Sensor, method implemented by sensor controller, and sensor controller

Abstract

An ultrasonic sensor having a self-cleaning function, a sensor controller, and a sensor control method are provided. An exemplary method includes driving a piezoelectric transducer to generate a short train of acoustic pulses for obstacle detection or distance measurement; obtaining a received signal to monitor reflections of the short acoustic pulse train; and operating to clean the sensor by driving the piezoelectric transducer to generate a long acoustic pulse train at the resonant frequency of the piezoelectric transducer. The method may be implemented by a sensor controller having: a transmitter configured to drive a piezoelectric transducer; a receiver coupled to the piezoelectric transducer and the microphone to detect a reflection of the acoustic pulse train within a measurement interval associated with the acoustic pulse train; and a microcontroller configured to control a length of the acoustic pulse train. The sensor controller may be incorporated into a sensor that further includes a piezoelectric transducer, and optionally one or more microphones.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of U.S. Provisional Application No. 63 / 626,354, entitled “Weather immune ultrasonic sensor,” filed on January 29, 2024 by inventors Marek Hustava and Jiri Kantor. The entire contents of that provisional application are hereby incorporated by reference.

[0003] This application also claims the benefit of commonly owned U.S. patent application Ser. No. 18 / 468,207, filed Sep. 15, 2023, by inventor Marek Hustava, entitled “Sensor array module for an advanced driver assistance system,” which claims the benefit of U.S. provisional application Ser. No. 63 / 482,193, filed Jan. 30, 2023. That commonly owned application is hereby incorporated by reference herein. Technical Field

[0004] The present application relates generally to sensors, and more particularly to sensors, methods implemented by a sensor controller of the sensors, and sensor controllers. Background Art

[0005] As explained, for example, in commonly owned U.S. Patent 11,269,067, entitled "Response-based determination of piezoelectric transducer state," filed by inventor Jiri Kutej et al. on February 5, 2018, the entire contents of which are hereby incorporated by reference, modern automobiles are equipped with a considerable number and variety of sensors. For example, cars are now commonly equipped with arrays of ultrasonic sensors to monitor the distance between the car and any nearby people, pets, vehicles, or obstacles. Due to environmental "noise" and safety concerns, each of the sensors may be required to provide dozens of measurements per second while the car is in motion. It is important for such sensor arrays to perform reliably or to alert the operator if their performance is questionable.

[0006] As the number of sensors increases, the incidence of sensor failures and the importance of correctly diagnosing such failures also increase, enabling an operator to be alerted and, if necessary, appropriate actions can be taken to correct or adjust the failure. Exemplary actions can include: repairing a damaged or disconnected transducer, washing the vehicle to remove dirt or grime from the sensor surface, exercising caution when the transducer is damaged by snow or rain, and ignoring transient noise events when the sensor is otherwise operating correctly. The previously mentioned U.S. Patent 11,269,067 describes techniques for diagnosing such failures.

[0007] Failures such as sensor contamination / blockage, e.g., where the sensor lens, antenna, surface, membrane, or port holes are coated with dirt, grime, snow, or ice. For some sensors, internal condensation can also be a problem. Ideal performance only occurs in the case of a clean or unblocked sensor. Thus, various sensor cleaning techniques have been developed, but typically require a significant additional manufacturing cost. Examples include: jets of high-pressure air, water, or mist; ultrasonic cleaning; heating; and mechanisms to prevent contamination by hiding the sensor when not in use.

[0008] Perhaps due to the increased manufacturing cost, existing ultrasonic sensing systems generally do not have such sensor cleaning techniques. At least some systems detect when a given ultrasonic sensor is blocked and ignore the associated sensor measurements while the sensor remains blocked. The sensor may continue to operate, but internal diagnostics monitor for sensor damage. The sensor controller attaches the diagnostic result to the measurement data, providing a flag or marker to indicate whether the sensor is blocked. The electronic control unit (ECU) omits unreliable measurement data from the distance measurement process to ensure that the system is not deceived by invalid measurement data. Of course, omitting measurements reduces the overall system performance. Thus, susceptibility to dirt and other blockages is a limiting factor for ultrasonic sensors. SUMMARY OF THE INVENTION

[0009] Accordingly, the present disclosure provides various sensors, sensor controllers, and sensor control methods having a self-cleaning function. An exemplary method includes: driving a piezoelectric transducer to generate a short burst of acoustic pulses for obstacle detection or distance measurement; obtaining a received signal to monitor the reflection of the short burst of acoustic pulses; and operating to clean the sensor by driving the piezoelectric transducer to generate a long burst of acoustic pulses at the resonant frequency of the piezoelectric transducer. The method can be implemented by a sensor controller having: a transmitter configured to drive the piezoelectric transducer; a receiver coupled to at least one of the piezoelectric transducer and a microphone to detect the reflection of the burst of acoustic pulses during a measurement interval associated with the burst of acoustic pulses; and a microcontroller configured to control the length of the burst of acoustic pulses. The sensor controller can be incorporated into a sensor that further includes a piezoelectric transducer and optionally includes one or more microphones.

