Echo wave peak-based method for detecting surface to be walked and mobile robot

Abstract

The invention discloses a to-be-walked surface detection method based on an echo peak and a mobile robot, the to-be-walked surface detection method is applied to the robot, and a robot body is provided with an ultrasonic sensor; the method for detecting the surface to be walked comprises the following steps: controlling an ultrasonic sensor to emit an ultrasonic excitation signal to the surface to be walked, and controlling the ultrasonic sensor to collect an echo signal; the echo signal comprises at least one echo wave peak; the ultrasonic excitation signal is reflected by the surface to be walked to form the echo signal; and determining the attribute of the to-be-walked surface according to the phase of each echo peak and the amplitude of each echo peak included in the echo signal. Therefore, non-destructive detection on the to-be-walked surface is carried out by utilizing the propagation difference of the multiple echoes in the to-be-walked surfaces made of different materials, good anti-interference performance is achieved, the computing power burden of the robot is not increased, and rapid detection on the attributes of the to-be-walked surface is achieved.

Description

Technical Field

[0001] The present application relates to the field of sensor technology, and in particular to a method for detecting a surface to be walked on based on echo peaks and a mobile robot. Background Art

[0002] The cleaning robot automatically navigates and cleans on floors of various materials. Usually, the single cleaning strategy configured for existing cleaning robots is only applicable to surfaces to be cleaned with fixed floor properties. When the cleaning robot switches and walks among a variety of floor properties, the single cleaning strategy cannot effectively meet the cleaning requirements required for switching floor properties, thereby reducing the cleaning adaptability of the cleaning robot. Moreover, the cleaning robot disclosed in the prior art determines the properties of the surface to be cleaned based on the motion posture data collected by the motion sensor and the echo data collected by the ultrasonic module. The motion posture data collected by the cleaning robot during its walking on the surface to be cleaned requires a certain amount of computing power to be allocated to configure the motion posture data required for judgment. Therefore, combining the motion posture data with the echo data for analysis will increase the computing power burden of the robot, and will easily introduce motion delay errors into the ultrasonic module's detection of floor materials, thereby affecting the detection effect of the ultrasonic module. Summary of the invention

[0003] The present application discloses a method for detecting a surface to be walked on based on echo peaks and a mobile robot. The specific technical solution is as follows: A method for detecting a surface to be walked on based on echo peaks, the method for detecting a surface to be walked on is applied to a robot, and an ultrasonic sensor is arranged on the body of the robot; the method for detecting a surface to be walked on comprises: controlling the ultrasonic sensor to transmit an ultrasonic excitation signal toward the surface to be walked on, and controlling the ultrasonic sensor to collect an echo signal; the echo signal comprises at least one echo peak; the ultrasonic excitation signal is reflected by the surface to be walked on to form the echo signal; and the properties of the surface to be walked on are determined according to the phases of the respective echo peaks and the amplitudes of the respective echo peaks included in the echo signal.

[0004] Compared with the prior art method of combining and analyzing motion posture data and echo data, multiple echo peaks included in the echo signal are directly selected to determine the properties of the surface to be walked on, including detecting whether the surface to be walked on is a carpet surface or a hard ground; limited by the hardware characteristics and timing control characteristics of the ultrasonic sensor, the properties of the surface to be walked on are also detected in combination with the ultrasonic excitation signal, but the motion posture data is not introduced into the frequency domain for analysis. Even if the error factor is considered, the inclination of the surface to be walked on relative to the ultrasonic sensor is not used as a factor affecting misjudgment, but it can further distinguish carpet surfaces with different material properties, such as distinguishing the thickness of the material, the softness and hardness of the hair, etc.

[0005] Therefore, the surface detection method disclosed in the present application is based on the ultrasonic excitation signal emitted toward the surface to be walked, and the echo with relatively obvious amplitude change is obtained, and then the property of the surface to be walked is judged according to the phase and amplitude of each echo peak. Compared with the traditional analog-to-digital conversion processing of a single echo, the surface detection method disclosed in the present application supports the judgment of multiple echo peaks, and realizes the non-destructive detection of the surface to be walked by using the propagation difference of multiple echoes in the surface to be walked of different materials, and has good anti-interference performance, and improves the reliability of ultrasonic sensors to detect the ground material required for the robot to walk. Furthermore, compared with the method of combining and analyzing motion posture data and echo data in the prior art, the present application does not allocate computing power to configure the motion posture data required to be judged, and concentrates computing power in the software environment to use the phase and amplitude of multiple echo peaks that have been collected to quickly and stably judge the property of the surface to be walked, and timely judge whether the robot needs to avoid the carpet; it does not increase the computing power burden of the robot, and it is not easy to introduce motion delay errors to the ultrasonic sensor to detect the ground material.

[0006] Furthermore, the method for determining the properties of the surface to be walked on based on the phases of each echo peak and the amplitudes of each echo peak included in the echo signal includes: determining whether the surface to be walked on is a carpet surface or a non-carpet surface based on the phases of each echo peak and the amplitudes of each echo peak in the echo signal; and determining the material properties of the carpet surface based on the phase information between the ultrasonic excitation signal and the echo peaks in the echo signal when the surface to be walked on is determined to be a carpet surface. Not only the properties of the surface to be walked on are determined based on the multiple echo peaks included in the echo signal, but also the specific material properties of the surface to be walked on are detected in combination with the ultrasonic excitation signal.

[0007] Furthermore, the method for determining whether the surface to be walked on is a carpet surface or a non-carpet surface according to the phase of each echo peak and the amplitude of each echo peak in the echo signal includes: extracting the phase of each echo peak and the amplitude of each echo peak in a target sampling interval in the echo signal collected by the ultrasonic sensor, and then calculating the phase difference between different echo peaks; judging whether the maximum phase difference between different echo peaks in the target sampling interval is greater than a preset phase difference threshold, and whether the amplitudes of each echo peak in the target sampling interval are greater than a preset intensity threshold; and determining whether the maximum phase difference between different echo peaks in the target sampling interval is greater than a preset phase difference threshold. When the maximum phase difference between different echo peaks in the interval is greater than the preset phase difference threshold, and the amplitudes of each echo peak in the target sampling interval are greater than the preset intensity threshold, it is determined that the surface to be walked on is a non-carpet surface and the corresponding judgment result in the target sampling interval is identified as a non-carpet surface; when the maximum phase difference between different echo peaks in the target sampling interval is less than or equal to the preset phase difference threshold, or the amplitude of an echo peak in the target sampling interval is less than the preset intensity threshold, it is determined that the surface to be walked on is a carpet surface and the corresponding judgment result in the target sampling interval is identified as a carpet surface. Therefore, it is possible to determine whether the surface to be walked on is a carpet surface or a non-carpet surface only by comprehensively judging the existence of more than two echo peaks, which improves the reliability of detection compared to the method of only judging whether there is one echo peak in the target sampling interval.

