frequency domain is obtained by performing a conversion process from a spatial range to a local frequency domain.

[0054] Next, in S604, the type of object to be observed is estimated. For example, the type of object to be observed is estimated by comparing a pre-prepared set of spectra (as a database) with a measured spectrum.

[0055] S605 then determines whether dimension estimation is necessary. If dimension estimation is not necessary, processing is terminated. If dimension estimation is necessary, processing is terminated after dimension estimation is performed in S606. Regarding the necessity of dimension estimation, for example, in cases where the object being observed is a curb or wheel stop, measures may be considered, such as dimension estimation in cases where it is better to determine whether a collision will occur by comparing the object's dimensions to the vehicle's bumper height.

[0056] For example, in dimension estimation in S606, a pre-prepared set of spectra is extracted for the estimated type only, and the measured spectrum and this set of spectra are compared to estimate the dimensions of the object being observed.

[0057] The Fig. Figures 7A to 7D show examples of waveforms when the technical validity of the shape estimation system according to the first embodiment of the present invention is verified by simulation. The calculations were performed for posts and blocks, and three types of posts with diameters of 40 mm, 80 mm, and 200 mm and three types of blocks with heights of 100 mm, 200 mm, and 300 mm were calculated.

[0058] Fig. Figure 7A shows the relationship between the level value of a scattered / reflected wave calculated by simulation and the distance between a vehicle and an object. As in Fig. As shown in Figure 7A, it can be seen that the closer the distance between the vehicle and the object, the more the level of the scattered / reflected wave generally increases, but since a block is in a blind spot if the distance to the block is too small, the level of the scattered / reflected wave tends to decrease.

[0059] Therefore, the comparison calculation unit 125 is able to estimate the shape of the detected object 3 after normalizing the set of spectra and all spectra of the detected object 3 output by the conversion processing unit 124 using their respective maximum values.

[0060] Fig. 7B shows a set of spectra after the level value of the signal is determined by the signal in Fig. Figure 7A illustrates the simulation of a calculated scattered / reflected wave that has undergone a Fourier transform into the spatial frequency domain. It should be noted that in this drawing, the maximum value is normalized to 1. A verification of principles was performed under the premise that this set of spectra is stored as a database.

[0061] Fig. Figure 7C shows the relationship between the level of the scattered / reflected wave in a measured simulation and the distance between a vehicle and an object. This data is created as data corresponding to the actual measurement by reducing noise to the values ​​provided by the simulation. Fig. Figure 7A illustrates the simulation overlaid with calculated data. It should be noted that during the principle verification, noise immunity was investigated while varying the amount of noise added, and this drawing illustrates an example of this.

[0062] Fig. 7D displays a set of spectra after the level value of the scattered wave / reflected wave in the Fig. 7C illustrated a measured simulation of a Fourier transform into the spatial frequency domain. Whether the shape of the object can be correctly estimated was determined by comparing this data with a pre-defined model. Fig. 7B illustrated the simulation as a database-prepared set of spectra.

[0063] Fig. Figure 8 shows an example of an estimation result when the technical validity of the shape estimation system according to the first embodiment of the present invention is verified by simulation. This result shows an estimation result of the shape estimation system shown in (d). Fig. 7C data shown. As shown Fig. As can be seen in Figure 8, it is evident that all types and dimensions are correctly estimated, and it is evident that the shape of the object to be observed can be estimated by applying the present invention.

[0064] Fig. Figure 9 shows an example of a relationship between the error rate and the noise level when the technical validity of the shape estimation system according to the first embodiment of the present invention is verified by simulation. Here, the noise level is expressed as a percentage, with the peak value for the level of the scattered / reflected wave at a block height of 100 mm being considered as 100%.

[0065] Particularly noteworthy is that the type is correctly estimated without error during type determination, even when a large amount of noise is superimposed, and it can be said that the technology of the present invention makes it possible to estimate the type with very high accuracy. Similarly, in dimensional estimation, if the noise level is less than 20%, the dimensions are estimated correctly and without error, and the dimensions of the object to be observed can be accurately estimated using the technology of the present invention.

[0066] Furthermore, by utilizing the tendency for noise to occur easily on the high-frequency side, it is possible to further improve the accuracy of the dimension estimation by performing filter processing such as a low-pass filter and separating a signal component from a noise component.

