Obstacle detection methods, ultrasonic sensor chips and ultrasonic sensor systems

Abstract

This application provides an obstacle detection method, an ultrasonic sensor chip, and an ultrasonic sensor system. When a target peak exceeding a dynamic comparison threshold exists in the envelope of a real-time mixed signal, the method determines whether the target peak is an obstacle peak based on characteristic peak parameters and the first effective descent point. This method uses pre-acquired clean aftershock signals as a dynamic comparison benchmark and combines characteristic peak parameters and the first effective descent point to perform a dual judgment on the real-time signal to distinguish between aftershock interference and true echoes. Through this approach, it is possible to effectively avoid misjudging aftershocks as targets while ensuring that true echoes are not missed, thereby achieving reliable detection of near-range obstacles in aftershock interference environments.

Description

Technical Field

[0001] This application relates to the field of ultrasonic signal processing, and in particular to an obstacle detection method, an ultrasonic sensor chip, and an ultrasonic sensor system. Background Technology

[0002] During the signal processing of an ultrasonic radar system, in the aftershock phase, due to interference from aftershock signals, the peak value of the obstacle echo is not significantly higher than the aftershock peak value.

[0003] Traditional detection methods use fixed thresholds, which make it difficult to effectively distinguish between aftershocks and true echoes, resulting in the inability to reliably detect nearby obstacles during the aftershock phase. Summary of the Invention

[0004] This application provides an obstacle detection method, an ultrasonic sensor chip, and an ultrasonic sensor system to effectively detect obstacles during aftershocks.

[0005] In a first aspect, embodiments of this application provide an obstacle detection method, including:

[0006] Acquire real-time mixed signals including obstacle echoes;

[0007] When there is a target peak value in the envelope of the real-time mixed signal that exceeds the dynamic comparison threshold, it is determined whether the target peak value is an obstacle peak value based on the characteristic peak parameters and the first effective descent point.

[0008] The dynamic comparison threshold is determined based on the peak envelope sequence of the pure aftershock signal; the first effective descent point is the first position point in the peak envelope sequence of the pure aftershock signal that meets the following conditions: the number of consecutive descents is greater than or equal to K, and the cumulative descent ratio is greater than or equal to R; where K is an integer greater than or equal to 2, and R is a real number greater than 0 and less than 1.

[0009] Optionally, determining whether the target peak value is an obstacle peak value based on the characteristic peak parameters and the first effective descent point includes:

[0010] When the time position of the target peak is after the first effective descent point, and the magnitude of the target peak exceeds the corresponding position threshold value in the dynamic comparison threshold by a proportion greater than a preset proportion threshold, the target peak is determined to be an obstacle peak.

[0011] When the time position of the target peak is before the first effective descent point, and the deviation between the amplitude of the target peak and the amplitude of the corresponding characteristic peak in the characteristic peak parameter is less than or equal to a preset deviation threshold, the target peak is determined to be a non-obstacle peak.

[0012] Optionally, the dynamic comparison threshold is updated by matching the peak envelope sequence of the pure aftershock signal with the aftershock characteristics of the current ultrasonic radar in real time, and is used to separate obstacle signals superimposed on the aftershock.

[0013] Optionally, a threshold calibration step may also be included, the threshold calibration step comprising:

[0014] The peak envelope sequence is obtained by performing envelope processing on the pure aftershock signal collected under unobstructed conditions.

[0015] Identify multiple characteristic peaks in the peak envelope sequence, and record the height and corresponding time position of each characteristic peak as the characteristic peak parameter;

[0016] In the peak envelope sequence, the starting time point at which the aftershock signal decays to a reliable threshold is determined as the first effective descent point.

[0017] Optionally, the envelope processing of the clean aftershock signal acquired under unobstructed conditions to obtain the peak envelope sequence includes:

[0018] The absolute value of the pure aftershock signal is taken to obtain the absolute value signal;

[0019] The absolute value signal is subjected to a sliding window maximum value calculation to generate a peak envelope sequence with the same length as the pure aftershock signal, wherein the sliding window maximum value calculation is to calculate the maximum amplitude of multiple sampling points before and after each sampling point.

[0020] Optionally, identifying multiple characteristic peaks in the peak envelope sequence and recording the height and corresponding time position of each characteristic peak includes:

[0021] In the peak envelope sequence, the height and time position corresponding to all peaks that meet the preset peak conditions are identified and recorded.

[0022] Secondly, this application provides an ultrasonic sensor chip for use in the method described in the first aspect.

[0023] Thirdly, this application provides an ultrasonic sensor system, including the ultrasonic sensor described in the second aspect;

[0024] The main control module is used to output detection and calibration commands;

[0025] The transducer module, connected to the main control module, is used to transmit detection pulses according to the detection command and receive real-time mixed signals; it is also used to transmit detection pulses according to the calibration command and receive pure aftershock signals.

[0026] The signal acquisition module, connected to the transducer module and the ultrasonic sensor chip, is used to acquire the real-time mixed signal and pure aftershock signal received by the transducer module and transmit them to the ultrasonic sensor chip so that the ultrasonic sensor chip can perform obstacle detection and threshold calibration.

[0027] Optionally, the main control module is further configured to output long and short sequence transmission commands; the transducer module is further configured to alternately transmit short sequence pulses and long sequence pulses according to the long and short sequence transmission commands;

[0028] The system also includes:

[0029] A gain control module, connected to the main control module and the transducer module, is used to set a first driving current and a first receiving gain when transmitting a short sequence, and to set a second driving current and a second receiving gain when transmitting a long sequence, according to the long and short sequence transmission commands.

[0030] Wherein, the first driving current is less than the second driving current, and the first receiving gain is less than the second receiving gain.

[0031] Optionally, the main control module is also used to control the transducer module to stop transmitting the long sequence and switch to transmitting a short sequence if the ultrasonic sensor chip detects an obstacle before the end of the long sequence working cycle during the transmission of the long sequence.

[0032] Optionally, the main control module is also used to output aftershock cancellation commands; the transducer module is also used to transmit cancellation waves according to the aftershock cancellation commands;

[0033] The system also includes:

[0034] The anti-vibration transmission module is connected to the main control module and the transducer module. It is used to generate an anti-vibration wave that is in the same phase and opposite in direction as the aftershock based on the anti-vibration power, phase and delay information output by the main control module, and drive the transducer module to transmit the anti-vibration wave until the peak amplitude of the aftershock signal is reduced to below a preset threshold.

[0035] Optionally, it may also include at least one of the following modules:

[0036] A storage module, connected to the main control module, is used to store a benchmark dataset and detection logs; wherein, the benchmark dataset includes a dynamic comparison threshold, feature peak parameters, and the first effective descent point;

[0037] The power module connects to each module and is used to supply power to each module;

[0038] The interface module is connected to the main control module and is used to communicate with external devices and output test results.

[0039] The obstacle detection method, ultrasonic sensor chip, and ultrasonic sensor system provided in this application determine whether a target peak is an obstacle peak based on characteristic peak parameters and the first effective descent point when a target peak exceeding a dynamic comparison threshold exists in the envelope of a real-time mixed signal. This method uses pre-acquired clean aftershock signals as a dynamic comparison benchmark and combines characteristic peak parameters and the first effective descent point to perform a dual judgment on the real-time signal, distinguishing between aftershock interference and true echoes. This approach effectively avoids misjudging aftershocks as targets while ensuring that true echoes are not missed, thus achieving reliable detection of near-range obstacles even in aftershock interference environments. Attached Figure Description

[0040] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0041] Figure 1 This is a schematic diagram of the detection architecture provided in this application;

[0042] Figure 2 Flowchart of the obstacle detection method provided in this application Figure 1 ;

[0043] Figure 3 A schematic diagram showing the comparison between the echo envelope and the calibration threshold provided in this application;

[0044] Figure 4 Flowchart of the obstacle detection method provided in this application Figure 2 ;

[0045] Figure 5 A schematic diagram of the aftershock signal and maximum value filtered data provided in this application;

[0046] Figure 6 A schematic diagram of the ultrasonic sensor system provided in this application;

[0047] Figure 7 A comparison diagram of the working modes provided for this application.