[0008] Furthermore, when it is determined that the surface to be walked on is a carpet surface, the method for determining the material properties of the carpet surface based on the phase information between the ultrasonic excitation signal and the echo peak in the echo signal includes: traversing each echo peak in turn within the target sampling interval, extracting the echo peak with the smallest absolute value of the phase difference with the ultrasonic excitation signal, and then marking the echo peak with the smallest absolute value of the phase difference with the ultrasonic excitation signal as the echo peak to be compared, and counting the number of echo peaks traversed within the target sampling interval; when the number of echo peaks is less than the preset number of echoes, if the absolute value of the phase difference between the ultrasonic excitation signal and the echo peak to be compared is smaller and the amplitude of the echo peak to be compared is smaller, it is determined that the thickness of the carpet surface is greater. Therefore, when it is determined that the surface to be walked on is a carpet surface, in order to ensure that the material properties of the carpet surface are reliably determined, the present application does not skip the ultrasonic excitation signal that is located in front of the echo signal in time sequence, but instead selects to combine the ultrasonic excitation signal and a certain number of echo peaks behind it for amplitude and phase analysis, so as to reliably and accurately detect the material properties of the carpet surface.

[0009] Further, the target sampling interval includes M echo peaks within a preset sampling time period; wherein M is an integer greater than or equal to 1; the time period occupied by the endpoint on one side of each echo peak to the endpoint on the other side is the preset sampling time period to cover multiple sampling frequency points; each sampling frequency point within the same echo peak has a frequency and an amplitude; the peak of each echo peak is represented by a sampling frequency point to determine the phase of each echo peak and the amplitude of each echo peak, so that the phase difference between different echo peaks is equal to the phase difference between the sampling frequency points at the peaks of different echo peaks, and the amplitude of an echo peak is equal to the amplitude of the sampling frequency point at the peak of an echo peak.

[0010] Further, when M is greater than or equal to 3, if the maximum phase difference between different echo peaks is greater than a preset phase difference threshold, and the amplitude of each echo peak is greater than a preset intensity threshold, it is determined that the surface to be walked on is a hard ground; wherein the non-carpet surface includes a hard ground; and the preset number of echoes is equal to 3. Thus, the hard ground is detected in combination with the number of echo peaks in the target sampling interval, the maximum phase difference between echo peaks, and the amplitude of echo peaks.

[0011] Furthermore, if the corresponding judgment results in N consecutive target sampling intervals are all identified as carpet surfaces, then the detection result of the surface to be walked on is confirmed to be a carpet surface; if the corresponding judgment results in N consecutive target sampling intervals are all identified as non-carpet surfaces, then the detection result of the surface to be walked on is confirmed to be a non-carpet surface. The corresponding judgment results in the target sampling interval are filtered to improve the fault tolerance of the detection results of the surface to be walked on.

[0012] Furthermore, the method for generating the ultrasonic excitation signal includes: keeping the relative position between the ultrasonic sensor and the surface of the carpet of the preset first thickness unchanged, within the preset frequency calibration range, starting from the sampling frequency point at one end of the preset frequency calibration range, traversing each sampling frequency point according to a preset step length, each time a sampling frequency point is traversed, the frequency of the ultrasonic signal to be detected is adjusted to the frequency of the currently traversed sampling frequency point, then controlling the ultrasonic sensor to transmit the ultrasonic signal to be detected toward the surface of the carpet of the first thickness, and controlling the ultrasonic sensor to collect the echo signal reflected by the surface of the carpet of the first thickness, and then recording the amplitude of the echo peak included in the echo signal reflected by the surface of the carpet of the first thickness, and marking the recorded amplitude of the echo peak as the amplitude corresponding to the currently traversed sampling frequency point; when all the sampling frequency points within the preset frequency calibration range are traversed, the frequency of the sampling frequency point corresponding to the echo peak with the largest amplitude is configured as the frequency of the ultrasonic excitation signal; wherein the frequency at the midpoint of the preset frequency calibration range is the preset nominal frequency. The frequency point with the strongest echo is selected under the surface of the first thickness carpet to overcome the processing error of the transmitting probe or the receiving probe of the ultrasonic sensor. The performance of the ultrasonic sensor can be maximized when the subsequent calibration frequency point is used to drive the ultrasonic sensor.

[0013] Furthermore, in the case where it is determined that the surface to be walked on is a carpet surface, the method for determining the material properties of the carpet surface based on the phase information between the ultrasonic excitation signal and the echo peak in the echo signal includes: traversing each echo peak in turn within the target sampling interval, extracting the echo peak with the smallest absolute value of the phase difference with the ultrasonic excitation signal, and then marking the echo peak with the smallest absolute value of the phase difference with the ultrasonic excitation signal as the echo peak to be compared, and counting the number of echo peaks traversed within the target sampling interval; the target sampling interval includes multiple sampling frequency points to represent the peak data points of the corresponding echo peaks; at the echo peak When the number of echoes is less than the preset number, if the absolute value of the phase difference between the ultrasonic excitation signal and the echo peak to be compared is greater than the preset phase threshold, and / or the amplitude of the echo peak to be compared is greater than the preset amplitude threshold, then the carpet surface is determined to be a carpet surface of the first thickness; when the number of echo peaks is less than the preset number, if the absolute value of the phase difference between the ultrasonic excitation signal and the echo peak to be compared is less than or equal to the preset phase threshold, and the amplitude of the echo peak to be compared is less than or equal to the preset amplitude threshold, then the carpet surface is determined to be a carpet surface of the second thickness; wherein the thickness of the carpet surface of the second thickness is greater than the thickness of the carpet surface of the first thickness. Therefore, when it is determined that the surface to be walked on is a carpet surface, in order to ensure reliable determination of the material properties of the carpet surface, the present application does not skip the ultrasonic excitation signal located in front of the echo signal, but selects to combine the ultrasonic excitation signal and a certain number of echo peaks located behind it for amplitude and phase analysis, so as to reliably and accurately detect the thickness properties of the carpet surface.

[0014] A mobile robot, wherein an ultrasonic sensor is provided at the bottom of the front end of the robot body, and the mobile robot is configured to control the ultrasonic sensor to execute the method for detecting the surface to be walked; wherein the ultrasonic sensor is used to emit an ultrasonic excitation signal toward the surface to be walked of the mobile robot, and control the ultrasonic sensor to collect an echo signal; the ultrasonic sensor outputs the execution result of the method for detecting the surface to be walked through a reserved interface of an asynchronous receiver or an interface of an asynchronous transmitter; the execution result of the method for detecting the surface to be walked includes a non-carpet surface and a carpet surface of various material properties. Thus, the method does not rely on the computing power and memory resources of the circuit board related to motion control in the mobile robot, and can realize rapid detection of the properties of the surface to be walked, thereby distinguishing between the non-carpet surface and the carpet surface of different material properties, and timely judging whether the mobile robot needs to avoid the carpet; the computing power burden of the mobile robot is not increased, and it is not easy to introduce motion delay errors to the ultrasonic sensor in detecting the ground material. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1It is a schematic diagram of the morphology of the ultrasonic excitation signal and the collected echo signal when the walking surface is a hard ground.