[0067] Next, the advantageous effects of the present embodiment will be described.

[0068] The shape estimation system 120 described above for a vehicle 1 according to the first embodiment of the present invention comprises the storage unit 122, which stores as recognition data the distance between the vehicle 1 and the detected object 3 as well as an amplitude / intensity value of the received wave 2B corresponding to the distance, wherein the distance is recorded on the basis of time, from the time at which the transmitting wave 2A is sent by the transmitting unit 111, which sends the transmitting wave 2A to a detection area to be observed, until the time at which the received wave 2B is received by the receiving unit 112, which receives the received wave 2B scattered / reflected by the detected object 3 present in the detection area;the re-sampling unit 123, which, based on the recognition data, re-samples an amplitude / intensity value of the received wave 2B, corresponding to the distance between the vehicle 1 and the recognized object 3, as sampling data of a desired distance; the conversion processing unit 124, which converts the re-sampling data output by the re-sampling unit 123 from a spatial domain into a spatial frequency domain; the storage unit 121, which stores a set of spectra in the spatial frequency domain for a variety of preset object shapes as a database in conjunction with the object shape;and the comparison computing unit 125, which estimates the shape of the recognized object 3 on the basis of the spectrum of the recognized object 3 output by the conversion processing unit 124 and the set of spectra stored in the storage unit 121, wherein the comparison computing unit 125 estimates as the shape of the recognized object 3 the object shape which corresponds to the spectrum under the set of spectra in the storage unit 121 which is closest to the spectrum of the recognized object 3 output by the conversion processing unit 124.

[0069] The present embodiment offers the advantage that, by using not only actual measurement data but also a database created in advance through simulation or the like, the type / dimensions of the object to be observed can be estimated with higher accuracy than when using conventionally disclosed estimation techniques. Furthermore, particularly in type estimation, a principle check demonstrates that the estimation can be performed without errors, even in cases where the noise level is high, and therefore the advantage lies in achieving high noise immunity.

[0070] It should be noted that type estimation and dimension estimation can also be performed using a stereo / multi-camera system, but the technology of the present invention has the advantage of being usable even at night or in bad weather, which is one of the problems of the aforementioned camera technology. Furthermore, improvements in system accuracy can be expected by combining detection information acquired by a stereo / multi-camera system with detection information acquired using the present technology.

[0071] Although lidar sensors are expected to become available in recent years, the technology of the present invention allows the use of a relatively inexpensive ultrasonic sonar or the like and offers the advantage that a system can be built cost-effectively.

[0072] Although a large number of transmitters and receivers are often used in dimensional estimation sensors, the technology of the present invention can be implemented with a single transmitter and receiver and can be implemented by improving the signal processing unit while maintaining the conventionally proposed configuration for the arrangement of transmitters and receivers, which offers the advantages of straightforward introduction and excellent price-performance ratio.

[0073] The technology of the present invention is also advantageous insofar as not only the vehicle control system, but also a distance sonar or the like, which have been widely used until now, are able to provide the driver with additional information about the type / dimensions of an object.

[0074] Furthermore, in conventional scatterometry, the spectral data are acquired by changing the frequency of the wave itself, but the technology of the present invention offers the advantage that the configuration is uncomplicated, since the technology is implemented using spectral data of spatial frequencies by using a wave with a single frequency.

[0075] Additionally, the re-sampling unit 123 calculates the sampling data from the data contained in the recognition data by calculating the average value of the amplitude / intensity values ​​of the received wave 2B that falls within a predetermined distance range from the desired distance, thus making it possible to reduce the effects of random noise and specific noise.

[0076] Furthermore, the conversion processing unit 124 combines the re-sampling data output by the re-sampling unit 123 and the data obtained by inverting the sign of the spacing of the re-sampling data, and then performs a conversion processing from a spatial domain to a spatial frequency domain, making it possible to perform preprocessing of the data suitable for the periodic boundary conditions, and since it is not necessary to intentionally destroy the original recognition waveform in a case where the discrete Fourier transform is performed as conversion processing, the accuracy can be improved.

[0077] Additionally, the comparison calculation unit 125 normalizes the set of spectra and all spectra of the detected object 3 output by the conversion processing unit 124 using the respective maximum values, and then estimates the shape of the detected object 3, making it possible to handle both an increase and a decrease in the level value of the scattered wave / reflected wave and to realize spectral processing of the detected object 3 at different heights.

