METHOD FOR DETECTING AN OBSTACLE BY MEANS OF REFLECTED ULTRASONIC WAVES

Abstract

A method for detecting an obstacle utilizing reflected ultrasonic waves, comprises transmitting an ultrasonic burst transmission signal by an ultrasonic transmitter to a detection area to be observed and receiving an ultrasonic signal reflected by an obstacle in the detection area by an ultrasonic receiver as an ultrasonic reception signal. In the ultrasonic reception signal at least one echo is detected resulting from an obstacle. The echo section of the ultrasonic reception signal belonging to the echo is transformed from the time domain into the frequency domain. The frequency spectrum of the echo section is then examined for the presence of at least one of a plurality of predetermined spectral characteristics, wherein each spectral characteristic is representative of a predetermined obstacle type or a plurality of predetermined obstacle types. The echo section is allocated to a predetermined obstacle type based on the examination.

Claims

1.-10. (canceled)

11. A method for detecting an obstacle with reflected ultrasonic waves, comprising: transmitting, by an ultrasonic transmitter, an ultrasonic burst transmission signal to a detection area to be observed; receiving an ultrasonic signal reflected by an obstacle in the detection area by an ultrasonic receiver as an ultrasonic reception signal; detecting at least one echo in the ultrasonic reception signal resulting from the obstacle; transforming an echo section of the ultrasonic reception signal belonging to the echo from a time domain into a frequency domain to generate a frequency spectrum of the echo section; examining the frequency spectrum of the echo section for a presence of at least one of a plurality of predetermined spectral characteristics representative respectively of a predetermined obstacle type or a plurality of obstacle types; and allocating the echo section to one of the predetermined obstacle types.

12. The method according to claim 11, wherein a degree of probability is specified with which the echo section can be allocated to the obstacle type, or wherein a plurality of probabilities is specified with which the echo section can be allocated to different respective obstacle types.

13. The method according to claim 11, wherein a detected obstacle is signalized optically and/or acoustically and/or tactilely, and wherein respective signals from the optical and/or acoustic and/or tactile signalization are different for predetermined obstacle types.

14. The method according to claim 11, wherein the spectral characteristics include a spectral center of gravity and a spectral width of the frequency spectrum of the echo section.

15. The method according to claim 11, wherein the predetermined obstacle types comprise as a first type a curb and a bump, as a second type a wall and a vehicle, and as a third type a post and a pile.

16. The method according to claim 15, wherein the pile can be one selected from a first set of a pile for information and traffic signs, a pile for traffic lights and a pile for street lights, and further wherein the post can be one selected from a second set of a post for information and traffic signs, a post for traffic lights and a post for street lights.

17. The method according to claim 11, wherein a sensitivity of the ultrasonic receiver is temperature-dependent, and wherein a detection of the echo section and the transformation of the echo section into the frequency domain is temperature-compensated or temperature-corrected or otherwise carried out by taking into account a current temperature of the ultrasonic receiver.

18. The method according to claim 17, wherein the current temperature of the ultrasonic receiver is determined by measurement of the current temperature or based on a temperature-dependent signal characteristic of the ultrasonic reception signal.

19. The method according to claim 11, wherein the ultrasonic burst transmission signal comprises one to twenty pulses.

20. The method according to claim 11, wherein the ultrasonic burst transmission signal comprises five to fifteen pulses.

21. The method according to claim 11, wherein the ultrasonic burst transmission signal comprises eight to twelve pulses.

22. The method according to claim 11, wherein the transformation into the frequency domain is carried out by means of a Fourier transform, or by an algorithm from signal processing of a weather radar, such as an autocovariance procedure.

23. The method according to claim 11, wherein the allocation of the echo section to the representative predetermined obstacle type is carried out based on classifiers or based on pattern recognition.