[0016] Figure 2 It is a schematic diagram of the morphology of the ultrasonic excitation signal and the collected echo signal when the walking surface is a carpet surface of the second thickness.

[0017] Figure 3 It is a schematic diagram of the morphology of the ultrasonic excitation signal and the collected echo signal when the walking surface is a carpet surface of the first thickness.

[0018] Figure 4 It is a flow chart of a method for detecting a surface to be walked on based on echo peaks according to an embodiment of the present application. Implementation

[0019] The technical solutions in the embodiments of the present application will be described in detail below in conjunction with the drawings in the embodiments of the present application. To further illustrate the embodiments, the present application provides drawings. These drawings are part of the disclosure of the present invention, which are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, ordinary technicians in this field should be able to understand other possible implementation methods and the advantages of the present application.

[0020] The present application discloses a method for detecting a surface to be walked on based on echo peaks, and the method for detecting a surface to be walked on is applied to a robot, and an ultrasonic sensor is arranged on the body of the robot; the body of the robot can be constructed into any suitable shape, such as a cylindrical shape, an elliptical cylindrical shape or a square shape, and the body of the robot provides a housing space to accommodate various components; the ultrasonic sensor can be installed at the bottom or front of the body of the robot; the robot also includes a controller, and the controller is used to drive the ultrasonic sensor to emit an ultrasonic signal, and process the ultrasonic signal reflected by the surface to be walked, or directly receive the ultrasonic sensor's judgment result on the amplitude and phase information of the echo signal, so as to obtain the detection result of the surface to be walked on.

[0021] Depend on Figure 4 It can be seen that the method for detecting the surface to be walked on includes: The ultrasonic sensor is controlled to emit an ultrasonic excitation signal toward the surface to be walked, and the ultrasonic sensor is controlled to collect an echo signal; the echo signal includes at least one echo peak; it should be noted that the ultrasonic excitation signal is an ultrasonic signal with an optimal frequency so that the echo generated by the ultrasonic signal reflected on a surface to be walked of a material is the most obvious. The ultrasonic excitation signal is reflected by the surface to be walked to form the echo signal, and the echo signal is an ultrasonic signal reflected by the surface to be walked but cannot be regarded as directly equivalent to the ultrasonic excitation signal, because when the ultrasonic sensor emits the same ultrasonic excitation signal toward the surface to be walked with different properties, the reflected echo signal will present different forms. Specifically, when the surface to be walked is a hard and smooth ground, since most of the ultrasonic waves can be reflected back to the ultrasonic sensor, the echo signal includes at least three echo peaks; when the surface to be walked is a carpet surface, since the carpet surface has a weak ability to reflect ultrasonic waves, there may be only one or two echo peaks in the ultrasonic sensor echo signal. Among them, the surface to be walked with a soft surface will reduce the vibration of the robot in the vertical direction relative to the surface to be walked with a hard surface.

[0022] In this embodiment, the transmitting unit (transmitting probe) in the ultrasonic sensor and the receiving unit (receiving probe) in the ultrasonic sensor are integrated to form an ultrasonic probe; the ultrasonic sensor first transmits an ultrasonic excitation signal toward the surface to be walked on, then stops transmitting the ultrasonic excitation signal within a certain period of time, and controls the ultrasonic sensor to receive the ultrasonic signal reflected by the surface to be walked on according to a certain acquisition frequency, thereby collecting the echo signal and the ultrasonic excitation signal reflected by the surface to be walked on. Since the ultrasonic sensor staggers the transmission and reception of ultrasonic signals, the process of transmitting the ultrasonic excitation signal and the process of receiving the echo signal do not interfere with each other, which is conducive to collecting more accurate and reliable echo peaks.

[0023] In addition, during the stage when the ultrasonic sensor transmits ultrasonic excitation signals toward the surface to be walked and collects echo signals, the robot may walk on the surface to be walked or stay at a certain position on the surface to be walked. It can be understood that the ultrasonic sensor does not contact the surface to be walked, and even the driving wheels of the robot may not contact the surface to be walked to be detected.

[0024] Moreover, limited by the hardware characteristics and timing control characteristics of the ultrasonic sensor, the echo signal and the ultrasonic excitation signal reflected by the surface to be walked on appear on the same waveform. Figure 1 The echo signal shown in the figure has multiple peaks, and Figure 1The echo peaks shown and the ultrasonic excitation signal are presented on a waveform; for example, the ultrasonic excitation signal presents an undulating waveform 10, and the waveform 10 includes a plurality of undulating peaks; Figure 1 The first echo peak 11, the second echo peak 12, the third echo peak 13 and the fourth echo peak 14 appear in sequence; among them, the phase and amplitude of the echo peaks in the interval a1b1 after the waveform 10 will show a pattern corresponding to the surface to be walked on.

[0025] Then, the properties of the surface to be walked are determined according to the phases of each echo peak and the amplitudes of each echo peak included in the echo signal, wherein the phases of each echo peak represent the frequency domain positions of each echo peak or each echo in the same waveform, and the amplitudes of each echo peak describe the maximum intensity of each echo peak in the same waveform. When the same ultrasonic excitation signal is emitted, the phases and amplitudes of the echo peaks reflected from the surfaces to be walked with different properties are different, i.e., the propagation differences are formed in the surfaces to be walked with different materials. Thus, the type of the surface to be walked is determined by using the positions and intensities of each peak in the echo signal, and the determination method includes threshold comparison of the phase difference and amplitude of each echo peak, and in some threshold comparison operations, the number of echo peaks and the sampling time period are also combined for determination.

[0026] In order to use ultrasonic sensors to reliably and quickly detect the material of the ground on which the robot needs to walk, compared with the method of combining and analyzing motion posture data and echo data in the prior art, this embodiment directly selects multiple echo peaks included in the echo signal to determine the properties of the surface to be walked, including detecting whether the surface to be walked is a carpet surface or a hard ground; due to the hardware characteristics and timing control characteristics of the ultrasonic sensor, the properties of the surface to be walked are also detected in combination with the ultrasonic excitation signal, but the motion posture data is not introduced into the frequency domain for analysis. Even if the error factor is considered, the inclination of the surface to be walked relative to the ultrasonic sensor is not used as a factor affecting misjudgment, but it can further distinguish carpet surfaces with different material properties, such as distinguishing the thickness of the material, the softness and hardness of the hair, etc.