[0048] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0049] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0050] Ultrasonic radar systems are widely used for short-range detection in fields such as automotive driver assistance, robot obstacle avoidance, and liquid level measurement due to their low cost and mature technology.

[0051] Its typical signal processing flow is as follows: Figure 1 As shown: An ultrasonic transducer emits ultrasonic detection pulses, which are reflected by obstacles to form echoes. These echoes are converted into weak analog electrical signals by a receiving transducer. This weak signal is first fed into a programmable gain amplifier (PGA) for dynamic amplitude adjustment, and then filtered by a bandpass filter to remove environmental noise and power frequency noise. The conditioned analog signal is then converted from analog to digital by an analog-to-digital converter (ADC), generating a digital data stream which is sent to a digital signal processor (DSP). The DSP sequentially performs down-conversion, high-precision digital filtering, and envelope processing. It can also perform chirp demodulation to accurately extract the clean echo envelope signal for obstacle detection.

[0052] However, after the transducer emits a pulse, it generates a persistent aftershock due to mechanical inertia. This aftershock signal exhibits time-varying characteristics, with its amplitude decaying exponentially over time and its frequency potentially drifting. During the sustained aftershock phase (typically corresponding to the near-field detection range), the echo signals from nearby obstacles highly overlap with the aftershock signal in the time domain, resulting in the echo peak value not being significantly higher than the aftershock peak value, making the two difficult to distinguish.

[0053] Existing technologies mostly employ fixed thresholds for obstacle detection, which are typically set based on a noise floor. However, during the aftershock phase, the amplitude of aftershock signals is much higher than the noise floor and varies drastically. Fixed thresholds cannot adapt to the rapid changes in aftershock amplitude, easily leading to false alarms (mistaking aftershocks for targets) or missed detections (submerging real echoes in aftershocks). Therefore, existing threshold methods cannot effectively distinguish between target-containing and target-free states under aftershock interference.

[0054] To address this, this application proposes an obstacle detection method that utilizes pre-acquired clean aftershock signals as a dynamic comparison benchmark. It introduces the peak envelope (as a dynamic threshold), characteristic peak parameters, and the first effective descent point to perform a dual judgment on the real-time signal: first, it compares whether the signal amplitude exceeds the dynamic threshold; then, it combines the timing (first effective descent point) and morphological characteristics (characteristic peak parameters) to distinguish between aftershock interference and true echoes. This method effectively avoids misjudging aftershocks as targets (i.e., false alarms) while ensuring that true echoes are not missed, thus achieving reliable detection of near-field obstacles even in aftershock interference environments.

[0055] The technical solution of this application is mainly applied to automotive reversing radar systems, and specific scenarios include:

[0056] Within 10-30cm behind a vehicle (such as low obstacles or ground protrusions), traditional radar struggles to detect nearby targets due to aftershock interference. This solution utilizes clean aftershock signals as a dynamic reference, combined with the first effective descent point, to effectively distinguish between aftershocks and real echoes, achieving reliable detection in near-range blind spots.

[0057] Aftershock characteristics may change under conditions such as temperature variations, transducer aging, and circuit parameter drift. This solution, by pre-collecting and modeling pure aftershock signals, can adaptively adjust dynamic thresholds and characteristic peak parameters to maintain detection accuracy and avoid misjudgments caused by environmental changes.

[0058] When aftershock signals are superimposed on target echoes, the false alarm rate of the static threshold method increases significantly. This scheme, through a dual judgment mechanism (first comparing whether the amplitude exceeds the threshold, and then combining time series and morphological characteristics for secondary verification), can effectively distinguish between aftershock interference and real echoes, reducing the risk of false alarms and missed detections.

[0059] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0060] Figure 2 Flowchart of the obstacle detection method provided in this application Figure 1 ,like Figure 2 As shown, the method includes:

[0061] S201. Acquire the real-time mixed signal containing obstacle echoes.

[0062] In this embodiment, the real-time mixed signal is a composite electrical signal acquired by the receiving circuit after the transducer emits a detection pulse. This composite signal includes obstacle echoes, aftershock signals generated by the transducer's residual mechanical oscillations, and environmental noise. It is understood that in the absence of obstacles, the real-time mixed signal is the superposition of the aftershock signal and environmental noise.

[0063] S202. When there is a target peak value in the envelope of the real-time mixed signal that exceeds the dynamic comparison threshold, determine whether the target peak value is an obstacle peak value based on the characteristic peak parameters and the first effective descent point.

[0064] In this embodiment, the dynamic comparison threshold is determined based on the peak envelope sequence of the pure aftershock signal. This sequence is used to characterize the maximum possible amplitude of the system's own aftershock signal at each time point under unobstructed conditions. This threshold changes as the aftershock signal decays and is called the dynamic comparison threshold.

[0065] For example, under unobstructed conditions, the pure aftershock signal collected after the ultrasonic transducer emits a detection pulse is envelope-processed to obtain a peak envelope sequence, which is used as a dynamic comparison threshold.

[0066] However, factors such as temperature, humidity, and device aging can cause aftershock characteristics to drift over time, making the pre-calibrated threshold no longer match the actual aftershock characteristics. When the threshold is too high, weak obstacle echoes are submerged, leading to missed detections; when the threshold is too low, aftershock fluctuations are misjudged as obstacles, leading to false alarms.

[0067] In one possible implementation, the dynamic comparison threshold is based on the real-time matching of the peak envelope sequence of the pure aftershock signal with the aftershock characteristics of the current ultrasonic radar, and is used to separate obstacle signals superimposed on the aftershock.

[0068] By matching in real time, the threshold is kept consistent with the current aftershock characteristics, enabling precise separation of weak obstacle signals superimposed on aftershocks. This significantly improves the target detection rate in blind zones while avoiding false alarms caused by aftershock fluctuations. Furthermore, it can automatically adapt to the effects of individual sensor differences, circuit parameter drift, temperature changes, and installation conditions, exhibiting robustness far superior to fixed-parameter methods.

[0069] For example, in actual operation, the system continuously monitors the received signal, separates the pure aftershock signal from the real-time mixed signal containing obstacle echoes through the signal processing algorithm, and dynamically updates its peak envelope sequence so that the dynamic comparison threshold matches the aftershock characteristics of the current ultrasonic radar in real time.

[0070] For example, envelope processing is performed on the real-time acquired mixed signal to obtain a real-time envelope signal. The real-time envelope signal is then compared point by point with a dynamic comparison threshold: if the real-time envelope value is greater than the dynamic threshold value, the point is marked as exceeding the threshold; otherwise, it is marked as not exceeding the threshold.

[0071] Subsequently, all consecutive points exceeding the threshold are aggregated into a candidate peak region. Specifically, starting from the first point exceeding the threshold, the process proceeds backward until the first non-threshold point is encountered; all points exceeding the threshold in between constitute a candidate peak region. This process continues, searching for the next point exceeding the threshold and repeating the above steps until all sampling points have been traversed.

[0072] For each candidate peak region, the maximum amplitude within that region is identified as the target peak corresponding to that region, and its amplitude and corresponding time position are recorded. If at least one candidate peak region exists, it is determined that there is a target peak exceeding the dynamic comparison threshold in the envelope of the real-time mixed signal, and each candidate peak region corresponds to one target peak.

[0073] In this embodiment, the characteristic peak parameter refers to the quantitative description of each characteristic peak in the peak envelope sequence of the pure aftershock signal, typically including the characteristic peak height val(n) and the characteristic peak position pos(n). Here, the characteristic peak height val(n) is the amplitude value of the nth characteristic peak, and the characteristic peak position pos(n) is the time point or sampling point number corresponding to the nth characteristic peak. The characteristic peak parameter reflects the inherent oscillation characteristics of the aftershock signal in the time domain; that is, aftershocks do not decay monotonically but exhibit multiple fluctuating peaks and troughs.