Description

[0024] In the following, the disclosure is described in detail with reference to the drawings. The individual figures show the following:

[0025] FIG. 1 an example time course of an ultrasonic reception signal with distinctive echo sections due to a bump approximately 80 cm away from the receiver, with a distinctive echo section due to a post approximately 120 cm away from the ultrasonic receiver, and with a distinctive echo due to a wall approximately 150 cm away from the ultrasonic receiver;

[0026] FIG. 2 an example time course of an extracted echo section;

[0027] FIG. 3 a diagram in which the spectral center of gravity of an example extracted echo sections is plotted for the obstacles bump, wall and post when measuring these obstacles at different distances from the ultrasonic receiver;

[0028] FIG. 4 a diagram in which the spectral width of example extracted echo sections is plotted for the obstacles bump, wall and post when measuring these obstacles at different distances from the ultrasonic receiver; and

[0029] FIG. 5 a summary of the diagrams of FIGS. 3 and 4 as a 2D diagram of the aforementioned obstacles to be measured during tests, illustrating that each obstacle type can be distinguished by the location of the two aforementioned spectral characteristics within the 2D diagrams.

[0030] As aforementioned, laboratory tests were carried out within the scope of the disclosure, in which ultrasonic burst transmission signals, for example, with 8 pulses were transmitted with an ultrasonic transducer into a detection area in which a post (75 mm tube with a height of 1 m), a simulated standard curb with a length of 1 m and aligned at a right angle with the direction of propagation of the ultrasonic waves, and a wall were located. This scene was measured, whereby for example the time course of the ultrasonic reception signal was determined according to FIG. 1. In FIG. 1, the continuous line represents the ultrasonic reception signal, while the interrupted line represents a threshold signal. Three characteristic echo sections can be seen, namely for the bump (standard curb), for the post and for the wall.

[0031] FIG. 2 shows as an example the course of an echo section extracted from the ultrasonic reception signal. It has been proven to be a good compromise in terms of testing to use eight pulses as an ultrasonic burst transmission signal. An echo section is detected by means of the echo maximum. A plurality of samples was selected to the left and to the right of the maximum (exemplary eight samples before and eight samples after the echo maximum). Then, the spectral center of gravity and the spectral width of the extracted echo section were determined. From the literature (e.g. as cited in: Keeler, R. J./Passarelli, R. E. (1990): Signal Processing for Atmospheric Radars. In: Atlas, D. (eds.): Radar in Meteorology, American Meteorological Society, Boston, Mass.), the formula correlations for the spectral center of gravity and the spectral width are known as follows:

1. Classification No. (Spectral Center of Gravity)

[0032] [00001]$R\left(1\right)=\underset{m=0}{\overset{M-1}{.Math.}}.Math.\left(I_m-{\mathrm{jQ}}_m\right).Math.\left(I_{m+1}+{\mathrm{jQ}}_{m+1}\right)$$S.Math..Math.1=\arctan \left(\frac{.Math.\left(R\right)}{.Math.\left(R\right)}\right)$

2. Classification No. (Spectral Width)

[0033] [00002]$R\left(0\right)=\underset{m=0}{\overset{M-1}{.Math.}}.Math.\left(I_m-{\mathrm{jQ}}_m\right).Math.\left(I_m+{\mathrm{jQ}}_m\right)$$S.Math..Math.2=\sqrt{1-.Math.\frac{R\left(1\right)}{R\left(0\right)}.Math.}$

[0034] FIGS. 3 and 4 show the distribution of the different measurement results after calculation of the spectral centers of gravity and the spectral widths of the echo signals sections for the stopper, the post and the wall. FIG. 3 shows that a stopper can be distinguished from a wall and a post by means of the spectral center of gravity, wherein the two obstacle types wall and post cannot unambiguously be distinguished from one another. In contrast, FIG. 4 shows that the spectral width can be used to distinguish a wall as an obstacle from a stopper and a post, whereas a stopper and a post cannot be distinguished from one another.

[0035] FIG. 5 finally shows that all three obstacle types can be well distinguished from one another when considering both spectral parameters.

[0036] Thus, once these preliminary examinations have been carried out and field tests are subsequently carried out, it can be seen that the determination of the two spectral characteristics spectral center of gravity and spectral width makes it possible to allocate an obstacle to one of the three aforementioned types. The actual measurement point from spectral center of gravity and spectral width of an echo section to be measured is finally allocated to one of the three obstacle classes based on known classifiers. The probability with which a currently measured obstacle can be allocated to one of the three classes can then also be specified. Alternatively, it is also possible to specify several probabilities in order to specify the allocation probabilities of an obstacle to several classes.