[0010] The foregoing method, sensor controller, and sensor can be employed individually or together, and they can also be used in any suitable combination with one or more of the following optional features: 1. The longer burst length completely occupies the measurement interval with driving vibrations or residual ringing vibrations. 2. The measurement interval is greater than 50 ms. 3. When operating to detect reflections, the microcontroller sets the length of the burst of acoustic pulses to be less than or equal to 4 ms. 4. When operating to clean the sensor, the transmitter drives the transducer at the resonant frequency within the length of the burst of acoustic pulses. 5. When operating to detect reflections, the transmitter drives the transducer at a frequency above or below the resonant frequency. 6. When operating to clean the sensor, the transmitter drives the transducer with a drive current greater than or equal to 400 milliamperes. 7. When operating to clean the sensor, the microcontroller employs the longer burst length in a plurality of consecutive measurement intervals. 8. When operating to clean the sensor, the microcontroller repeatedly employs the longer burst length until the sensor controller reaches a target temperature. 9. When operating to clean the sensor, after the sensor controller returns from the target temperature to a lower temperature threshold, the microcontroller resumes employing the longer burst length. 10. The sensor controller heats the microphone port to remove ice. 11. When operating to clean the sensor, the sensor controller heats the piezoelectric transducer without relying on a heater. 12. The microcontroller is configured to monitor sensor damage and is configured to automatically start operating to clean the sensor after detecting sensor damage. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 A top view of an exemplary vehicle equipped with a parking assist sensor.

[0012] Figure 2 A block diagram of an exemplary parking assist system.

[0013] Figure 3 is a perspective view of an exemplary parking assist sensor.

[0014] Figure 4 is a block diagram of an exemplary parking assist sensor.

[0015] Figures 5A to 5B The graphs are for comparing short sound pulse trains and long sound pulse trains.

[0016] Figure 6 is a flow chart of an exemplary sensing method. DETAILED DESCRIPTION

[0017] The following description and drawings are provided for the purpose of explanation, not limitation of the present disclosure. In other words, they provide a basis for one of ordinary skill in the art to identify and understand all modifications, equivalents, and alternatives that fall within the scope of the claims, and do not impose any implicit limitations on the scope of the claims.

[0018] Figure 1 An exemplary vehicle 102 is shown equipped with a set of ultrasonic parking assist sensors 104. The number and configuration of sensors in the sensor arrangement vary, and it is not uncommon to have four sensors on each bumper, with two additional sensors on each side acting as blind spot detectors. The vehicle can employ this sensor arrangement to detect and measure distances to objects within various detection zones, using the sensors for both individual measurements and collaborative measurements (e.g., triangulation, multi-receiver).

[0019] Ultrasonic sensors are transceivers, meaning each sensor can transmit and receive ultrasonic sound pulses. The transmitted sound pulse propagates outward from the vehicle until it encounters and reflects from an object or some other form of acoustic impedance mismatch. The reflected sound pulse returns to the vehicle as an "echo" of the transmitted pulse. The time between the transmitted sound pulse and the received echo indicates the distance to the reflection point. In many systems, only one sensor transmits at a time, but all sensors can be configured to measure the resulting echo. However, by using orthogonal waveforms or transmitting to non-overlapping detection areas, multiple simultaneous transmissions can be supported.

[0020] Figure 2An electronic control unit (ECU) 202 is shown coupled to various ultrasonic sensors 204 as the center of a star topology. Of course, other topologies, including serial, parallel, and hierarchical (tree) topologies are also suitable, and it is contemplated that these topologies will be used in accordance with the principles disclosed herein. To provide automatic parking assistance, the ECU 202 may be further coupled to a set of actuators, such as a turn signal actuator 206, a steering actuator 208, a brake actuator 210, and a throttle actuator 212. The ECU 202 may be further coupled to a user interactive interface 214 to accept user input and provide a display of various measurements and system status. Using the interface, sensors, and actuators, the ECU 202 may provide automatic parking, assisted parking, lane change assistance, obstacle and blind spot detection, and other desired features. Although Figure 2 Only one ECU is shown in the figure, but many systems may distribute ECU tasks among multiple interconnected ECUs.

[0021] Figure 3 is an isometric view of an exemplary sensor 104 having an ultrasonic transducer 302, which may include a piezoelectric element with a support member to configure the transducer's resonant frequency and resonant peak width. In the case of a concealed sensor, the ultrasonic transducer 302 may contact the backside of an overlying surface. Alternatively, the ultrasonic transducer 302 may be positioned behind an opening or acoustic window in the overlying surface. Where aesthetics permit, the ultrasonic transducer 302 may be mounted externally or otherwise exposed to view. The ultrasonic transducer 302 generates a train of acoustic pulses when driven and may be configured to sense reflections of the acoustic pulses from objects within a sensing range.

[0022] The sensor 104 also includes a substrate 304, such as a PCB (printed circuit board) or an MLO (multi-layer organic) laminate, which has control electronics for the sensor components. A sensor controller 306 and one or more optional microphones 308 are mounted on the substrate 304. The microphones 308 can be a bare die having a microelectromechanical system (MEMS) structure for acoustic sensing. To enhance sensitivity, the microphones 308 can be aligned with corresponding holes 310 in the sensor housing and, if the sensor is hidden, with unobtrusive openings or acoustic windows in the overlying surface. Alternatively, the sensor 104 can position the microphones 308 near the edge of the overlying surface, or on one side of the surface around the corners of the ultrasonic transducer 302, where the microphones can be sensitive to acoustic signals without being obtrusive.