[0074] The first effective descent point refers to the starting time point in the peak envelope sequence that characterizes the initial decay of the aftershock signal to a confidence threshold. Before this point, the aftershock signal has a high amplitude and fluctuates violently, making it difficult to distinguish the target echo; after this point, the aftershock signal has decayed to a sufficiently low level, making the target echo easier to identify.

[0075] The first effective descent point is the first location point in the envelope signal of the pure aftershock signal that meets the following conditions: the number of continuously descent sampling points is greater than or equal to K, and the cumulative descent ratio is greater than or equal to R; where K is an integer greater than or equal to 2, and R is a real number greater than 0 and less than 1.

[0076] The cumulative decline percentage refers to the overall decline from the first decline point to the point where the Kth decline action ends.

[0077] By constraining the number of consecutive descents and the descent ratio, misjudgments caused by single-point noise fluctuations or accidental descents are avoided, ensuring that the first effective descent point can accurately reflect the true attenuation turning point of the aftershock signal, thus providing a reliable time boundary basis for subsequent obstacle detection.

[0078] Understandably, when both of the above conditions are met simultaneously, this time point is determined as the first effective descent point. This time point marks the first transition of the aftershock signal from a stage of high amplitude and violent fluctuations to a stage of lower amplitude and more stable attenuation.

[0079] In other implementations, the first effective descent point can also be determined by the following methods: the time point when the signal value first falls below a preset amplitude threshold, the time point when the signal value descent rate first exceeds a preset rate threshold, or the time point when the slope of the signal value envelope first changes from negative to positive, etc.

[0080] By limiting the detection of the first effective descent point to the envelope signal of the pure aftershock signal, the interference of high-frequency oscillation components in the original aftershock signal on the continuous descent determination is eliminated, making the descent trend clearer and more identifiable, improving the reliability of the first effective descent point detection, while reducing data processing volume and improving computational efficiency. Furthermore, by employing a cumulative descent ratio, the overall effect of multiple consecutive descent steps can be accumulated, avoiding missed detections due to insufficient single-step descent ratios, and also avoiding misjudgments due to large fluctuations at a single point, thus more accurately reflecting the overall attenuation degree of the aftershock signal.

[0081] For example, if K=5 and R=0.3, when the signal value drops 5 times consecutively and the cumulative drop rate is ≥30%, it is determined to be the first effective drop point. The position of this point is recorded. After this point, the aftershock signal attenuates significantly, and the detection reliability is significantly improved.

[0082] For example, after confirming the existence of a target peak exceeding the threshold, a comprehensive judgment is further made using characteristic peak parameters and the first effective descent point:

[0083] (1) Time and location judgment

[0084] If the target peak occurs earlier than the first effective descent point, it indicates that the peak occurred before the aftershock signal had sufficiently decayed. In this case, the peak is likely caused by fluctuations in the aftershock itself, rather than echoes from actual obstacles, and is preliminarily identified as aftershock interference.

[0085] If the target peak's time position is later than or equal to the first valid descent point, it indicates that the peak occurred after the aftershock signal had decayed to a reliable level. In this case, the probability that the peak was caused by obstacle echoes increases significantly, and it is preliminarily identified as a candidate obstacle peak.

[0086] (2) Feature peak matching and amplitude threshold judgment

[0087] For target peak values ​​whose time position is earlier than the first effective descent point, and for target peak values ​​whose time position is later than or equal to the first effective descent point, their amplitude is further compared with the threshold value of the corresponding position in the dynamic comparison threshold envp(t).

[0088] In one possible implementation, when the time position of the target peak is after the first effective descent point, and the amplitude of the target peak exceeds the corresponding position threshold value in the dynamic comparison threshold by a proportion greater than a preset proportion threshold, the target peak is determined to be an obstacle peak; when the time position of the target peak is before the first effective descent point, and the deviation between the amplitude of the target peak and the amplitude of the corresponding characteristic peak in the characteristic peak parameter is less than or equal to a preset deviation threshold, the target peak is determined to be a non-obstacle peak.

[0089] After the first effective descent point, the aftershock signal has attenuated to a sufficiently low level, and the signal background tends to stabilize. If a peak value with an amplitude significantly exceeding the dynamic threshold appears at this point, it indicates that the signal energy is much higher than the aftershock energy in the targetless state, and it is highly likely to be caused by the echo of a real obstacle. Therefore, it is determined to be a reliable obstacle peak.

[0090] Before the first effective descent point, the aftershock signal has a high amplitude and fluctuates violently, with multiple inherent characteristic peaks. If the amplitude of the target peak closely matches the amplitude of the corresponding characteristic peak (within the preset range), it indicates that the peak is consistent with the inherent oscillation characteristics of the aftershock and belongs to the fluctuation of the aftershock itself, rather than the echo of the obstacle. Therefore, it is determined to be a non-obstacle peak (i.e., aftershock interference).

[0091] Through the aforementioned judgment mechanism, the amplitude exceeding the threshold proportion is used as the judgment criterion after the first effective descent point, avoiding misjudgments caused by residual aftershock fluctuations. Since aftershocks have sufficiently attenuated by this period, almost all peak values ​​exceeding the threshold originate from real obstacles, thus significantly reducing the false alarm rate. Before the first effective descent point, characteristic peak matching is used for identification, excluding peak values ​​whose amplitude is highly consistent with the inherent characteristic peaks of aftershocks, effectively preventing aftershock fluctuations from being misjudged as obstacles, further reducing false alarms.

[0092] For example, the preset ratio threshold and the preset deviation threshold can be determined according to the actual situation. The preset ratio threshold can be adjusted according to the actual application scenario. For example, in a scenario with a high requirement for false alarm rate, the preset ratio threshold can be set to 20%~50% (e.g., 30%); in a scenario with a high requirement for detection rate, the preset ratio threshold can be set to 15%~20% (e.g., 15%). The preset deviation threshold is any value between 5% and 10%, for example, 8%.

[0093] In a specific example, taking a high detection rate scenario, when a car is reversing, the system continuously emits detection pulses to collect a real-time mixed signal containing obstacle echoes. The peak envelope sequence envp(t) stored during the threshold calibration phase is used as a dynamic comparison threshold to extract the envelope of the real-time mixed signal. When a peak value exceeding the threshold occurs, a judgment is made based on the height and position of the characteristic peak and the first valid descent point fa_rd.

[0094] If the peak value appears before the first valid descent point, and the peak height deviates from the characteristic peak height by ≤10%, it is determined to be aftershock interference and is not considered an obstacle; if the peak value appears after the first valid descent point, and the peak height exceeds 15% of the corresponding position threshold in the dynamic comparison threshold, it is determined to be a reliable obstacle peak value, and obstacle distance information is output. Figure 3 As shown, the second peak exceeding the threshold is the obstacle peak.

[0095] As an alternative approach, for all target peak values ​​exceeding the threshold, both amplitude exceeding the threshold and characteristic peak matching are performed simultaneously. Only when both conditions are met is the peak value identified as a real obstacle. Specifically: if the amplitude of the target peak exceeds the corresponding position threshold value in the dynamic comparison threshold by a proportion greater than a preset proportion threshold, and the deviation from the characteristic peak height is greater than a preset deviation threshold, it is determined to be a reliable obstacle peak; otherwise, it is determined to be aftershock interference or noise.

[0096] As another implementation method, the first effective descent point is used as the dividing point, and the detection time axis is divided into a front window and a back window. Within the front window, a judgment strategy is adopted that prioritizes feature peak matching and supplements it with amplitude exceeding a threshold: if the deviation between the amplitude of the target peak and the height of the feature peak is less than or equal to a preset deviation threshold, it is judged as aftershock interference; otherwise, a comprehensive judgment is made based on the amplitude exceeding threshold ratio. Within the back window, a judgment strategy is adopted that prioritizes amplitude exceeding a threshold and supplements it with feature peak matching: if the proportion of the target peak's amplitude exceeding the corresponding position threshold value in the dynamic comparison threshold is greater than a preset proportion threshold, it is judged as a reliable obstacle peak; otherwise, a comprehensive judgment is made based on feature peak matching.