[0078] Furthermore, by estimating the type of the detected object 3 as a shape estimate of the detected object 3, the comparison calculation unit 125 is able to narrow down the types of targets for which the dimensions are estimated at the time of the subsequent dimension estimation processing. It is therefore possible to increase the processing speed and further reduce the probability of an erroneous determination occurring.

[0079] In addition, the comparison calculation unit 125 is able to estimate the shape of a curb / wheel stopper with high accuracy by estimating at least the curb / wheel stopper as the type of recognized object 3, which was conventionally difficult.

[0080] Furthermore, the comparison computation unit 125 can efficiently improve the accuracy of the type estimation by performing machine learning of the features of each type by using a set of spectra as training data and estimating the type from the spectrum of the detected object 3 based on the result of the machine learning.

[0081] In addition, the comparison calculation unit 125 calculates for each type the average value of the difference between the set of spectra and the spectrum of the detected object 3 and estimates the type for which the average value of this difference is essentially minimal as the type of detected object 3, thereby enabling an improvement in the accuracy of the type estimation.

[0082] Furthermore, the comparison calculation unit 125 can more accurately estimate what type of object the detected object 3 is by estimating the dimensions of the detected object 3 as a shape estimate of the detected object 3.

[0083] Additionally, the comparison calculation unit 125 estimates the type of the detected object 3 at the time of the shape estimation of the detected object 3 and then estimates the dimensions of the detected object 3 after referring only to data under the set of spectra relating to the identified type of the detected object 3, thus making it possible to realize the estimation processing with higher speed and higher accuracy.

[0084] Furthermore, the comparison calculation unit 125 estimates that the dimensions at which the difference between the set of spectra and the spectrum of the detected object 3 is minimized are the dimensions of the detected object 3, thus further improving the accuracy of the dimension estimation. <Zweite Ausführungsform>

[0085] A shape estimation system, an object recognition system and a communication system, as well as a shape estimation method, an object recognition method and a communication method according to a second embodiment of the present invention, are described with reference to Fig. 10 described. Fig. Figure 10 shows a block diagram illustrating an example of the overall configuration of a vehicle equipped with a shape estimation system and a communication system according to a second embodiment of the present invention.

[0086] As in Fig.As illustrated in Figure 10, the present embodiment differs from the first embodiment in that a communication system 4 has been added which can communicate with the storage unit 121 in the vehicle 1.

[0087] The communication system 4 is a system that communicates via a network with the vehicle 1, which includes the object recognition system 100, and which includes, for example, a database update unit 410 that updates a database of the database storage unit 420, a database storage unit 420 that stores a database that is used when the vehicle 1 estimates the shape of the recognized object 3, and a database transfer unit 430 that transfers the database stored in the database storage unit 420 to the storage unit 121 of the vehicle 1.

[0088] The database update unit 410 updates, as needed, a set of spectra in a spatial frequency range relating to various objects for which the vehicle should have a database pre-configured. For example, in a case where this database is created through simulation, it is conceivable to update the database if the simulation accuracy can be improved compared to an initial state, or if errors in the initial database state become problematic.

[0089] The database storage unit 420 inputs the database updated by the database update unit 410 and stores the database.

[0090] The database transfer unit 430 inputs the database updated by the database storage unit 420 and updates the database stored in storage unit 121 of vehicle 1. Ideally, the database transfer unit 430 transfers the updated database to vehicle 1 at the same time that the database update unit 410 updates the database. A configuration without the database storage unit 420 is possible, but it is preferable for the updated database to be managed by the database storage unit 420 for traceability purposes.

[0091] Therefore, the database stored in storage unit 121 can be updated via communication system 4 or a physical medium.

[0092] Other configurations and operations are essentially the same as those of the shape estimation system, the object recognition system and the communication system, as well as the shape estimation procedure, the object recognition procedure and the communication procedure according to the first embodiment described above, and details thereof are omitted.

[0093] The shape estimation system, the object recognition system and the communication system, as well as the shape estimation method, the object recognition method and the communication method according to the second embodiment of the present invention, can also achieve essentially similar advantageous effects as those of the shape estimation system, the object recognition system, the communication system, the shape estimation method, the object recognition method and the communication method according to the first embodiment described above.