[0023] In some contemplated sensor configurations, at least one MEMS microphone 308 is vertically displaced from the ultrasonic transducer 302 and can be operatively coupled with the transducer 302 to effect direction-of-arrival (DoA) detection from which the elevation angle of a reflector can be estimated. Alternatively, two vertically spaced apart microphones 308 can be used for this purpose. In a similar manner, a second MEMS microphone can be horizontally displaced from the ultrasonic transducer or from the first MEMS microphone to enable estimation of the azimuth angle of the reflector. Some contemplated sensor assemblies have multiple MEMS microphones that provide horizontal and vertical displacement relative to the transducer or relative to each other, enabling estimation of both the elevation angle and the azimuth angle.

[0024] Figure 4 is a block diagram of an exemplary sensor controller 306. It employs IO (input and output) pins that are multiplexed to a selected one of various IO protocol modules. Modules are shown for I2C (Inter-Integrated Circuit), SPI (Serial Peripheral Interface), and DSI3 (Distributed System Interface, 3rd Generation) communication bus protocols, but these are merely non-limiting examples. Other contemplated protocols include those set forth in the LIN, CAN, and SENT standards. The IO bus 402 couples the various IO protocol modules to an embedded microcontroller (μC). An AMBA (Advanced Microcontroller Bus Architecture) bus 404 (such as, for example, AHB-Lite (Advanced High-Performance Bus, Lite)) can be used to couple the embedded microprocessor to various other functional blocks of the sensor controller 306. Such blocks can include, for example, non-volatile firmware (FW) memory, volatile working memory (MEM), digital signal processor (DSP), power-on reset (POR) module, an ADC (analog-to-digital converter) for each microphone signal receiver (RX M ), an additional ADC for the transducer signal receiver (RX T ), a drive current transmitter (TX I ), and a drive voltage transmitter (TX V ). (The transmitters respectively generate drive signals for the ultrasonic transducer as a controlled current or a controlled voltage.) A multiplexer can couple a selected one of the transmitters to an output pin for driving the ultrasonic transducer 302. Figure 4 An embedded oscillator (OSC) is further shown that generates one or more clock signals for the various other blocks, optionally including the carrier frequency for the transmitters and receivers. The POR module can control the distribution of the clock signals to enable a sleep mode for unneeded components and other power saving features.

[0025] The IO pins couple the sensor controller 306 directly or indirectly (e.g., via a DSI3 bus master) to an ECU (electronic control unit). The ECU can transmit commands to the various sensor controllers to, for example, set the values of various configuration parameters of the sensor, initiate the transmission of an acoustic burst, and collect signal data or other measurements. The microprocessor in each sensor controller operates according to firmware and stored configuration parameters to interpret commands from the ECU and perform appropriate operations, including the transmission of acoustic bursts and the reception of acoustic signals.

[0026] In various implementations, a chirp modulated signal is used, such as a linear frequency modulation ("LFM") chirp. A chirp is a pulse that changes frequency during transmission. An up-chirp is a single pulse whose frequency increases during transmission, and a down-chirp is a single pulse whose frequency decreases during transmission. For clarity, the examples used herein will consider a linear increase or decrease, however, in various implementations, the increase or decrease is not linear. The chirped echo can be compressed in the correlator without introducing much or any correlated noise. Therefore, peak detection of the echo is facilitated without reducing the time resolution. In addition, the LFM chirp can withstand Doppler shift without any or minimal increase in correlated noise. The LFM chirp can be used as a transmit pulse for measuring the distance and direction to obstacles or objects within the sensing range of the sensor system.

[0027] In other embodiments, an AM (amplitude modulated) signal is used, such as a shaped pulse of a fixed frequency carrier. The AM signaling mode can enable the use of shorter pulse trains (e.g., approximately 200 to 300 microseconds), thereby reducing transmission time and increasing sensitivity to nearby obstacles. Other embodiments may employ pulses with a modulated carrier (e.g., using binary phase shift keying (BPSK) modulation). For clarity, the term "burst" as used herein refers to an AM (fixed frequency), BPSK (modulated), or chirped (swept frequency) pulse, which can be one of a series of pulse trains generated by driving a piezoelectric element or other ultrasonic transducer. Chirped modulated pulses can have a duration longer than a typical AM pulse, for example, greater than 1 millisecond, such as in the range of 2 to 3 milliseconds. It is noted here that the burst length can vary, with shorter bursts being used to facilitate detection of nearby obstacles and longer bursts being used to increase the burst energy (and echo energy) for obstacles that are farther away. The pulse train length for detecting nearby obstacles may be half or possibly a quarter of the pulse train length for obstacles further away.The sensor can be switched between modes for different detection distances.

[0028] While systematically varying the characteristic frequency (e.g., the starting frequency, or equivalently, the center or ending frequency) of the chirped modulated pulses in the sequence is considered particularly useful, such frequency variation can also be applied to the carrier frequency of the AM pulses in the sequence. The frequency variation of each pulse can be expressed as a frequency shift from a nominal characteristic frequency (e.g., the nominal starting frequency or the nominal carrier frequency).