[0097] Based on the above analysis, this method, through a three-layer judgment logic of target peak, time location, and characteristic peak matching, can effectively distinguish between real obstacle echoes and aftershock fluctuations in close-range scenarios with severe aftershock interference, and significantly reduce false alarm rate and missed detection rate.

[0098] The obstacle detection method provided in this application does not rely on a fixed amplitude threshold, but uses the pure aftershock signal itself as a dynamic comparison benchmark and introduces the temporal characteristics of the aftershock signal as an auxiliary criterion. This can effectively avoid misjudging aftershocks as targets (i.e., false alarms) while ensuring that real echoes are not missed, thereby achieving reliable detection of near-field obstacles in aftershock interference environments.

[0099] Figure 4 Flowchart of the obstacle detection method provided in this application Figure 2 ,like Figure 4 As shown, in this embodiment... Figure 2 Based on the embodiments, the method further includes: a threshold calibration step, which includes:

[0100] S301. Envelope processing is performed on the pure aftershock signal collected under unobstructed conditions to generate a peak envelope sequence.

[0101] For example, a probe pulse is emitted and the echo signal is acquired in an open environment without any obstacles (such as an anechoic chamber) to obtain a clean aftershock signal (containing weak background noise). The signal is then envelope-processed to extract its amplitude profile and generate a peak envelope sequence.

[0102] For example, control the ultrasonic transducer to emit a detection pulse with a frequency of 40kHz, and simultaneously collect the output signal through the receiving transducer to filter out the pure aftershock signal x(t) that does not contain obstacle echoes. The acquisition time is 10ms and the sampling frequency is 1MHz.

[0103] In one possible implementation, the absolute value of the pure aftershock signal is first taken to obtain an absolute value signal; then, the maximum value of the sliding window is calculated on the absolute value signal to generate a peak envelope sequence of the same length as the original signal. This peak envelope sequence accurately characterizes the peak envelope change of the pure aftershock signal; wherein, the maximum value of the sliding window is calculated by calculating the maximum amplitude of multiple sampling points before and after each sampling point.

[0104] Ultrasonic echo signals are typically attenuating oscillating signals containing alternating positive and negative waveforms. Directly processing the envelope of the original signal would distort the result due to the alternating positive and negative values. Taking the absolute value ensures all amplitudes are positive, facilitating subsequent amplitude profile extraction.

[0105] For each sampling point, the sliding window maximum algorithm finds the maximum value within its neighborhood (window), thus following the peak changes of the signal. The window width determines the smoothness of the envelope: a larger width results in a smoother envelope, but a slower response to rapid changes; a smaller width makes the envelope closer to the original signal, but may retain too much fluctuation. This algorithm can effectively extract the peak profile of the signal while suppressing negative fluctuations (valleys), and has low computational cost, making it suitable for real-time operation in resource-constrained embedded systems.

[0106] For example, calculating the absolute value of the pure threshold signal x(t) yields the absolute value signal y(t), and setting N=8 (i.e., calculating the maximum amplitude of 8 points before and after y(t)) results in a peak envelope sequence envp(t) with the same length as x(t), such as... Figure 5 As shown, this peak envelope sequence accurately characterizes the peak envelope variation of the pure aftershock signal.

[0107] It should be noted that taking the absolute value of the pure aftershock signal first to obtain the absolute value signal, and then performing sliding window maximum calculation on the absolute value signal, is only one example. It can also be implemented in other ways, such as performing Hilbert transform on the pure aftershock signal to obtain the analytic signal, and then taking its magnitude as the envelope signal. This method can accurately extract the instantaneous amplitude of the signal and is suitable for scenarios that require high-precision envelope.

[0108] S302. Identify multiple characteristic peaks in the peak envelope sequence and record the height of each characteristic peak and its corresponding time position as characteristic peak parameters.

[0109] For example, the generated peak envelope curve is analyzed to identify obvious peaks that do not decrease monotonically, i.e., characteristic peaks of aftershocks. The system automatically identifies these peaks and records the height (amplitude value) and location (time point or sampling point number) of each peak, forming characteristic peak parameters. For example: [(peak 1 height, peak 1 location), (peak 2 height, peak 2 location),...].

[0110] In one possible implementation, the height and time position of all peaks that satisfy preset peak conditions are identified and recorded in the peak envelope sequence. By recording all characteristic peaks that satisfy the preset peak conditions, the most comprehensive reference information is provided for subsequent characteristic peak matching, thereby improving matching accuracy.

[0111] The preset peak condition is a configurable set of parameters that can include:

[0112] Local maximum condition: The value of this point is greater than the values ​​of its immediate and next-to-immediate adjacent points, ensuring that the identified local peak point in the signal is a local peak point.

[0113] Amplitude significance condition: The value of this point is greater than the preset minimum amplitude threshold in order to eliminate noise interference and avoid misjudging low-amplitude noise as a characteristic peak;

[0114] Morphological salience condition: The difference between the point and its nearest preceding and following troughs is greater than a preset threshold to ensure that a distinct peak is identified, rather than a gentle undulation.

[0115] For example, using a peak detection algorithm, six characteristic peaks are identified in the peak envelope sequence. The heights val(1)-val(6) and corresponding positions pos(1)-pos(6) of each characteristic peak are recorded, where val(1) is the height of the aftershock main peak and pos(1) is the time point corresponding to the aftershock main peak.

[0116] For example, in subsequent obstacle determination, the detected peak can be compared peak by peak with a list of characteristic peaks. If the height and location match perfectly, it is determined to be an aftershock rather than an obstacle, and the matching accuracy is much higher than that of simple amplitude comparison.

[0117] It should be noted that identifying and recording the height and time position of all peaks that meet the preset peak conditions is only one example. It can also be achieved through other methods. For example, the peak detection threshold can be dynamically adjusted according to the overall amplitude of the signal, which is suitable for scenarios where the signal amplitude changes greatly with distance or environment.

[0118] S303. In the peak envelope sequence, determine the starting time point when the aftershock signal decays to the confidence threshold, which is taken as the first effective descent point.

[0119] For example, after the peak envelope curve experiences initial sharp fluctuations, find a turning point from which the signal value continues to decline significantly without any large rebounds. This point is the first effective decline point, marking the end of the severe aftershock interference zone and the beginning of the stable signal zone.

[0120] For example, the point in time when the signal value (such as envelope amplitude) decreases continuously for a preset number of times and the decrease ratio reaches a preset percentage threshold is considered the first valid decrease point. Reaching the preset number of consecutive decreases, i.e., the number of consecutive decreasing sampling points of the signal value reaching or exceeding a preset value (e.g., 2 points, 3 points, 5 points, etc.), ensures that the downward trend is continuous, rather than a single-point random fluctuation. The decrease ratio condition ensures that the decrease amplitude is sufficiently significant to reflect the transition of the aftershock signal from a high-amplitude oscillation phase to a stable decay phase.

[0121] The descent ratio can be either a single-step descent ratio or a cumulative descent ratio: a single-step descent ratio refers to the descent ratio of the current signal value relative to the previous signal value being greater than or equal to a preset ratio threshold; a cumulative descent ratio refers to the overall descent ratio from the starting point of the descent to the current point being greater than or equal to a preset ratio threshold.

[0122] When both of the above conditions are met simultaneously, this time point is determined as the first effective descent point. This time point marks the transition of the aftershock signal from a stage of high amplitude and violent fluctuations to a stage of lower amplitude and more stable attenuation. By constraining the signal with both the number of consecutive descents and the descent ratio, misjudgments caused by single-point noise fluctuations or accidental descents are avoided, ensuring that the first effective descent point accurately reflects the true attenuation turning point of the aftershock signal, thus providing a reliable time boundary for subsequent obstacle detection.

[0123] For example, if K=5 and R=0.3, when the signal value drops 5 times consecutively and the cumulative drop rate is ≥30%, it is determined to be the first effective drop point. The position of this point is recorded. After this point, the aftershock signal attenuates significantly, and the detection reliability is significantly improved.

[0124] In one possible implementation, the triggering conditions for the threshold calibration step include at least one of the following: system initialization, a preset change in the usage scenario, circuit board aging, and a change in transducer position.