[0094] Furthermore, the advantage is that, since the pre-prepared set of scattered wave / reflected wave spectra can be updated as a database, it is possible to address a case where the accuracy of the object shape estimation becomes a problem, or a case where the estimation accuracy can be improved.

[0095] It should be noted that the example described uses a communication system to update the database, but the database can also be updated via a physical medium. <Zusätzliche Bemerkungen>

[0096] The embodiments of the present invention have been described above. It should be noted that the present invention is not limited to or restricted by the embodiments described above and includes various modifications and equivalent configurations within the scope of the appended claims. For example, the embodiments described above have been described in detail to facilitate understanding of the present invention, but the present invention is not necessarily limited to including all the described configurations. Furthermore, part of the configuration of one embodiment can be replaced by the configuration of another embodiment. Additionally, the configuration of another embodiment can be added to the configuration of a particular embodiment.Furthermore, a part of the configuration of each embodiment can be added to, deleted from, or replaced by another configuration.

[0097] Furthermore, some or all of the configurations, functions, processing units, processing means, and the like described above can be implemented by hardware through a design using an integrated circuit, or can be implemented by software as a result of a processor interpreting and executing programs to implement each function.

[0098] Information such as programs, tables and files for implementing each function can be stored on a storage device such as memory, a hard disk or an SSD (Solid State Drive) or on a recording medium such as an IC card, an SD card or a DVD.

[0099] Furthermore, control lines and information lines indicate what is considered necessary for the description and do not necessarily represent all control lines and information lines required for implementation. In practice, it can be assumed that almost all configurations are interconnected. Reference symbol list 1 vehicle 2A transmitting wave 2B receive wave 3 detected objects 4 Communication system 100 object recognition systems 110 Transmit / Receive Unit 111 Transmitter unit 112 Receiving unit 113 Distance calculation unit 120 shape estimation system 121 storage unit 122 storage unit 123 New scanning unit 124 Conversion processing unit 125 Comparative calculation unit (shape estimation unit) 126 Type estimation unit 127 Dimensional estimation unit 130 Vehicle control unit 140 HMI 410 Database Update Unit 420 database storage unit 430 Database Transfer Unit QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] JP 2022-142252 A

[0003]