[0029] To transmit an acoustic pulse train, the microprocessor instructs the selected transmitter to drive an output pin for the ultrasonic transducer, which is coupled to the piezoelectric element PZ. A transformer and / or resonant tuning network may be provided to amplify the voltage and control the resonant frequency of the transducer. The transmitter may receive a carrier frequency signal having a nominal frequency of, for example, 50 kHz, from an oscillator. The transmitter may use this carrier frequency signal to generate a series of AM (amplitude modulated) or chirped pulses, each corresponding to an acoustic pulse train. An example of a chirped pulse is a pulse whose frequency sweeps upward from 7 kHz below the carrier frequency to 7 kHz above the carrier frequency (up-chirp). Alternatively, down-chirp may be employed, in which the frequency sweeps linearly downward rather than upward. In some contemplated implementations, the transmitter provides a customized frequency offset pattern to the acoustic pulse train to serve as a unique signature for the ultrasonic transducer. Alternatively, frequency offsetting may be employed to support upper or lower sideband operation (above and below the resonant frequency, respectively), thereby enabling multiple transducers to transmit simultaneously without interfering with each other.

[0030] To receive the acoustic signal, the microprocessor instructs one or two ADCs to digitize the electrical receive signal from the MEMS microphone 308 and / or the piezoelectric element 302. The digitized signal can be provided directly to the DSP for real-time processing, or buffered in memory for subsequent processing by the DSP or ECU. To reduce IO bandwidth requirements, the DSP can implement data compression to reduce the number of bits required to represent the ZIF IQ data or to represent the magnitude of the baseband signal. To further reduce bandwidth requirements, the DSP can perform on-chip processing to perform peak detection and distance estimation. Various suitable processing techniques for detecting reflections of acoustic pulse trains are known in the art, including commonly owned U.S. Patent Publication 2024 / 0069192, "Motion-compensated distance sensing with concurrent up-chirp down-chirp waveforms," which is incorporated herein by reference.

[0031] Since the received electrical signal is usually in the millivolt or microvolt range, the receiver RX M RX TAn amplifier may be included to buffer and amplify the signal from the receiving terminal. An analog or digital mixer may be included to down-convert the received signal to baseband for further filtering and processing by the DSP. In one specific implementation, the mixer is an in-phase / quadrature (I / Q) digital mixer, giving zero intermediate frequency (ZIF) IQ data as its output. (Although the term "ZIF" is used herein, the down-converted signal may in practice be a low intermediate frequency or "near baseband" signal.)

[0032] The DSP applies programmable methods to acquire the received signal and detect any echoes and measure their parameters such as time of flight (ToF), direction of arrival (DoA), duration, and peak amplitude. Such methods may employ threshold comparison, minimum spacing, peak detection, zero-crossing detection and counting, noise level determination, and other customizable techniques tailored for improved reliability and accuracy. It is noted that the peak detection process itself has variants, some of which perform rising edge detection, falling edge detection, or peak maximum detection. The processing of nearby obstacle detection may be performed entirely in the controller 306, or may be shared with or delegated to the ECU or host processor, which receives certain data via the communication bus as previously described.

[0033] Figure 5A is a graph showing the vibration VIBR of the piezoelectric element during a typical measurement interval 501. The measurement interval 501 may optionally be separated from the previous and subsequent measurement intervals 503 by measurement intervals 500, 502. The measurement intervals may be periodic, repeating at a fixed cycle period as long as the sensor is enabled. Alternatively, each measurement interval may be initiated as needed by, for example, the ECU. Each measurement interval begins with the emission of an acoustic pulse train 510 having a drive portion 512 and a ringing portion 514, in the drive portion, the transmitter supplies a drive current or drive voltage to the ultrasonic transducer, and in the ringing portion, the ultrasonic transducer undergoes residual vibration. The measurement interval lasts for a predetermined time, which may be set or adjusted based on the desired sensing range. If the acoustic pulse train is reflected from an object within the sensing range, the energy of the reflected acoustic pulse train causes the ultrasonic transducer to vibrate with a delayed echo 516. The delay of the echo corresponds to the distance of the object. Multiple objects may cause multiple echoes, each echo having a delay corresponding to the distance of the corresponding object.

[0034] It should be noted here that Figure 5A and Figure 5BNot drawn to scale. In a typical measurement interval, the drive portion 512 of the acoustic burst may extend for 400 microseconds (about 20 cycles of a 50 kHz signal) for an AM signaling mode, or 2.5 milliseconds (about 125 cycles) for a chirp signaling mode. The measurement interval may be, for example, 30 to 100 milliseconds, while the optional inter-measurement interval may be a few milliseconds or shorter, or may be long enough to avoid operation during measurement intervals employed by other sensors in the system. With active damping, the ringing portion 514 of the acoustic burst may not exceed a few cycles of the resonant frequency. The echo amplitude may be one or more orders of magnitude smaller than the acoustic burst amplitude.

[0035] Figure 5B Figure 5 is a graph showing the vibration of the VIBR or piezoelectric element during a measurement interval when the sensor is in cleaning mode. Acoustic burst 520 is initiated at the beginning of measurement interval 501 and continues for the entire duration of the measurement interval. In some cases, the driven portion of the acoustic burst can be maintained throughout the measurement interval, such that the ringing portion occurs during the inter-measurement interval. In other implementations, the driven portion is terminated shortly before the end of measurement interval 501, allowing the residual vibration to drop below a threshold before the end of the measurement interval, thereby maximizing the sensor's readiness to return to normal operation in the subsequent measurement interval. A similar readiness can be achieved by further shortening the acoustic burst to approximately 50% of the measurement interval to allow the echo energy to dissipate before the subsequent measurement interval, but such shortening may not be necessary in systems employing sideband or chirp signaling. The sensor controller can maximize the amplitude of the piezoelectric element during acoustic burst 520 to maximize the displacement of any dust, dirt, mud, water, or other obstructions that may impair sensor operation. As part of maximizing the vibration, the sensor controller may apply a maximum rated drive current or drive voltage level and may operate at the resonant frequency of the ultrasonic transducer when in the cleaning mode.