[0125] System initialization refers to the first operation of the ultrasonic radar system after its initial power-on or reset. At this time, no threshold reference data has been stored in the system, so a threshold calibration procedure needs to be performed to establish an initial dynamic threshold reference dataset.

[0126] For example, when the vehicle starts, the on-board ultrasonic radar system is powered on and initialized, automatically triggering the threshold calibration step. After confirming that there are no obstacles in front, it collects clean aftershock signals and generates a benchmark dataset to provide a basis for subsequent obstacle detection.

[0127] Changes in application scenarios refer to significant changes in the working environment of the ultrasonic radar system, such as changes in environmental parameters like temperature, humidity, and air pressure, or changes in the layout of obstacles around the installation location. These changes may cause shifts in the characteristics of aftershock signals, rendering the original threshold reference data inapplicable, thus requiring recalibration.

[0128] For example, when a vehicle moves from an indoor parking lot into an outdoor environment, the temperature drops sharply from 25°C to 5°C, which may change the resonant frequency and aftershock attenuation characteristics of the ultrasonic transducer. At this point, if the system detects that the temperature change exceeds a preset threshold, it automatically triggers a threshold calibration step, re-acquires clean aftershock signals, and updates the baseline dataset.

[0129] Circuit board aging refers to the slow drift of electrical characteristics in electronic components (such as amplifiers, filters, ADCs, etc.) of an ultrasonic radar system as the usage time increases. This drift may cause changes in parameters such as signal amplification and noise floor, thereby affecting the accuracy of threshold reference data.

[0130] For example, after the system has accumulated a preset running time, the threshold calibration step is automatically triggered to reacquire clean aftershock signals and update the benchmark dataset to compensate for changes in signal characteristics caused by circuit aging.

[0131] Transducer position change refers to a change in the installation location or orientation of an ultrasonic transducer, such as due to vibration, impact, or human adjustment causing transducer displacement. This change in transducer position alters its coupling characteristics with the surrounding environment, thus affecting the morphology of the aftershock signal.

[0132] For example, after a minor collision, the system detects that the installation angle of the transducer has deviated beyond a preset threshold, automatically triggers the threshold calibration step, re-acquires clean aftershock signals and updates the benchmark dataset to ensure the accuracy of obstacle detection.

[0133] The obstacle calibration method provided in this embodiment obtains benchmark data such as dynamic comparison threshold, characteristic peak parameters, and first effective descent point through pre-calibration, avoiding the computational delay and resource consumption caused by dynamically calculating the threshold in real-time detection, and improving the efficiency and reliability of real-time detection.

[0134] This application also provides an ultrasonic sensor chip for performing the obstacle detection method described above.

[0135] The chip integrates the hardware modules and / or software algorithms required to perform the above-mentioned obstacle detection methods, and is capable of determining obstacles.

[0136] This application also provides an ultrasonic sensor system, such as Figure 6 As shown, the system includes the aforementioned ultrasonic sensor chip, main control module, transducer module, and signal acquisition module. The main control module outputs detection and calibration commands. The transducer module, connected to the main control module, transmits detection pulses according to the detection commands and receives real-time mixed signals; it also transmits detection pulses according to the calibration commands and receives pure aftershock signals. The signal acquisition module, connected to the transducer module and the ultrasonic sensor chip, acquires the real-time mixed signals and pure aftershock signals received by the transducer module and transmits them to the ultrasonic sensor chip, enabling the ultrasonic sensor chip to perform obstacle detection and threshold calibration.

[0137] Based on the aforementioned ultrasonic sensor chip, the ultrasonic sensor system further integrates a main control module, a transducer module, and a signal acquisition module, forming a more complete system architecture. Each module has a clear division of labor and works collaboratively to complete obstacle detection and threshold calibration functions.

[0138] In this embodiment, the main control module is the control core of the system, used to output detection and calibration commands and coordinate the working status of each module in the system. The detection command is used to trigger the transducer module to emit detection pulses and start the signal acquisition and processing process, entering the real-time obstacle detection mode. The calibration command is used to trigger the transducer module to emit detection pulses and start the threshold calibration process, entering the threshold calibration mode.

[0139] For example, the main control module is used to determine whether threshold calibration needs to be performed based on preset triggering conditions (such as system initialization, environmental changes, device aging, etc.) and to generate corresponding calibration instructions. It can also receive obstacle detection results output by the ultrasonic sensor chip and perform subsequent operations based on the results (such as issuing an alarm, controlling vehicle braking, etc.). The main control module can also manage the system's power status and control the switching between detection mode and calibration mode for each module.

[0140] For example, the main control module can be a microcontroller, which serves as the control core of the hardware device, with a main frequency of 168MHz. This main control module is used to trigger the threshold calibration process, control the transmission of long and short sequences, drive the aftershock cancellation closed-loop control, process detection results, and output instructions. Its built-in timer is used for phase detection and delay control to ensure real-time performance and control accuracy, thereby meeting the requirements of dynamic threshold control, long and short sequence switching, and aftershock cancellation closed-loop control.

[0141] In this embodiment, the transducer module is connected to the main control module and is used to perform corresponding transmission and reception operations according to the received instructions. For example, according to the detection instructions sent by the main control module, it transmits detection pulses (such as 40kHz ultrasonic pulses) and receives a real-time mixed signal containing obstacle echoes and aftershock signals. It can also transmit detection pulses according to the calibration instructions sent by the main control module and, if it is confirmed that there are no obstacles, receive pure aftershock signals (containing weak background noise).

[0142] For example, the transducer module can employ four 40kHz ultrasonic transducers, symmetrically mounted on the rear bumper of a vehicle. Two transducers are used to transmit probe pulses and cancel out echoes, while the other two are used to receive aftershock signals and obstacle echoes. The transducer has a rated power of 1W, a transmit voltage of 12V, and a receive sensitivity of no less than -70dB. It supports both short-sequence (low current) and long-sequence (high current) transmission modes to meet the requirements of gain control and echo cancellation.

[0143] For example, in calibration mode, the transducer module emits a detection pulse and receives a clean aftershock signal when the vehicle is stationary and there are no obstacles in front; in detection mode, the transducer module periodically emits a detection pulse while the vehicle is in motion and receives a real-time mixed signal containing obstacle echoes.

[0144] In this embodiment, the signal acquisition module is connected to the transducer module and the ultrasonic sensor chip, and is used to acquire the signal received by the transducer module and transmit it to the ultrasonic sensor chip for processing.

[0145] In detection mode, the transducer module acquires a real-time mixed signal containing obstacle echoes and aftershock signals, which is then transmitted to the ultrasonic sensor chip for obstacle detection. In calibration mode, the transducer module acquires a pure aftershock signal, which is then transmitted to the ultrasonic sensor chip for threshold calibration.

[0146] For example, the signal acquisition module can consist of an ADC acquisition chip and a filtering circuit. The ADC has an adjustable sampling frequency, ranging from 100kHz to 1MHz, and a 24-bit resolution, used to acquire pure aftershock signals, real-time mixed signals, and canceled ultrasonic signals. The filtering circuit uses a second-order low-pass filter with a cutoff frequency of 100kHz to filter out high-frequency interference, ensuring the integrity of the acquired signal and thus providing an accurate signal source for dynamic threshold establishment and aftershock detection.

[0147] In practical applications, the main control module detects that the triggering conditions are met (such as system initialization) and sends a calibration command to the transducer module. The transducer module then emits probe pulses according to the calibration command and receives the clean aftershock signal. The signal acquisition module acquires the clean aftershock signal and transmits it to the ultrasonic sensor chip. The ultrasonic sensor chip performs a threshold calibration step, generating a peak envelope sequence, characteristic peak parameters, and the first effective descent point, which are stored in the storage module. The ultrasonic sensor chip can also report the calibration completion status back to the main control module.