[0027] Therefore, the method for detecting the surface to be walked disclosed in the present embodiment is based on an ultrasonic excitation signal emitted toward the surface to be walked, to obtain an echo with relatively obvious amplitude change, and then judges the properties of the surface to be walked according to the phase and amplitude of each echo peak. Compared with the traditional analog-to-digital conversion processing of a single echo, the method for detecting the surface to be walked disclosed in the present embodiment supports the judgment of multiple echo peaks, and realizes non-destructive detection of the surface to be walked by utilizing the propagation differences of multiple echoes in surfaces to be walked of different materials. It has good anti-interference performance and improves the reliability of ultrasonic sensors in detecting the material of the ground on which the robot needs to walk.

[0028] Furthermore, compared with the prior art method of combining and analyzing motion posture data and echo data, this embodiment does not allocate computing power to configure the motion posture data required for judgment, but concentrates computing power in the software environment to use the phase and amplitude of multiple echo peaks that have been collected to quickly and stably judge the properties of the surface to be walked on, and promptly judge whether the robot needs to avoid the carpet; it does not increase the computing power burden of the robot, and is not easy to introduce motion delay errors to the ultrasonic sensor in detecting the ground material.

[0029] In some embodiments, the method of determining the properties of the surface to be walked on according to the phases of each echo peak and the amplitudes of each echo peak included in the echo signal includes: determining whether the surface to be walked on is a carpet surface or a non-carpet surface according to the phases of each echo peak and the amplitudes of each echo peak in the echo signal. Then, when it is determined that the surface to be walked on is a carpet surface, the material properties of the carpet surface are determined according to the phase information between the ultrasonic excitation signal and the echo peaks in the echo signal. Thus, based on the multiple echo peaks included in the echo signal, not only the properties of the surface to be walked on are determined, including detecting whether the surface to be walked on is a carpet surface or a hard ground; but also the specific material properties of the surface to be walked on are detected in combination with the ultrasonic excitation signal.

[0030] Generally, before identifying each echo peak, the ultrasonic signal reflected from the walking surface will be filtered and then converted into an echo signal. The echo signal can be copied to a related register through a DMA (Direct Memory Access) channel, and the register stores the echo data. Then, each echo peak is identified by executing a peak detection algorithm.

[0031] Specifically, the method for determining whether the surface to be walked on is a carpet surface or a non-carpet surface based on the phase of each echo peak and the amplitude of each echo peak in the echo signal includes: in the echo signal collected by the ultrasonic sensor, extracting the phase of each echo peak and the amplitude of each echo peak within a target sampling interval, and then calculating the phase difference between different echo peaks to facilitate subsequent corresponding threshold comparison. It can be understood that the target sampling interval is customized by the user based on echo collection experience; preferably, the target sampling interval includes 3 echo peaks, representing the primary echo peak, the secondary echo peak and the tertiary echo peak respectively.

[0032] Then, it is determined whether the maximum phase difference between different echo peaks in the target sampling interval is greater than the preset phase difference threshold, and whether the amplitude of each echo peak is greater than the preset intensity threshold. If yes, it is determined that the surface to be walked on is a non-carpet surface and the corresponding judgment result in the target sampling interval is identified as a non-carpet surface. Otherwise, it is determined that the surface to be walked on is a carpet surface and the corresponding judgment result in the target sampling interval is identified as a carpet surface. Therefore, it is possible to determine whether the surface to be walked on is a carpet surface or a non-carpet surface only by comprehensively judging more than two echo peaks, which improves the detection reliability compared to the method of only judging whether there is one echo peak in the target sampling interval.

[0033] When the maximum phase difference between different echo peaks in the target sampling interval is greater than the preset phase difference threshold, and the amplitude of each echo peak in the target sampling interval is greater than the preset intensity threshold, it is determined that the surface to be walked on is a non-carpet surface and the corresponding judgment result in the target sampling interval is identified as a non-carpet surface; for example, for conventional hard floor materials, the echo will be stronger, and there will be 3 or more echo peaks with large phase differences. Figure 1 As shown, when the surface to be walked on is a hard ground, four fluctuating echo peaks appear in the a1b1 time period (representing the target sampling interval), which are the first echo peak 11, the second echo peak 12, the third echo peak 13 and the fourth echo peak 14, respectively; in this embodiment, the ultrasonic excitation signal is presented as a waveform 10. In order to overcome the influence of the fluctuating peaks in the waveform 10 in the process of distinguishing the non-carpet surface from the carpet surface, the present embodiment skips the waveform 10 and selects the first echo peak 11, the second echo peak 12, the third echo peak 13 and the fourth echo peak 14 that are located behind the waveform 10 in time sequence for analysis. At this time, according to the phase of each echo peak included in the echo signal and the amplitude of each echo peak, it can be judged that the surface to be walked on is a non-carpet surface.

[0034] When the maximum phase difference between different echo peaks in the target sampling interval is less than or equal to the preset phase difference threshold, or the amplitude of an echo peak in the target sampling interval is less than the preset intensity threshold, it is determined that the surface to be walked on is a carpet surface and the corresponding judgment result in the target sampling interval is identified as a carpet surface. In this embodiment, the surface to be walked on is detected as a carpet surface, and the ultrasonic sensor transmits the ultrasonic excitation signal toward the surface to be walked on. Most of the ultrasonic waves are absorbed by the carpet, and only a small amount of ultrasonic signals are reflected back to the ultrasonic sensor, thereby collecting the echo signal. Therefore, relative to Figure 1 , Figure 2 The shapes of echo peak 21 and echo peak 22 in the echo signal shown do not show obvious peak shapes, that is, the amplitude of the echo peak is less than the preset intensity threshold and the maximum phase difference between different echo peaks is less than or equal to the preset phase difference threshold, that is, the amplitude of echo peak 21 and the amplitude of echo peak 22 fluctuate within a smaller amplitude range and a smaller phase range; therefore, for carpet material, one echo peak or two echo peaks generally appear, and the echo intensity is small, at least smaller than the echo intensity reflected by the hard floor material.

[0035] It is understandable that when the ultrasonic sensor transmits ultrasonic excitation signals toward the walking surfaces with different properties, the reflected echo signals will show different forms. Figure 1 and Figure 2 It can be seen that the echo signals formed by reflections from a carpet surface and a non-carpet surface will have significant differences in morphology. Based on the differences, those skilled in the art can configure a preset phase difference threshold and a preset intensity threshold for detecting the properties of the surface to be walked on, so that the robot can quickly determine whether the surface to be walked on is a carpet surface based on the relationship between the phase of each echo peak in the echo signal and the preset phase difference threshold, combined with the relationship between the amplitude of each echo peak and the preset intensity threshold.