[0036] When in cleaning mode, the sensor controller can further operate to maximize the energy dissipation and heating of the sensor within each measurement interval. Experiments have shown that this is achieved by maximizing the duration of the acoustic pulse train and the associated current in the transmitter driving the ultrasonic transducer. When the drive current was approximately 400mA, the measurement interval repetition in measurement mode was 50ms, and the acoustic chirp was 2.5ms, the sensor controller achieved a temperature rise of 7°C after 20 seconds, which translated into a temperature rise of 2°C on the surface of the sealed sensor assembly. This is a proof of concept. In subsequent tests, using 300mA, 400mA, and 500mA drive currents and a 50ms acoustic pulse train at the resonant frequency, after 20 seconds, the temperature of the sensor controller increased by 26°C, 27°C, and 32°C above the ambient temperature (25°C), respectively, which translated into surface temperature rises of 3°C, 5°C, and 7°C, respectively. Such energy dissipation within the controller helps melt and remove ice and snow from the sensor surface without the need for dedicated heating elements.

[0037] In sensors with a close proximity microphone port, piezoelectric element vibration and sensor controller heating can operate to clear blockages from the microphone port and from the piezoelectric element surface. It may be preferable to limit the acoustic burst 520 to the associated measurement interval to facilitate autonomous switching between cleaning mode and measurement mode.

[0038] Figure 6 6 is a flow chart of an exemplary sensing method that can be implemented by a sensor controller. In blocks 602 and 604, the sensor controller performs noise level detection while waiting for the initiation of a measurement interval. The measurement interval can be initiated by, for example, a command from an ECU, and / or periodically by, for example, an internal timer. When the measurement interval is initiated, the sensor controller checks in block 606 whether the cleaning mode is activated. In some implementations, the sensor controller can set a bit in a status register to indicate when the cleaning mode is activated, and the bit can be cleared when the sensor is in normal operating mode. If the sensor is in normal operating mode, the sensor controller can proceed to block 612 to generate a short acoustic pulse train for object detection and distance measurement.

[0039] Otherwise, in block 608, the sensor controller performs a temperature check. A dedicated thermocouple or other external temperature sensor may be used, but in at least some contemplated embodiments, the sensor controller measures the internal PN junction temperature. The junction temperature may be measured indirectly by measuring the voltage across a current-biased diode or transistor, or by measuring the current flowing through a voltage-biased diode or transistor. Various temperature-compensated voltage and current reference circuits are known in the literature and can be readily adapted to provide temperature measurement signals derived from their temperature-compensated elements. The sensor controller may compare the temperature measurement to an upper temperature threshold, T Uand the lower temperature threshold T L The sensor controller may further maintain a heating status bit to indicate whether the sensor controller temperature is being intentionally increased or allowed to cool. If the junction temperature is cold (below the lower threshold T L ), or if the junction temperature is medium (between the upper and lower thresholds) and the heating status bit is set, the sensor controller proceeds to block 624.

[0040] Otherwise, in block 610, the sensor controller clears the heating status bit and proceeds to block 612. In block 612, the sensor controller generates a short pulse to detect the object and measure the distance. In block 614, the sensor controller may measure one or more transducer response parameters to sense the piezoelectric transducer state using any suitable method, such as the method outlined in the aforementioned U.S. Patent No. 11,269,067 ("Response-Based Determination of Piezoelectric Transducer State"). The sensor controller may dynamically adjust the acoustic pulse train waveform based on one or more parameters to, for example, track the ultrasonic transducer resonant frequency.

[0041] In box 616, the sensor controller can check to determine whether the sensor is blocked. As discussed in the referenced patent, snow, ice, water, mud, or other materials can cover the surface of the sensor and, in doing so, significantly affect the response of the transducer. In the event that such material blocks the microphone port, the blockage can be detected as a significant attenuation of the acoustic pulse train relative to the default state. If such a blockage is detected, the sensor controller can set the cleaning mode bit in the status register in box 618 before returning to box 602. Otherwise, in box 620, the sensor controller can clear the cleaning mode status bit to enable normal sensor operation. In box 622, the sensor controller can perform signal processing to detect the echo, determine the associated distance, and report the measurement results to the ECU before returning to box 602.

[0042] If the sensor controller has verified in block 608 that the cleaning mode is active and heating is required, the sensor controller sets or maintains the heating state bit in block 624. In block 626, the sensor controller generates a long acoustic pulse train to maximize vibration and / or energy dissipation during the current measurement interval. In block 614, the sensor controller may monitor the transducer response to determine if the sensor is still blocked.

[0043] The above-mentioned sensors, controllers, and methods can achieve a self-cleaning function, thereby reducing the impact of ultrasonic sensor blockage without significantly increasing manufacturing costs and enhancing the demand for short-range backup sensors used in camera-based systems. The self-cleaning extends beyond the piezoelectric element, enabling the cleaning of nearby microphones and microphone ports in 3D ultrasonic sensors. The self-cleaning function does not require additional components that might otherwise increase manufacturing costs, such as dedicated heating elements, independent vibration elements, or frames designed to couple ultrasonic element vibrations to the sensor bracket. Internal temperature monitoring enables self-cleaning without generating excessive temperature stress in the piezoelectric element, which might otherwise degrade sensor performance and shorten the sensor's service life. The sensor can autonomously diagnose the need for self-cleaning and automatically perform self-cleaning, but can alternatively perform such an operation when triggered by the ECU. Since the acoustic pulse train is restricted within the measurement interval, the self-cleaning function can remain transparent to the operation of other sensors in the system.