[0148] In one possible implementation, the main control module periodically sends detection commands to the transducer module. The transducer module emits detection pulses according to the detection commands and receives real-time mixed signals. The signal acquisition module acquires the real-time mixed signals and transmits them to the ultrasonic sensor chip. The ultrasonic sensor chip executes an obstacle detection method, comparing the peak envelope of the real-time mixed signal with a dynamic threshold, and making a comprehensive judgment based on characteristic peak matching and the first effective descent point. The ultrasonic sensor chip can also feed back obstacle detection results (such as the presence or absence of obstacles, obstacle distance, etc.) to the main control module. The main control module performs subsequent operations (such as issuing an alarm, controlling vehicle braking, etc.) based on the detection results.

[0149] By setting up a main control module, a transducer module, and a signal acquisition module, the main control module is responsible for control logic and decision-making, the transducer module is responsible for acoustic-to-electric conversion, the signal acquisition module is responsible for signal conditioning and digitization, and the ultrasonic sensor chip is responsible for core algorithm processing. The modules work in parallel, which improves the overall processing efficiency of the system.

[0150] Optionally, the main control module is also used to output long and short sequence transmission commands; the transducer module is also used to alternately transmit short sequence pulses and long sequence pulses according to the long and short sequence transmission commands;

[0151] The system also includes a gain control module, which is connected to the main control module and the transducer module, and is used to set a first drive current and a first receiving gain when transmitting a short sequence, and to set a second drive current and a second receiving gain when transmitting a long sequence, according to the long and short sequence transmission commands; wherein the first drive current is less than the second drive current, and the first receiving gain is less than the second receiving gain.

[0152] In this embodiment, in addition to outputting detection and calibration commands, the main control module is also used to output long and short sequence transmission commands to control the transducer module to alternately transmit short sequence pulses and long sequence pulses.

[0153] Short pulse sequences are used for short-range detection (e.g., 10cm~30cm), employing fewer pulses (e.g., 8), smaller drive current, and lower receiver gain to avoid oversaturation of the short-range aftershock signal. Long pulse sequences are used for long-range detection (e.g., 25cm~500cm), employing more pulses (e.g., 96), larger drive current, and higher receiver gain to ensure the signal-to-noise ratio of the long-range echo signal.

[0154] The transducer module alternately transmits short-sequence pulses and long-sequence pulses according to the long and short sequence transmission commands from the main control module. For example, after the system starts up, it repeatedly alternates between transmitting short and long sequences, that is, first transmitting a short sequence (lasting 5ms), then transmitting a long sequence (lasting 20ms), and repeating the cycle.

[0155] The gain control module is connected to the main control module and the transducer module, and is used to set different drive currents and receive gains when transmitting different sequences of transmission commands, based on the length of the transmission commands.

[0156] Short sequence mode: Set the first drive current (small, such as 50mA) and the first receive gain (small, such as 20dB). The small drive current and small receive gain effectively reduce the saturation region of the ADC, avoid excessive saturation of the near-field aftershock signal, and thus improve the detection conditions in the near-field blind zone.

[0157] Long sequence mode: Set the second drive current (larger, such as 150mA) and the second receive gain (larger, such as 40dB). The large drive current and large receive gain ensure that the long-distance echo signal has sufficient amplitude for reliable detection.

[0158] In this embodiment, the short sequence employs a small drive current and a small receiver gain, reducing the ADC saturation region and avoiding oversaturation from near-range aftershocks. Combined with a dynamic threshold and the first effective descent point, it can accurately identify obstacles in a 10-30cm blind zone, improving blind zone detection accuracy by approximately 30%. The long sequence employs a large drive current and a large receiver gain, ensuring the signal-to-noise ratio of the long-range echo signal, enabling stable detection of targets from 25cm to 500cm, improving long-range detection stability by approximately 20%. Figure 7 As shown, where, Figure 7 (a) in the diagram illustrates the working mode of this application. Figure 7 (b) in the diagram illustrates the existing working mode.

[0159] Furthermore, the dynamic switching mechanism between long and short sequences allows the system to simultaneously cover near- and far-range detection within a single duty cycle. When a target is detected earlier using a long sequence, the system immediately switches to a short sequence to continue detection, improving energy utilization and maintaining a data rate of 28Hz to meet real-time requirements.

[0160] Furthermore, in short sequence mode, low gain and low current combined with dynamic threshold improve blind zone detection; in long sequence mode, high gain and high current combined with characteristic peak matching improve long-distance detection accuracy. The combination of these two approaches achieves high-precision, high-reliability obstacle detection across the entire range.

[0161] For example, the gain control module may consist of a gain adjustment chip and a drive circuit. This module supports continuous adjustment of the receive gain from 0 to 60 dB and adjustment of the drive current from 10 to 200 mA. The main control module outputs corresponding control signals according to the transmission sequence type (short sequence or long sequence) to adjust the gain and drive current. Specifically, in short sequence mode, it outputs a drive current of 50 mA and a receive gain of 20 dB; in long sequence mode, it outputs a drive current of 150 mA and a receive gain of 40 dB, thereby meeting the gain control requirements.

[0162] Optionally, the main control module is also used to control the transducer module to stop transmitting the long sequence and switch to transmitting the short sequence if the ultrasonic sensor chip detects an obstacle before the end of the long sequence working cycle during the transmission of the long sequence.

[0163] In the alternating long and short sequence transmission mode, the long sequence is used for long-range detection and has a longer duty cycle. If an obstacle is detected before the long sequence's duty cycle ends, continuing to transmit the remaining long sequence pulses will result in unnecessary energy consumption and wasted time.

[0164] The main control module monitors the detection results of the ultrasonic sensor chip in real time. Once an obstacle is detected in advance within the long sequence working cycle, the remaining pulse transmission of the current long sequence is terminated to avoid ineffective energy consumption. The system then immediately switches to short sequence transmission mode to continue detecting the close-range area (e.g., 10cm~30cm) without waiting for the current long sequence working cycle to end. This shortens the system's response time to close-range obstacles and improves real-time performance.

[0165] To facilitate understanding of the technical solution of this application, the following section combines long and short sequence combination detection and gain control techniques to provide a detailed explanation of how to balance blind zone detection performance and long-distance detection effect.

[0166] 1. Threshold Calibration Stage

[0167] Completely consistent with the aforementioned threshold calibration method, the benchmark dataset {envp(t), val(n), pos(n),fa_rd} was collected and stored.

[0168] 2. Setting Operating Parameters

[0169] Short sequence: using fixed frequency mode (fix mode), 8 pulses, detection range of 10cm to 30cm, drive current of 50mA, and receiving gain of 20dB.

[0170] Long sequence: Chirp mode is used, with 96 pulses, a detection range of 25cm to 500cm, a drive current of 150mA, and a receiving gain of 40dB.

[0171] 3. Alternating long and short sequence transmissions

[0172] After the system starts, it repeatedly transmits short and long sequences alternately: first a short sequence with a duration of 5ms, then a long sequence with a duration of 20ms, and so on. The initial data rate is 28Hz (i.e., 1 / 35ms).

[0173] 4. Real-time detection and sequence switching

[0174] During the long sequence transmission, if a target (range from 25cm to 500cm) is detected before the end of the duty cycle (i.e., within 20ms), the long sequence transmission is immediately stopped, and the system switches to a short sequence to continue detection, maintaining a data rate of 28Hz. If no target is detected during the long sequence, the long and short sequences continue to be transmitted in a loop.

[0175] 5. Blind Spot Detection and Judgment

[0176] During short-sequence transmissions, the low-gain, low-current design effectively reduces the saturation region of the ADC, avoiding excessive saturation of near-field aftershock signals. Combined with the aforementioned dynamic threshold and the first effective descent point (fa_rd), the signals acquired during short-sequence transmissions are analyzed to accurately identify obstacles within a 10cm to 30cm blind zone.

[0177] Optionally, the main control module is also used to output aftershock cancellation commands; the transducer module is also used to transmit cancellation waves according to the aftershock cancellation commands; the system also includes: a cancellation wave transmitting module, connected to the main control module and the transducer module, used to generate cancellation waves that are in the same phase and opposite in direction as the aftershocks according to the cancellation wave power, phase and delay information output by the main control module, and drive the transducer module to transmit cancellation waves until the peak amplitude of the aftershock signal is reduced to below a preset threshold.