[0036] On the basis of the above embodiment, when it is determined that the surface to be walked on is a carpet surface, the method for determining the material properties of the carpet surface based on the phase information between the ultrasonic excitation signal and the echo peak in the echo signal includes: traversing each echo peak in turn within the target sampling interval, extracting the echo peak with the smallest absolute value of the phase difference with the ultrasonic excitation signal, and then marking the echo peak with the smallest absolute value of the phase difference with the ultrasonic excitation signal as the echo peak to be compared, and counting the number of echo peaks traversed within the target sampling interval; when the number of echo peaks is less than the preset number of echoes, if the absolute value of the phase difference between the ultrasonic excitation signal and the echo peak to be compared is smaller and the amplitude of the echo peak to be compared is smaller, it is determined that the thickness of the carpet surface is greater.

[0037] The preset number of echoes is preferably 3; Figure 2 The number of echo peaks in Figure 3 The number of echo peaks in is less than 3, and Figure 2 The amplitudes of the echo peak 21 and the echo peak 22 are both less than half of the amplitude of the ultrasonic excitation signal. Figure 3 The amplitudes of the echo peaks 31 and 32 in the ultrasonic wave do not exceed half of the amplitude of the ultrasonic wave excitation signal 10. Figure 2 and Figure 3 It can be seen that Figure 2 The amplitude of the echo peak in Figure 3 The amplitude of the echo peak in Figure 2 The first echo peak on the left side is 21 Figure 3 The first echo peak 31 on the left side is closer to the ultrasonic excitation signal 10 (i.e. Figure 2 The echo time in is shorter). Therefore, Figure 2 The carpet surface detected by the echo peak in Figure 3 The carpet surface detected by the echo peak in (corresponding to the carpet surface of the first thickness) is large, and Figure 2 The echo peak in the image may detect the material of the carpet surface. Figure 3 The echo peaks in the image detect different materials on the carpet surface.

[0038] Therefore, when it is determined that the surface to be walked on is a carpet surface, in order to ensure that the material properties of the carpet surface are reliably determined, this embodiment does not skip the ultrasonic excitation signal that is located in front of the echo signal in time sequence, and selects to combine the ultrasonic excitation signal and a certain number of echo peaks behind it for amplitude and phase analysis, so as to reliably and accurately detect the material properties of the carpet surface.

[0039] It should be noted that the target sampling interval includes M echo peaks within a preset sampling time period; wherein M is an integer greater than or equal to 1; the time period occupied by the endpoint on one side of each echo peak to the endpoint on the other side is the preset sampling time period to cover multiple sampling frequency points, for example, on the waveform extending from the left endpoint of the echo peak to the right endpoint thereof, multiple sampling frequency points can be distributed sequentially according to the sampling interval or multiple sampling frequency points can be distributed continuously. Each sampling frequency point in the same echo peak has a frequency and an amplitude. The peaks of each echo peak are represented by sampling frequency points to determine the phase of each echo peak and the amplitude of each echo peak, so that the phase difference between different echo peaks is equal to the phase difference between the sampling frequency points at the peaks of different echo peaks, and the amplitude of an echo peak is equal to the amplitude of the sampling frequency point at the peak of an echo peak.

[0040] The method for determining the phase of each echo peak and the amplitude of each echo peak is to identify the echo peak through a peak detection algorithm, and then determine the phase of the echo peak and the amplitude of the echo peak. The basic principle of the peak detection algorithm is to identify the peak value in the echo signal by analyzing continuous sampling frequency points (traversing the maximum amplitude value in sequence). The peak value will specifically adopt the comparison method and the differential recognition method. Due to the conventional recognition means, it will not be repeated here. Therefore, those skilled in the art can construct the peak detection algorithm according to the difference between the initial echo and the secondary echo, so as to extract the phase and amplitude of the sampling frequency points in various echoes from the target sampling interval according to the peak detection algorithm.

[0041] On the basis of the above embodiment, when M is greater than or equal to 3, if the maximum phase difference between different echo peaks is greater than the preset phase difference threshold, and the amplitude of each echo peak is greater than the preset intensity threshold, then it is determined that the surface to be walked on is a hard ground, that is, it can be determined that the surface to be walked on is a hard and smooth ground; wherein, the non-carpet surface includes a hard ground, and the hard ground can be a cement surface, a wooden surface or a tile surface; accordingly, the preset number of echoes is equal to 3. In the case where the surface to be walked on is a hard and smooth ground, since most of the ultrasonic waves can be reflected back to the ultrasonic sensor, for example, a conventional cement floor, the echo will be stronger, and there will be 3 or more echo peaks with large phase differences. Figure 1 As shown, in the a1b1 time period (representing the target sampling interval), there are four fluctuating echo peaks with amplitudes greater than the preset intensity threshold, which are the first echo peak 11, the second echo peak 12, the third echo peak 13 and the fourth echo peak 14, respectively; wherein, the phase difference between the peak of the first echo peak 11 and the peak of the fourth echo peak 14, as the maximum phase difference between different echo peaks, is not only greater than the preset phase difference threshold, but also much greater than the phase difference occupied by the two sides of the ultrasonic excitation signal 10. Thus, the hard ground is detected by combining the number of echo peaks in the target sampling interval, the maximum phase difference between the echo peaks and the amplitude of the echo peaks.

[0042] In some embodiments, if the corresponding judgment results in N consecutive target sampling intervals are all identified as carpet surfaces, then the detection result of the surface to be walked on is confirmed to be a carpet surface; if the corresponding judgment results in N consecutive target sampling intervals are all identified as non-carpet surfaces, then the detection result of the surface to be walked on is confirmed to be a non-carpet surface. Wherein, N is preferably 3.

[0043] The judgment result obtained in the method of determining whether the surface to be walked on is a carpet surface or a non-carpet surface according to the phase of each echo peak and the amplitude of each echo peak in the echo signal in the aforementioned embodiment is a preliminary detection result; on this basis, the method of determining whether the surface to be walked on is a carpet surface or a non-carpet surface according to the phase of each echo peak and the amplitude of each echo peak in the echo signal is performed N times in succession. If the judgment results obtained N times in succession are all carpet surfaces, that is, the corresponding judgment results in N target sampling intervals that are continuous in time are all identified as carpet surfaces, then it is confirmed that the detection result of the surface to be walked on is a carpet surface, that is, the final detection result of the robot for the surface to be walked on is a carpet surface, otherwise the final detection result of the robot for the surface to be walked on is not a carpet surface; for example, the robot obtains an echo peak in a target sampling interval at an interval of 50 milliseconds, and continuously obtains 3 target sampling intervals. If the judgment results corresponding to the 3 target sampling intervals are all carpet surfaces, then the robot confirms that the surface to be walked on is a carpet surface, otherwise it is not.

[0044] Similarly, the method of determining whether the surface to be walked on is a carpet surface or a non-carpet surface according to the phase of each echo peak in the echo signal and the amplitude of each echo peak is performed N times in succession. If the judgment results obtained N times in succession are all non-carpet surfaces, that is, the corresponding judgment results in N time-continuous target sampling intervals are all identified as non-carpet surfaces, then it is confirmed that the detection result of the surface to be walked on is a non-carpet surface, that is, the final detection result of the robot on the surface to be walked on is a non-carpet surface, otherwise, the final detection result of the robot on the surface to be walked on is not a non-carpet surface; In summary, filtering is performed on the corresponding judgment results within the target sampling interval to improve the fault tolerance of the detection results of the walking surface.