[0044] In a typical ultrasonic sensor, the duration of a single measurement interval may be approximately 50 ms, which provides a sensing range of approximately 6.5 m. If operated in chirp mode, the acoustic pulse train can be restricted to no more than approximately 2.5 ms, or if operated in amplitude modulation mode, no more than approximately 450 ms. As a proof of concept, thermographic imaging of a piezoelectric-based ultrasonic sensor and an integrated circuit controller chip showed that 2.5 ms chirp measurements repeated at 50 ms intervals for approximately 20 seconds caused sufficient heat dissipation to raise the chip temperature by 7 °C and the sensor membrane temperature by approximately 2 °C. In these experiments, the chip-driven transmit current was 400 mA.

[0045] In a specific implementation, the sensor controller uses vibration to remove dirt from the piezoelectric membrane without any special requirements for the piezoelectric bracket design. To this end, the piezoelectric transducer can be driven at the resonant frequency with a continuous pulse train (amplitude modulation mode) throughout the measurement duration (approximately 50 ms). This maximizes the vibration of the transducer to ensure the highest possible efficiency in removing dirt from the piezoelectric membrane and achieving the highest possible sound pressure level (SPL).

[0046] In one embodiment, the sensor controller uses energy dissipation to heat the sensor without the need for a separate heater and without requiring a large amount of temperature stress on the piezoelectric element. To this end, the piezoelectric transducer can be driven at a resonant frequency at a relatively high drive current (greater than or equal to about 400mA) (to maximize the efficiency of energy transfer between the piezoelectric and the air). The current level is programmable and can be adjusted. The high drive current leads to high on-chip dissipation, thereby ensuring that the overall temperature of the sensor rises. Since the chip temperature is a heat source, the on-chip junction temperature can be used to monitor the heating process. Therefore, the cleaning sequence can provide a series of measurement intervals, each measurement interval containing a continuous AM pulse train, and the series is paused when the junction temperature T exceeds an upper temperature threshold. The upper temperature threshold is preferably configurable. The series can be restarted when the junction temperature drops below a lower temperature threshold that can also be configurable.

[0047] In one specific implementation, the cleaning operation is performed autonomously. To enable the ultrasonic sensor to perform autonomous cleaning, the integrated circuit control chip can detect when the sensor is blocked, preferably without direct involvement of the ECU or any interruption of the ongoing measurement sequence. When the sensor is blocked, the control chip can inform the ECU that the measurement data should be ignored as long as the sensor remains blocked, and can initiate a continuous train of amplitude modulated acoustic pulses during unused measurement intervals to automatically clear the blockage.

[0048] In one specific implementation, the sensor controller can (automatically or upon instruction from the ECU) detect sensor blockage and malfunction while performing measurements. Techniques for performing such detection are described, for example, in U.S. Patent 11,269,067 (Response-Based Determination of Piezoelectric Transducer State). Exemplary techniques can be based on measurements of drive response or reverberation decay. Measurements such as decay time, reverberation period, transducer signal phase, and amplitude can be used for such diagnostics. If the membrane is determined to be covered with dirt or ice, the sensor controller can activate a cleaning mode using a long acoustic pulse train that completely occupies each measurement interval. A long acoustic pulse train is a continuous pulse train at the resonant frequency with a drive current of 400 mA for substantially all of each measurement interval, shaped to minimize crosstalk with other unobstructed sensors operating in dual-channel operation. The shaping and current level can be configurable to further limit such crosstalk. To this end, the frequency can also be configurable. The ECU can configure whether to enable or disable the automatic / autonomous sensor cleaning mode for each sensor.

[0049] Measurement data can be captured for each measurement interval, including the interval when cleaning mode is activated. A status bit can be used to report the cleaning activation status to the ECU along with the measurement results from the sensor. If the sensor controller detects that the sensor is no longer blocked, the status bit will be cleared and cleaning mode will be terminated.

[0050] If the sensor control chip detects that the junction temperature T is greater than the upper temperature threshold when the cleaning activation status bit is high, a long acoustic pulse train can be interrupted until the control chip detects that the junction temperature T is lower than the lower temperature threshold. At this time, if the cleaning activation status bit is high, the long acoustic pulse train is resumed.

[0051] In a contemplated system employing dual-channel sensing, when a given sensor is performing self-cleaning, the unblocked sensors in the system should be able to continue normal operation. Normal measurements operate at side frequencies rather than at the resonant frequency and are thus not expected to be disturbed by the cleaning operation of a blocked sensor that utilizes beamforming shaped at the resonant frequency and current-limited. Additional information regarding dual-channel operation can be found, for example, in U.S. Application No. 15 / 888,471, filed on February 5, 2018, entitled "Composite Acoustic Bursts for Multi-channel Sensing", which is hereby incorporated by reference.