[0178] In this embodiment, after threshold calibration and before real-time detection, an aftershock cancellation mechanism is introduced. Through closed-loop control, a cancellation wave with the same phase and opposite direction as the aftershock signal is actively emitted to cancel the energy of the aftershock signal, thereby further reducing the interference of aftershocks on blind zone detection.

[0179] In addition to outputting detection commands, calibration commands, and long and short sequence transmission commands, the main control module is also used to output aftershock cancellation commands to initiate the aftershock cancellation closed-loop control process. Simultaneously, based on the detection results during the cancellation process, the main control module outputs the power, phase, and delay information of the cancellation wave for use by the cancellation wave transmitting module.

[0180] In addition to transmitting probe pulses and receiving echo signals, the transducer module is also used to transmit cancellation waves according to aftershock cancellation commands. These cancellation waves have the same frequency (e.g., 40kHz), phase, and opposite direction as the aftershock signal, canceling out the aftershock energy through the principle of acoustic interference.

[0181] The core functions of the wave suppression transmission module include: dynamically adjusting the drive current for wave suppression based on the amplitude and saturation of the aftershock signal to avoid introducing larger, reverse aftershocks; determining the positions of wave peaks, troughs, and zero points by analyzing the period of the aftershock signal to ensure that the wave suppression is in phase with the aftershock; and adding a delay to the phase-corresponding transmission time based on the time loss of ultrasonic sampling, reception, and wave suppression transmission (total delay of approximately 0.1ms) to ensure that the wave suppression is transmitted at the correct time.

[0182] For example, the anti-canceling transmission module may consist of a D / A converter chip and a power amplifier circuit. The D / A converter has a 16-bit resolution and an output voltage range of 0 to 5V, used to amplify the anti-canceling signal to the required power.

[0183] To facilitate understanding of aftershock closed-loop control, its workflow is explained in detail below:

[0184] 1. Threshold Calibration Stage

[0185] Completely consistent with the aforementioned threshold calibration method, the benchmark dataset {envp(t), val(n), pos(n),fa_rd} was collected and stored.

[0186] 2. Aftershock cancellation closed-loop control

[0187] Start the system after threshold calibration and before real-time detection, and continue running until the aftershock signal amplitude decreases below the preset threshold. The specific steps are as follows:

[0188] (1) Aftershock signal detection: The aftershock signal is detected in real time by the transducer module after the threshold calibration and during the real-time detection process. The acquisition time is 5ms and the sampling frequency is 1MHz.

[0189] (2) Measurement of amplitude, phase, and delay information:

[0190] Transmission power determination: ADC data for one drive cycle is collected as a signal amplitude reference. If the proportion of ADC saturated data exceeds 20%, and the slope from saturation to unsaturation is ≥0.5V / ms, the suppression drive current will be reduced to 30mA (approximately 1 / 5 of the original probe pulse drive current) to avoid introducing a larger reverse aftershock.

[0191] Phase determination: The period of the aftershock signal is recorded by a timer, and the positions of the wave peaks, troughs and zero points within one aftershock period are analyzed to determine the phase of the aftershock signal and ensure that the phase of the wave suppression is consistent with that of the aftershock.

[0192] Delay information determination: Considering the time loss of ultrasonic sampling, reception and the effect of wave cancellation transmission (total delay is about 0.1ms), a delay of 0.1ms is added to the transmission time corresponding to the phase.

[0193] (3) Counter-wave transmission: Through the transmission module, the power of the transmission is 30mA, the phase is consistent with the aftershock, and the direction is opposite. The counter-wave frequency is consistent with the detection pulse frequency (40kHz).

[0194] (4) Closed-loop iteration: Detect the ultrasonic signal after cancellation again, repeat steps (1) to (3) above until the peak amplitude of the aftershock signal is reduced to less than 10% of the peak value of envp(t) in the threshold calibration stage, stop the cancellation wave transmission, and enter the real-time detection stage.

[0195] 3. Real-time detection phase

[0196] Blind zone target detection is achieved by utilizing dynamic thresholds, characteristic peaks, and the first effective descent point fa_rd.

[0197] Through the closed-loop control of aftershock cancellation described above, the amplitude of aftershock signals is effectively reduced, with an aftershock attenuation rate exceeding 85%. This makes it easier to identify weak obstacle signals within the blind zone (10cm to 30cm), further increasing the detection rate to 99% and reducing the false alarm rate to below 0.5%. Aftershock cancellation works in conjunction with dynamic thresholds and the first effective descent point: the significantly reduced aftershock amplitude after cancellation allows for a more lenient dynamic threshold setting and more accurate identification of the first effective descent point, further improving the reliability of blind zone detection.

[0198] In one possible implementation, combining long and short sequences, gain control, and aftershock cancellation achieves superior blind zone detection performance. Through the synergistic effect of these three technical features, it simultaneously addresses the difficulties in close-range blind zone detection, insufficient stability in long-range detection, and false alarms caused by aftershock interference. Specifically:

[0199] 1. Threshold Calibration Stage

[0200] Complete the collection and storage of the benchmark dataset {envp(t), val(n), pos(n), fa_rd}.

[0201] 2. Aftershock cancellation closed-loop control

[0202] By using closed-loop control to eliminate the aftershock signal, the peak amplitude of the aftershock signal is reduced to less than 10% of the peak value of envp(t), effectively suppressing aftershock interference.

[0203] 3. Alternating long and short sequence emission and gain control

[0204] Set the following operating parameters:

[0205] Short sequence: using fixed frequency mode (fix mode), 8 pulses, detection range of 10cm to 30cm, drive current of 50mA, and receiving gain of 20dB.

[0206] Long sequence: Chirp mode is used, with 96 pulses, a detection range of 25cm to 500cm, a drive current of 150mA, and a receiving gain of 40dB.

[0207] After the system starts up, it alternately transmits short and long sequences. Once the long sequence detects a target, it immediately switches to the short sequence to continue detection.

[0208] 4. Real-time detection and judgment

[0209] By combining dynamic thresholds, characteristic peak matching, and the first effective descent point fa_rd, signals acquired from both long and short sequences are analyzed.

[0210] Short sequence: Focuses on detecting blind zones of 10cm to 30cm, with low gain, low current and dynamic threshold to accurately identify nearby obstacles.

[0211] Long sequence: Focuses on detecting targets at a distance of 25cm to 500cm. High gain and high current combined with characteristic peak matching ensure the stability of long-distance detection.

[0212] The three technical features work synergistically: aftershock cancellation further reduces aftershock interference; gain control of long and short sequences avoids ADC saturation; and dynamic threshold and feature peak matching improve detection accuracy. Combined, these three features enable full-range, high-precision target detection.

[0213] In this embodiment, the three technical features work synergistically to achieve a blind zone (10cm to 30cm) detection rate of over 99.5%, a false alarm rate of less than 0.3%, improved long-distance detection stability by 30%, maintain a data rate of 28Hz to meet real-time requirements, and improve energy utilization by 40%. Furthermore, it can adapt to various complex scenarios such as individual sensor differences, circuit parameter drift, temperature changes, and variations in installation conditions, fully meeting the practical application needs of automotive reversing radar.

[0214] Optionally, it may also include at least one of the following modules:

[0215] The storage module, connected to the main control module, is used to store the benchmark dataset and detection logs; the benchmark dataset includes dynamic comparison thresholds, feature peak parameters, and the first effective descent point.

[0216] The power module connects to each module and is used to supply power to each module;

[0217] The interface module connects to the main control module and is used to communicate with external devices and output test results.

[0218] For example, the storage module can use an SD card and a Flash chip. The SD card is used to store the benchmark dataset {envp(t), val(n), pos(n), fa_rd}, detection logs, and system parameters from the threshold calibration phase; the Flash chip is used to store program code and commonly used configuration parameters to ensure the stability of data storage and the speed of retrieval, thereby meeting the storage requirements of the benchmark dataset.