Claims

[1] Shape estimation system for a vehicle, wherein the system comprises: a storage unit that stores as recognition data the distance between the vehicle and a detected object as well as an amplitude value / intensity value of a wave corresponding to the distance, wherein the distance is measured on the basis of time, from the time at which a transmission wave is sent by a transmitting unit that sends the transmission wave to a detection area to be observed, until the time at which the wave is received by a receiving unit that receives the wave scattered / reflected by the detected object present in the detection area; a re-sampling unit which, based on the detection data, re-samples an amplitude / intensity value of the wave corresponding to the distance between the vehicle and the detected object as sampling data for a desired distance; a conversion processing unit that converts the re-sampling data output by the re-sampling unit from a spatial range into a spatial frequency range; a storage unit that stores a set of spectra in the spatial frequency domain for a variety of preset object shapes as a database in conjunction with the object shape; and a shape estimation unit that estimates the shape of the detected object based on the spectrum of the detected object output by the conversion processing unit and the set of spectra stored in the memory unit, wherein The shape estimation unit estimates the shape of the recognized object as the object shape that corresponds to the spectrum among the set of spectra in the storage unit that is closest to the spectrum of the recognized object output by the conversion processing unit. [2] Shape estimation system according to claim 1, wherein the re-scanning unit calculates the scan data from among the data contained in the recognition data by calculating the average value of the amplitude / intensity values ​​of the wave that falls within a predetermined distance range from the desired distance. [3] Shape estimation system according to claim 1, wherein the conversion processing unit performs conversion processing from a spatial domain to a spatial frequency domain after combining the resample data output by the resample unit with data obtained by inverting the sign of the spacing of the resample data. [4] Shape estimation system according to claim 1, wherein the shape estimation unit normalizes the set of spectra and all spectra of the recognized object output by the conversion processing unit using the respective maximum values ​​and then estimates the shape of the recognized object. [5] Shape estimation system according to claim 1, wherein the shape estimation unit estimates the type of the recognized object as the shape estimation of the recognized object. [6] Shape estimation system according to claim 5, wherein the shape estimation unit estimates at least one curb / wheel stop as the type of detected object. [7] Shape estimation system according to claim 5, wherein the shape estimation unit performs machine learning of the features of each type by using the set of spectra as training data and estimates the type from the spectrum of the recognized object based on the result of the machine learning. [8] Shape estimation system according to claim 5, wherein the shape estimation unit calculates for each type the average value of the difference between the set of spectra and the spectrum of the detected object and estimates the type for which the average value of this difference is minimal as the type of detected object. [9] Shape estimation system according to claim 1, wherein the shape estimation unit estimates the dimensions of the detected object as the shape estimation of the detected object. [10] Shape estimation system according to claim 1, wherein the shape estimation unit estimates the type of the detected object at the time of shape estimation of the detected object and then estimates the dimensions of the detected object after referring only to data under the set of spectra relating to the identified type of the detected object. [11] Shape estimation system according to claim 9, wherein the shape estimation unit estimates that the dimensions at which the difference between the set of spectra and the spectrum of the detected object is minimized are the dimensions of the detected object. [12] Shape estimation system according to claim 1, wherein the set of spectra is created on the basis of a simulation. [13] Shape estimation system according to claim 1, wherein the set of spectra is created on the basis of actual measurement data. [14] Shape estimation system according to claim 1, wherein the database stored in the storage unit can be updated via a communication system or a physical medium. [15] Shape estimation system according to claim 1, wherein the transmitting wave sent by the transmitting unit and the wave received by the receiving unit are ultrasonic waves. [16] Object detection system for a vehicle, the system comprising: a transmitting unit that sends a transmission wave to a detection area to be observed; a receiving unit that receives a wave scattered / reflected by a detected object present in the detection range; and a distance calculation unit that determines the distance between the vehicle and the detected object based on the time from when the transmitting unit sends the transmission wave until when the receiving unit receives the wave; and the shape estimation system according to any one of claims 1 to 15. [17] Communication system which communicates via a network with a vehicle which includes the object recognition system according to claim 16, wherein the communication system comprises: a database storage unit that stores a database used when the vehicle estimates the shape of a detected object; a database update unit that updates the database of the database storage unit; and a database transfer unit that transfers the database stored in the database storage unit to the vehicle's storage unit, wherein The database transfer unit transmits an updated database to the vehicle at the same time as the database update unit updates the database. [18] Shape estimation method for a vehicle, wherein the method comprises: a storage step of the storage, as recognition data, the distance between the vehicle and a detected object as well as an amplitude value / intensity value of a wave corresponding to the distance, wherein the distance is measured on the basis of time, from the time at which a transmission wave is sent by a transmitting unit that sends the transmission wave to a detection area to be observed, until the time at which the wave is received by a receiving unit that receives the wave scattered / reflected by the detected object present in the detection area; a re-scanning step of the re-scanning process, based on the detection data, an amplitude / intensity value of the wave corresponding to the distance between the vehicle and the detected object, as scanning data of a desired distance; a conversion processing step of converting the re-sampling data output in the re-sampling step from a spatial area into a spatial frequency area; a storage step of storing a set of spectra in the spatial frequency domain for a multitude of preset object shapes as a database in conjunction with the object shape; and a shape estimation step of estimating the shape of the recognized object based on the spectrum of the recognized object output in the conversion processing step and the set of spectra stored in the storage step, wherein In the shape estimation step, the object shape that corresponds to the spectrum among the set of spectra stored in the storage step, which is closest to the spectrum of the recognized object output in the conversion processing step, is estimated as the shape of the recognized object. [19] Object recognition methods, including: a transmission step of sending a wave to a detection area to be observed; a reception step of receiving a wave scattered / reflected by a detected object present in the detection range; a distance calculation step of determining the distance between the vehicle and the detected object based on the time from the time the wave is sent in the transmit step until the time the wave is received in the receive step; and each step of the shape estimation method according to claim 18. [20] Communication method for communicating via a network with a vehicle comprising a processing device performing the shape estimation method according to claim 18, wherein the communication method comprises: a database storage step of storing a database that is used when the vehicle estimates the shape of a detected object; a database update step of updating the database; and a database transfer step of transferring the database to a storage unit of the processing device, wherein In the database transfer step, the updated database is transferred to the storage unit at the same time as the database is updated in the database update step.

Images