[0052] Thermograms taken during testing at an ambient temperature of (24.8 °C) show that after operating in the cleaning mode at the above drive currents of 300 mA, 400 mA, and 500 mA for 20 seconds, the centers of the sensor membranes reached 28.0 °C, 30.1 °C, and 30.7 °C, respectively. The corresponding regions of the control chip reached 50.7 °C, 51.9 °C, and 57.0 °C, respectively. In the normal operating mode, the chip temperature was measured to be 31.6 °C. The experiment demonstrated the effective melting of the thin frost layer on the sensor membrane.

[0053] The class of ultrasonic sensors includes not only traditional transmit-receive piezoelectric transducers but also 3D ultrasonic sensors that measure not only the distance to a reflector but also the azimuth and elevation angles of the reflector relative to the sensor. Such 3D sensors can employ microphone arrangements that are closely spaced from each other (about half a wavelength) and closely spaced from the piezoelectric transducer (about half a wavelength). Such microphones can be implemented using microelectromechanical systems (MEMS) technology. Sensor blockage can occur in the form of dirt, water, ice, grime, etc. on the piezoelectric transducer membrane or in the form of blockage of one or more of the ports of the microphone. As long as blockage of the microphone ports can be detected, the previously described cleaning mode can be applied to remove those blockages because these ports are positioned close to the piezoelectric transducer and are subject to a significant amount of vibration and heating.

[0054] In a 3D ultrasonic sensor, a burst of acoustic pulses emitted by a piezoelectric transducer is coupled as crosstalk to a MEMS microphone. This crosstalk can be captured for each microphone and compared to the expected crosstalk of an unobstructed sensor. If a change in crosstalk is detected without a detected piezoelectric membrane load, such a change indicates an obstruction of the corresponding microphone. Thus, for a 3D ULS, the method of operation can include: (1) ultrasonic beamforming (i.e., emitting a burst of acoustic pulses); (2) capturing and analyzing the crosstalk signals from ultrasonic beamforming of each MEMS microphone; (3) evaluating the phase, frequency, and envelope of the crosstalk signals to diagnose fouling obstruction of the MEMS (MEMS + lid + fouling form a complex resonator); and (4) when an obstructed MEMS microphone is detected, enabling a self-cleaning mode to perform continuous amplitude-modulated beamforming during each measurement interval as long as the MEMS obstruction is detected. The beamforming current and beamforming frequency are programmable and can be optimized to maximize the cleaning efficiency of the MEMS microphone and limit MEMS stress (both temperature and mechanical stress).

[0055] Since hot air rises, the piezoelectric transducer and the sensor controller should be located below the MEMS microphone in the same module, so that the piezoelectric and the controller can act as heaters while providing a "high-pressure" wave to clean the microphone port hole. The burst of pulses from the piezoelectric element also enables the MEMS to monitor the crosstalk signal to determine when the obstruction is cleared.

[0056] It is noted here that a microphone operating as an audio sensor (e.g., for siren detection) can be co-packaged with the piezoelectric transducer to enable obstruction detection and enable self-cleaning. Even when the piezoelectric transducer operates in the ultrasonic frequency band, the audio sensor can continue to operate to detect signals in the audio frequency band to enable obstruction detection and self-cleaning of the audio sensor. The sensor controller chip can perform processing in both the audio channel (filtering out frequencies above 20 kHz) and the ultrasonic channel (filtering out frequencies below 30 kHz and above 90 kHz). Such a sensor can be described as a "self-cleaning MEMS microphone". The self-cleaning function can be used for ultrasonic sensors, 3D ultrasonic sensors, and MEMS microphones without direct involvement of the ECU and without reconfiguring the system measurement sequence. These sensors can thus be equipped with this feature as an independent safety mechanism.

[0057] Some contemplated embodiments may include one or more of the following features: 1. Shaped amplitude modulated beamforming is performed at the transducer's resonant frequency during the entire measurement interval (approximately 50 ms) used to clean a blocked sensor—normal operation of a properly operating sensor is not affected / impeded by the cleaning operation of a blocked sensor. 2. The beamforming frequency and beamforming current are configurable and can be set to minimize crosstalk into the high / low channels of other sensors (a dual-channel chirp mode may be preferred to help reduce crosstalk between the long acoustic bursts used for cleaning and the short acoustic chirps used for measurement). 3. In cleaning mode, the ultrasonic transducer can be driven at its resonant frequency to maximize the SPL and thereby optimize ultrasonic cleaning. 4. The sensor's self-cleaning function can be autonomously activated based on ultrasonic transducer and / or MEMS microphone diagnostics. MEMS microphone diagnostics can employ phase-based, decay-time-based, frequency-based, or envelope-based analysis of the crosstalk signal from the piezoelectric transducer. 5. Sensor cleaning can be performed without modifying or interrupting the system's ongoing measurement sequence. Such interruptions could undesirably halt system-level operations (e.g., parking assistance), whereas a sensor reporting an obstruction while self-cleaning is occurring can enable normal operation to continue while relying on a fully functional, unobstructed sensor. 6. The sensor can perform self-cleaning of both the piezoelectric element cover and the MEMS microphone via ultrasonic energy and self-heating of the piezoelectric element and sensor controller chip. 7. The solution can be implemented using existing sensor components and can therefore provide a low-cost solution to achieve immunity to adverse weather conditions.

[0058] For automotive applications, features made possible and more robust by using the disclosed ultrasonic sensor with self-cleaning functionality include kick-to-open, parking assist, automatic parking, and door protection. Applications are not limited to the automotive environment. As an example of an industrial application, the disclosed sensor can be used for level gauging in harsh environments.