[0219] For example, the power module can use a DC-DC step-down chip, with an input voltage of the automotive 12V power supply and output voltages of 5V and 3.3V to power each module respectively. This module incorporates overcurrent and overvoltage protection circuits to ensure stable operation of the hardware in the complex power supply environment of a vehicle, adapting to automotive application scenarios.

[0220] For example, the interface module may include a CAN bus interface and a GPIO interface. The CAN bus interface is used to communicate with the vehicle's central control system, outputting information such as obstacle distance and detection status, enabling linkage with the vehicle's onboard system. The GPIO interface is used to connect indicator lights and alarm devices; when an obstacle is detected in the blind spot, the indicator lights illuminate and the alarm devices sound, thereby improving safety.

[0221] The following section provides a detailed explanation of the hardware device's workflow, taking into account all the aforementioned technical solutions.

[0222] 1. Power-on initialization

[0223] After the car starts, the power module supplies power to all hardware modules. The main control module completes system initialization, performs a self-test via the GPIO interface, and after confirming that all modules are working properly, triggers the threshold calibration process.

[0224] 2. Threshold Calibration

[0225] The main control module controls the transducer module to emit detection pulses. The signal acquisition module acquires clean aftershock signals and transmits them to the signal processing module. The signal processing module performs envelope processing, characteristic peak identification, and first effective descent point determination on the signals to obtain a benchmark dataset. This dataset is stored in the storage module, thus completing the threshold calibration.

[0226] 3. Aftershock cancellation

[0227] The main control module initiates aftershock cancellation closed-loop control. The signal acquisition module detects aftershock signals, and the signal processing module analyzes the signals to obtain the power, phase, and delay information of the aftershocks. Based on this information, the main control module controls the aftershock transmission module to transmit aftershocks, and repeats the process iteratively until the aftershock signal meets the detection requirements, thus completing aftershock cancellation.

[0228] 4. Real-time detection

[0229] The main control module controls the gain control module to adjust the drive current and receive gain according to the type of long and short sequences. The transducer module then alternately transmits short and long sequences accordingly. The signal acquisition module acquires real-time mixed signals, and the signal processing module uses a benchmark dataset to determine the obstacle peak value. Based on the detection results, the main control module outputs information through the interface module. When a target is detected using a long sequence, the system immediately switches to a short sequence to continue detection, thus completing the long and short sequence detection and gain control.

[0230] The ultrasonic sensor system provided in this application can be widely used in the following scenarios:

[0231] Vehicle-mounted ultrasonic radar system: used for reversing radar, forward collision warning, blind spot detection, automatic parking, etc.;

[0232] Robot navigation and obstacle avoidance system: used for obstacle detection and avoidance in robotic vacuum cleaners, service robots, drones, etc.

[0233] Industrial automation systems: used for liquid level detection, material level detection, object positioning, distance measurement, etc.

[0234] Smart home systems: used for human body sensing, distance triggering, smart door locks, etc.

[0235] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. An obstacle detection method, characterized in that, include: Acquire real-time mixed signals including obstacle echoes; When there is a target peak value in the envelope of the real-time mixed signal that exceeds the dynamic comparison threshold, it is determined whether the target peak value is an obstacle peak value based on the characteristic peak parameters and the first effective descent point. The dynamic comparison threshold is determined based on the peak envelope sequence of the pure aftershock signal; the first effective descent point is the first position point in the peak envelope sequence of the pure aftershock signal that meets the following conditions: the number of consecutive descents is greater than or equal to K, and the cumulative descent ratio is greater than or equal to R; where K is an integer greater than or equal to 2, and R is a real number greater than 0 and less than 1.

2. The method according to claim 1, characterized in that, The step of determining whether the target peak value is an obstacle peak value based on the characteristic peak parameters and the first effective descent point includes: When the time position of the target peak is after the first effective descent point, and the magnitude of the target peak exceeds the corresponding position threshold value in the dynamic comparison threshold by a proportion greater than a preset proportion threshold, the target peak is determined to be an obstacle peak. When the time position of the target peak is before the first effective descent point, and the deviation between the amplitude of the target peak and the amplitude of the corresponding characteristic peak in the characteristic peak parameter is less than or equal to a preset deviation threshold, the target peak is determined to be a non-obstacle peak.

3. The method according to claim 1, characterized in that, The dynamic comparison threshold is updated by matching the peak envelope sequence of the pure aftershock signal with the aftershock characteristics of the current ultrasonic radar in real time, and is used to separate obstacle signals superimposed on the aftershock.

4. The method according to any one of claims 1-3, characterized in that, It also includes a threshold calibration step, which includes: The peak envelope sequence is obtained by performing envelope processing on the pure aftershock signal collected under unobstructed conditions. Identify multiple characteristic peaks in the peak envelope sequence, and record the height and corresponding time position of each characteristic peak as the characteristic peak parameter; In the peak envelope sequence, the starting time point at which the aftershock signal decays to a reliable threshold is determined as the first effective descent point.

5. The method according to claim 4, characterized in that, The envelope processing of the clean aftershock signal acquired under unobstructed conditions to obtain the peak envelope sequence includes: The absolute value of the pure aftershock signal is taken to obtain the absolute value signal; The absolute value signal is subjected to a sliding window maximum value calculation to generate a peak envelope sequence with the same length as the pure aftershock signal, wherein the sliding window maximum value calculation is to calculate the maximum amplitude of multiple sampling points before and after each sampling point.

6. The method according to claim 4, characterized in that, The process of identifying multiple characteristic peaks in the peak envelope sequence and recording the height and corresponding time position of each characteristic peak includes: In the peak envelope sequence, the height and time position corresponding to all peaks that meet the preset peak conditions are identified and recorded.

7. An ultrasonic sensor chip, characterized in that, Used to perform the method according to any one of claims 1-6.

8. An ultrasonic sensor system, characterized in that, Includes the ultrasonic sensor chip as described in claim 7; The main control module is used to output detection and calibration commands; The transducer module, connected to the main control module, is used to transmit detection pulses according to the detection command and receive real-time mixed signals; it is also used to transmit detection pulses according to the calibration command and receive pure aftershock signals. The signal acquisition module, connected to the transducer module and the ultrasonic sensor chip, is used to acquire the real-time mixed signal and pure aftershock signal received by the transducer module and transmit them to the ultrasonic sensor chip so that the ultrasonic sensor chip can perform obstacle detection and threshold calibration.

9. The system according to claim 8, characterized in that, The main control module is also used to output long and short sequence transmission commands; the transducer module is also used to alternately transmit short sequence pulses and long sequence pulses according to the long and short sequence transmission commands; The system also includes: A gain control module, connected to the main control module and the transducer module, is used to set a first driving current and a first receiving gain when transmitting a short sequence, and to set a second driving current and a second receiving gain when transmitting a long sequence, according to the long and short sequence transmission commands. Wherein, the first driving current is less than the second driving current, and the first receiving gain is less than the second receiving gain.

10. The system according to claim 9, characterized in that, The main control module is also used to control the transducer module to stop transmitting the long sequence and switch to transmitting the short sequence if the ultrasonic sensor chip detects an obstacle before the end of the long sequence working cycle during the transmission of the long sequence.

11. The system according to claim 9, characterized in that, The main control module is also used to output aftershock cancellation commands; the transducer module is also used to transmit cancellation waves according to the aftershock cancellation commands. The system also includes: The anti-vibration transmission module is connected to the main control module and the transducer module. It is used to generate an anti-vibration wave that is in the same phase and opposite in direction as the aftershock based on the anti-vibration power, phase and delay information output by the main control module, and drive the transducer module to transmit the anti-vibration wave until the peak amplitude of the aftershock signal is reduced to below a preset threshold.

12. The system according to any one of claims 8-11, characterized in that, It also includes at least one of the following modules: A storage module, connected to the main control module, is used to store a benchmark dataset and detection logs; wherein, the benchmark dataset includes a dynamic comparison threshold, feature peak parameters, and the first effective descent point; The power module connects to each module and is used to supply power to each module; The interface module is connected to the main control module and is used to communicate with external devices and output test results.

Images