[0045] As an embodiment, the method for generating the ultrasonic excitation signal includes: keeping the relative position between the ultrasonic sensor and the surface of the carpet of the first thickness unchanged, within the preset frequency calibration range, starting from the sampling frequency point at one end of the preset frequency calibration range, traversing each sampling frequency point according to a preset step length, each time a sampling frequency point is traversed, the frequency of the ultrasonic signal to be detected is adjusted to the frequency of the currently traversed sampling frequency point, and at the register level, each time a sampling frequency point is queried, the automatic load value of the register for configuring the driving frequency inside the ultrasonic sensor is modified to the frequency of the currently traversed sampling frequency point, which is recorded as a calibration of the frequency of the ultrasonic signal to be detected; then controlling the ultrasonic sensor to transmit the ultrasonic signal to be detected toward the surface of the carpet of the first thickness, and controlling the ultrasonic sensor to collect the echo signal reflected by the surface of the carpet of the first thickness, and then recording the amplitude of the echo peak included in the echo signal, and marking the amplitude of the echo peak as the amplitude corresponding to the currently traversed sampling frequency point; the mode of controlling the ultrasonic sensor to transmit is determined by the duty cycle adjustment mechanism, so that the ultrasonic signal to be detected is generated by multiple PWM square wave drives.

[0046] When all sampling frequency points within the preset frequency calibration range are traversed, the frequency of the sampling frequency point corresponding to the echo peak with the largest amplitude is configured as the frequency of the ultrasonic excitation signal, the correction of the ultrasonic signal to be detected is completed, and the calibration frequency is obtained; and the sampling frequency point corresponding to the echo peak with the largest amplitude is stored in the flash memory inside the ultrasonic sensor and saved as the calibration frequency point, so as to select the frequency point with the strongest echo under the surface of the carpet of the first thickness, and the subsequent calibration frequency point is used to drive the ultrasonic sensor to maximize the performance of the ultrasonic sensor.

[0047] Each time the ultrasonic sensor is powered on and starts transmitting, it reads the frequency of the calibration frequency point from the flash memory to drive the ultrasonic sensor to transmit the ultrasonic excitation signal. The ultrasonic sensor specifically obtains the ultrasonic excitation signal through duty cycle modulation to obtain the echo signal with the largest echo intensity, thereby selecting the calibration frequency point to drive the ultrasonic sensor to work, thereby improving measurement consistency. Subsequently, the carpet surface can be accurately identified when the robot just enters the carpet, and it can be promptly determined whether the robot needs to avoid the carpet or sweep and mop the carpet, thereby overcoming the processing errors of the transmitting probe or receiving probe of the ultrasonic sensor.

[0048] The pre-set nominal frequency is preferably 300KHz and the preset frequency calibration range is set to 280KHz to 320KHz, and the preset step size is set to 5Khz. When all the sampling frequency points in the preset frequency calibration range are traversed in sequence according to the preset step size, a total of 8 calibrations of the ultrasonic signal to be detected are completed, and then the reflected ultrasonic signal corresponding to the echo peak with the largest amplitude is selected as the ultrasonic excitation signal, forming a frequency calibration function of the ultrasonic sensor.

[0049] Preferably, the frequency at the midpoint of the preset frequency calibration range is a preset nominal frequency; the preset nominal frequency is the driving frequency set for the ultrasonic sensor to emit an excitation signal of corresponding frequency without calibrating the frequency of the ultrasonic signal to be detected, and is set to the frequency of the ultrasonic excitation signal by default.

[0050] As an embodiment, when it is determined that the surface to be walked on is a carpet surface, the method for determining the material properties of the carpet surface according to the phase information between the ultrasonic excitation signal and the echo peak in the echo signal includes: Each echo peak is traversed in turn within the target sampling interval; generally, each echo peak is identified by executing the aforementioned peak detection algorithm before traversing each echo peak. Each echo peak is derived from an echo signal collected by an ultrasonic sensor; the echo signal can be an ultrasonic signal received by the ultrasonic sensor and reflected from the surface to be walked on, and then converted to a digital signal. The echo signal can be copied to a related register through a DMA (Direct Memory Access) channel.

[0051] Then, the echo peak with the smallest absolute value of phase difference with the ultrasonic excitation signal can be extracted from the register to obtain the echo peak closest to the ultrasonic excitation signal in the target sampling interval. The echo peak with the smallest absolute value of phase difference with the ultrasonic excitation signal is then marked as the echo peak to be compared, and the number of echo peaks traversed in the target sampling interval is counted; wherein the target sampling interval includes multiple sampling frequency points to represent the peak data points of the corresponding echo peaks; the peak data points of the echo peaks are all configured with phase and amplitude to represent the phase and amplitude of the echo peaks respectively.

[0052] When it is determined that the surface to be walked on is a carpet surface, a statistical result of the number of echo peaks traversed within the target sampling interval is less than the preset echo number.

[0053] When the number of echo peaks is less than the preset number of echoes, if the absolute value of the phase difference between the ultrasonic excitation signal and the echo peak to be compared is greater than the preset phase threshold, and / or the amplitude of the echo peak to be compared is greater than the preset amplitude threshold, it is determined that the carpet surface is a carpet surface of the first thickness.

[0054] When the number of echo peaks is less than the preset number of echoes, if the absolute value of the phase difference between the ultrasonic excitation signal and the echo peak to be compared is less than or equal to the preset phase threshold, and the amplitude of the echo peak to be compared is less than or equal to the preset amplitude threshold, it is determined that the carpet surface is a carpet surface of the second thickness.

[0055] In combination with the foregoing embodiment, when the number of echo peaks is less than the preset number of echoes, the thickness of the second thickness carpet surface is greater than the thickness of the first thickness carpet surface because the smaller the absolute value of the phase difference between the ultrasonic excitation signal and the echo peak to be compared and the smaller the amplitude of the echo peak to be compared, the greater the thickness of the carpet surface.