[0059] Although for the purpose of explanation Figure 6The operations shown and described herein are considered to occur sequentially, but in practice, the method may be implemented simultaneously by multiple integrated circuit components, which may even operate speculatively to achieve out-of-order operation. This sequential discussion is not intended to be limiting. In addition, the above embodiments may omit complicating factors, such as parasitic impedances, current limiting resistors, level shifters, clamps, etc., which may be present but do not meaningfully affect the operation of the disclosed circuits. Further, the focus of the above discussion has been on ultrasonic sensors, but the principles apply to any acoustic sensor whose operating parameters may be impaired or affected by transducer loading. These and many other modifications, equivalents, and alternatives will become apparent to those skilled in the art once the above disclosure is fully understood. It is intended that the following claims be interpreted as including all such modifications, equivalents, and alternatives where applicable.

Claims

1. A sensor, characterized in that, The sensor includes: a piezoelectric transducer; and a sensor controller, the sensor controller including: a transmitter coupled to the piezoelectric transducer to generate a burst of acoustic pulses; a receiver coupled to at least one of the piezoelectric transducer and a microphone to detect reflections of the burst of acoustic pulses during a measurement interval associated with the burst of acoustic pulses; and a microcontroller configured to control the length of the burst of acoustic pulses, the microcontroller setting a longer burst length when operating to clean the sensor, the longer burst length exceeding half of the measurement interval.

2. The sensor according to claim 1, wherein, The longer burst length completely occupies the measurement interval having driving vibrations or residual ringing vibrations.

3. The sensor according to claim 1, wherein, The measurement interval is greater than 50 ms, and wherein, when operating to detect the reflection, the microcontroller sets the length of the burst of acoustic pulses to be less than or equal to 4 ms.

4. The sensor according to claim 1, wherein, When operating to clean the sensor, the transmitter drives the transducer at a resonant frequency within the length of the burst of acoustic pulses, and when operating to detect the reflection, the transmitter drives the transducer at a frequency above or below the resonant frequency.

5. The sensor according to claim 1, wherein When operating to clean the sensor, the transmitter drives the transducer with a drive current greater than or equal to 400 milliamps.

6. The sensor according to claim 1, wherein, When operating to clean the sensor, the microcontroller employs the longer burst length in a plurality of consecutive measurement intervals.

7. The sensor according to claim 1, wherein, When operating to clean the sensor, the microcontroller repeatedly employs the longer burst length until the sensor controller reaches a temperature threshold.

8. The sensor according to claim 7, wherein, When operating to clean the sensor, after the sensor controller returns from the temperature threshold to a lower temperature threshold, the microcontroller resumes employing the longer burst length.

9. The sensor according to claim 7, wherein, The sensor controller heats a microphone port to clear ice.

10. The sensor according to claim 1, wherein, When operating to clean the sensor, the sensor controller heats the piezoelectric transducer without relying on a heater.

11. The sensor according to claim 1, wherein, The microcontroller is configured to monitor for sensor damage and is configured to automatically initiate operation to clean the sensor after detecting sensor damage.

12. A method implemented by a sensor controller of a sensor, characterized in that, The method includes: driving a piezoelectric transducer to generate a short burst of acoustic pulses for obstacle detection or distance measurement; obtaining a received signal to monitor reflections of the short burst of acoustic pulses; and operating to clean the sensor by driving the piezoelectric transducer to generate a long burst of acoustic pulses at the resonant frequency of the piezoelectric transducer.

13. The method according to claim 12, wherein, The sensor implements a series of sequential measurement intervals, and wherein the long burst of acoustic pulses occupies the entire measurement interval having driving vibrations or residual ringing vibrations.

14. The method according to claim 12, wherein, The short burst of acoustic pulses drives the piezoelectric transducer at a frequency above or below the resonant frequency.

15. The method according to claim 12, wherein When operating to clean the sensor, the driving of the piezoelectric transducer is performed with a drive current greater than or equal to 400 milliamps.

16. The method according to claim 13, wherein, As part of the operation to clean the sensor, the method includes repeatedly generating the long burst of acoustic pulses until the sensor controller reaches a temperature threshold.

17. A sensor controller, characterized in that, The sensor controller includes: A transmitter configured to drive a piezoelectric transducer to generate a train of acoustic pulses; A receiver coupled to at least one of the piezoelectric transducer and a microphone to detect reflections of the train of acoustic pulses during a measurement interval associated with the train of acoustic pulses; and A microcontroller configured to control a length of the train of acoustic pulses, the microcontroller setting a longer train of acoustic pulses length when operating to clean the sensor, the longer pulse train length exceeding half of the measurement interval.

18. The sensor controller according to claim 17, wherein, The microcontroller is configured to monitor for sensor damage and is configured to automatically initiate operation to clean the sensor after detecting sensor damage.

19. The sensor controller according to claim 18, wherein, When operating to clean the sensor, the microcontroller repeatedly employs the longer train of acoustic pulses length to reach and maintain an elevated temperature of the sensor controller while maintaining the sensor damage.

20. The sensor controller according to claim 19, wherein, The measurement interval is greater than 50 ms, and wherein when operating to detect the reflections, the microcontroller sets the length of the train of acoustic pulses to be less than or equal to 4 ms.

Images