[0056] The preset number of echoes is preferably 3; Figure 2 and Figure 3 The number of echo peaks in is less than 3, and Figure 2 The amplitudes of the echo peak 21 and the echo peak 22 are both less than half of the amplitude of the ultrasonic excitation signal. Figure 3 The amplitudes of the echo peaks 31 and 32 in the ultrasonic wave do not exceed half of the amplitude of the ultrasonic wave excitation signal 10. Figure 2 and Figure 3 It can be seen that Figure 2 The amplitude of the echo peak in Figure 3 The amplitude of the echo peak in Figure 2 The first echo peak on the left side is 21 Figure 3 The first echo peak 31 on the left side is closer to the ultrasonic excitation signal 10 (i.e. Figure 2 The echo time in is shorter). Therefore, Figure 2 The carpet surface detected by the echo peak in is the carpet surface of the second thickness, Figure 3 The carpet surface detected by the echo peak in is the carpet surface of the first thickness, Figure 2 The carpet surface detected by the echo peak in Figure 3 The carpet surface detected by the echo peak in (corresponding to the carpet surface of the first thickness) is large, and Figure 2 The echo peak in the image may detect the material of the carpet surface. Figure 3 The echo peaks in the image detect different materials on the carpet surface.

[0057] It should be noted that, compared Figure 2 and Figure 3 It can be seen that the echo signals formed by reflections from carpet surfaces of different thicknesses will have great differences in morphology. Those skilled in the art can configure preset phase thresholds and preset amplitude thresholds for detecting the properties of the surface to be walked on according to the difference in carpet thickness.

[0058] Therefore, when it is determined that the surface to be walked on is a carpet surface, in order to ensure that the material properties of the carpet surface are reliably determined, this embodiment does not skip the ultrasonic excitation signal that is located in front of the echo signal in time sequence, but selects to combine the ultrasonic excitation signal and a certain number of echo peaks behind it for amplitude and phase analysis, so as to reliably and accurately detect the thickness properties of the carpet surface.

[0059] The present application also discloses a mobile robot, wherein an ultrasonic sensor is arranged at the bottom of the front end of the robot body, and the mobile robot is configured to control the ultrasonic sensor to execute the method for detecting the surface to be walked on; the carpet surface can be identified by executing the method for detecting the surface to be walked on when the mobile robot just enters the carpet, and when the mobile robot is set as a cleaning robot, the cleaning robot can promptly determine whether it needs to avoid the carpet or sweep and mop the carpet.

[0060] Among them, the ultrasonic sensor is used to emit an ultrasonic excitation signal toward the surface to be walked on by the mobile robot, and control the ultrasonic sensor to collect an echo signal, which is formed by the ultrasonic excitation signal being reflected by the surface to be walked; the frequency of the ultrasonic excitation signal emitted by the ultrasonic sensor is the calibration frequency disclosed in the aforementioned embodiment.

[0061] Generally, before traversing each echo peak of the echo signal, the ultrasonic signal reflected from the surface to be walked on will be filtered, such as by using any suitable filtering circuit; then the analog-to-digital conversion of the ultrasonic signal after filtering is controlled to obtain the echo signal; then the echo signal can be copied to the relevant register through the DMA (Direct Memory Access) channel, and then the peak detection algorithm is executed to identify each echo peak from the register.

[0062] The execution results of the detection method of the surface to be walked include non-carpet surfaces and carpet surfaces of various material properties. In order to use ultrasonic sensors to reliably and quickly detect the ground material that the mobile robot needs to walk on, compared with the method of combining and analyzing motion posture data and echo data in the prior art, the present application directly selects multiple echo peaks included in the echo signal to determine the properties of the surface to be walked, including detecting whether the surface to be walked is a carpet surface or a hard ground; limited by the hardware characteristics and timing control characteristics of the ultrasonic sensor, the material properties of the carpet surface are also detected in combination with the ultrasonic excitation signal, but the motion posture data is not introduced into the frequency domain for analysis. Even if the error factor is considered, the inclination of the surface to be walked relative to the ultrasonic sensor is not used as a factor affecting misjudgment, but it can further distinguish carpet surfaces with different material properties, such as distinguishing the thickness of the material, the softness and hardness of the hair, etc.

[0063] The ultrasonic sensor outputs the execution result of the detection method of the surface to be walked by reserving an interface of an asynchronous receiver (corresponding to the output interface of the receiving probe) or an interface of an asynchronous transmitter (corresponding to the output interface of the transmitting probe) to obtain the detection result of the surface to be walked; it can also indicate soft materials (carpet surface) and hard materials (hard ground) by outputting two levels, wherein the circuit carrier relied on by the ultrasonic sensor to execute the detection method of the surface to be walked is a receiving circuit in the ultrasonic sensor that supports receiving multiple echoes; it does not rely on the computing power and storage resources of the circuit board related to motion control in the mobile robot, and realizes the rapid detection of the properties of the surface to be walked. Therefore, the present application does not allocate additional computing power to configure the motion posture data required for judgment, and concentrates computing power in the software environment to use the phase and amplitude of the multiple echo peaks that have been collected to quickly and stably judge the properties of the surface to be walked, and promptly judge whether the mobile robot needs to avoid the carpet; it does not increase the computing power burden of the mobile robot, and it is not easy to introduce motion delay errors to the ultrasonic sensor to detect the ground material.

[0064] Since the cleaning strategy or walking strategy corresponding to the surface to be walked on is a carpet surface, it will be different from the cleaning strategy or walking strategy under normal cleaning conditions, such as avoiding the carpet or slowing down. In order to provide a detection basis for subsequent cleaning strategies or walking strategies more reliably and safely, the present application can not only detect whether the surface to be walked on is a carpet surface, but also determine the material properties of the carpet surface. This approach can improve the working reliability and safety of the self-moving robot. Therefore, the present application can reliably identify the properties of the surface to be walked on, provide a basis for judging the cleaning strategy or walking strategy of the mobile robot, and enable the mobile robot to adapt to the cleaning or walking of various surfaces to be walked on, which is beneficial to improving the working reliability and user experience of the mobile robot.

[0065] The above embodiments are only for illustrating the technical concept and features of the present invention, and their purpose is to enable technicians familiar with the technical field to understand the content of the present invention and implement it accordingly, and they cannot be used to limit the protection scope of the present invention. Any equivalent changes or modifications made according to the spirit of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for detecting a surface to be walked on based on echo peaks, characterized in that: The method for detecting the surface to be walked is applied to a robot, and an ultrasonic sensor is provided on the robot body; The method for detecting a surface to be walked on comprises: Controlling the ultrasonic sensor to emit an ultrasonic excitation signal toward the surface to be walked on, and controlling the ultrasonic sensor to collect an echo signal; the echo signal includes at least one echo peak; the ultrasonic excitation signal is reflected by the surface to be walked on to form the echo signal; The property of the surface to be walked on is determined according to the phase of each echo peak and the amplitude of each echo peak included in the echo signal.

2. The method for detecting a surface to be walked on according to claim 1, characterized in that: The method for determining the property of the surface to be walked on according to the phase of each echo peak and the amplitude of each echo peak included in the echo signal includes: Determining whether the surface to be walked on is a carpet surface or a non-carpet surface according to the phase of each echo peak and the amplitude of each echo peak in the echo signal; When it is determined that the surface to be walked on is a carpet surface, the material properties of the carpet surface are determined according to the phase information between the ultrasonic excitation signal and the echo peak in the echo signal.

3. The method for detecting a surface to be walked on according to claim 2, characterized in that: The method for determining whether the surface to be walked on is a carpet surface or a non-carpet surface according to the phase of each echo peak and the amplitude of each echo peak in the echo signal comprises: In the echo signal collected by the ultrasonic sensor, the phase of each echo peak and the amplitude of each echo peak are extracted within a target sampling interval, and then the phase difference between different echo peaks is calculated; Determine whether the maximum phase difference between different echo peaks in the target sampling interval is greater than a preset phase difference threshold, and whether the amplitudes of each echo peak in the target sampling interval are greater than a preset intensity threshold; When the maximum phase difference between different echo peaks in the target sampling interval is greater than a preset phase difference threshold, and the amplitudes of the echo peaks in the target sampling interval are greater than a preset intensity threshold, it is determined that the surface to be walked on is a non-carpet surface and the corresponding judgment result in the target sampling interval is identified as a non-carpet surface; When the maximum phase difference between different echo peaks in the target sampling interval is less than or equal to a preset phase difference threshold, or when the amplitude of an echo peak in the target sampling interval is less than a preset intensity threshold, it is determined that the surface to be walked on is a carpet surface and the corresponding judgment result in the target sampling interval is identified as a carpet surface.

4. The method for detecting a surface to be walked on according to claim 3, characterized in that: In the case where it is determined that the surface to be walked on is a carpet surface, the method for determining the material properties of the carpet surface according to the phase information between the ultrasonic excitation signal and the echo peak in the echo signal includes: Traversing each echo peak in turn in the target sampling interval, extracting the echo peak with the smallest absolute value of phase difference with the ultrasonic excitation signal, marking the echo peak with the smallest absolute value of phase difference with the ultrasonic excitation signal as the echo peak to be compared, and counting the number of echo peaks traversed in the target sampling interval; When the number of echo peaks is less than the preset number of echoes, if the absolute value of the phase difference between the ultrasonic excitation signal and the echo peak to be compared is smaller and the amplitude of the echo peak to be compared is smaller, it is determined that the thickness of the carpet surface is greater.

5. The method for detecting a surface to be walked on according to claim 4, characterized in that: The target sampling interval includes M echo peaks within a preset sampling time period; wherein M is an integer greater than or equal to 1; The time period from the endpoint on one side to the endpoint on the other side of each echo peak is the preset sampling time period to cover multiple sampling frequency points; each sampling frequency point in the same echo peak has a frequency and an amplitude; The peak of each echo peak is represented by a sampling frequency point to determine the phase and amplitude of each echo peak, so that the phase difference between different echo peaks is equal to the phase difference between the sampling frequency points at the peaks of different echo peaks, and the amplitude of an echo peak is equal to the amplitude of the sampling frequency point at the peak of an echo peak.

6. The method for detecting a surface to be walked on according to claim 5, characterized in that: When M is greater than or equal to 3, if the maximum phase difference between different echo peaks is greater than the preset phase difference threshold, and the amplitude of each echo peak is greater than the preset intensity threshold, it is determined that the surface to be walked on is a hard ground; wherein the non-carpet surface includes a hard ground; and the preset number of echoes is equal to 3.

7. The method for detecting a surface to be walked on according to claim 3, characterized in that: If the corresponding judgment results in N time-continuous target sampling intervals are all identified as carpet surfaces, then it is confirmed that the detection result of the surface to be walked on is a carpet surface; If the corresponding judgment results in N time-continuous target sampling intervals are all identified as non-carpet surfaces, then it is confirmed that the detection result of the surface to be walked on is a non-carpet surface.

8. The method for detecting a surface to be walked on according to claim 1, characterized in that: The method for generating the ultrasonic excitation signal comprises: While keeping the relative position between the ultrasonic sensor and the surface of the carpet of the first thickness unchanged, within the preset frequency calibration range, starting from a sampling frequency point at one end of the preset frequency calibration range, traversing each sampling frequency point according to a preset step length, each time a sampling frequency point is traversed, the frequency of the ultrasonic signal to be detected is adjusted to the frequency of the currently traversed sampling frequency point, and then the ultrasonic sensor is controlled to transmit the ultrasonic signal to be detected toward the surface of the carpet of the first thickness, and the ultrasonic sensor is controlled to collect the echo signal reflected by the surface of the carpet of the first thickness, and then the amplitude of the echo peak included in the echo signal reflected by the surface of the carpet of the first thickness is recorded, and the recorded amplitude of the echo peak is marked as the amplitude corresponding to the currently traversed sampling frequency point; When all sampling frequency points within the preset frequency calibration range are traversed, the frequency of the sampling frequency point corresponding to the echo peak with the largest amplitude is configured as the frequency of the ultrasonic excitation signal.

9. The method for detecting a surface to be walked on according to claim 8, characterized in that: In the case where it is determined that the surface to be walked on is a carpet surface, the method for determining the material properties of the carpet surface according to the phase information between the ultrasonic excitation signal and the echo peak in the echo signal includes: Traversing each echo peak in turn in the target sampling interval, extracting the echo peak with the smallest absolute value of phase difference with the ultrasonic excitation signal, marking the echo peak with the smallest absolute value of phase difference with the ultrasonic excitation signal as the echo peak to be compared, and counting the number of echo peaks traversed in the target sampling interval; the target sampling interval includes multiple sampling frequency points to represent the peak data points of the corresponding echo peaks; In the case where the number of echo peaks is less than the preset number of echoes, if the absolute value of the phase difference between the ultrasonic excitation signal and the echo peak to be compared is greater than a preset phase threshold, and / or the amplitude of the echo peak to be compared is greater than a preset amplitude threshold, it is determined that the carpet surface is the carpet surface of the first thickness; In the case where the number of echo peaks is less than the preset number of echoes, if the absolute value of the phase difference between the ultrasonic excitation signal and the echo peak to be compared is less than or equal to the preset phase threshold, and the amplitude of the echo peak to be compared is less than or equal to the preset amplitude threshold, it is determined that the carpet surface is a carpet surface of the second thickness; The thickness of the surface of the carpet with the second thickness is greater than the thickness of the surface of the carpet with the first thickness.

10. A mobile robot, characterized in that: An ultrasonic sensor is arranged at the bottom of the front end of the robot body, and the mobile robot is configured to control the ultrasonic sensor to perform the method for detecting the surface to be walked on according to any one of claims 1 to 9; The ultrasonic sensor is used to transmit an ultrasonic excitation signal toward the surface to be walked on by the mobile robot, and control the ultrasonic sensor to collect the echo signal; The ultrasonic sensor outputs the execution result of the to-be-walked surface detection method through a reserved asynchronous receiver interface or an asynchronous transmitter interface; the execution result of the to-be-walked surface detection method includes non-carpet surfaces and carpet surfaces with various material properties.

Images