A hail detection method and related apparatus

Abstract

A hail detection method and related device applied to the technical field of protection control. The elastic wave generated when hail hits the vehicle body has specific signal characteristics. By comparing the elastic wave signals propagating on the surface of the vehicle body and the signal characteristics in the hail scene, the method can effectively distinguish between hail, raindrops, tree branch scratches, stone splashes, and human knocking, thereby improving the accuracy of hail detection. In addition, the elastic wave propagates extremely fast in the vehicle body metal or composite material, and the initial collision of hail can be detected earlier through elastic wave detection of hail, so as to protect the vehicle from hail as early as possible and minimize user loss.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of protection control, in particular to a hail detection method and related device. BACKGROUND

[0002] In recent years, the number of hail weather occurrences is increasing, and hail particles have high kinetic energy when falling, which can cause the body sheet metal of a vehicle to be dented, the vehicle window glass to be broken, the vehicle lamp shell to be cracked, and the outside rearview mirror to be damaged, etc., causing great loss to the user. At the same time, hail particles can also damage radar, camera and other important sensing devices of intelligent driving, causing the intelligent driving function to degrade or even be unavailable.

[0003] Some solutions consider automatically providing hail protection for vehicles in hail weather to reduce damage to the vehicle. However, hail weather has the characteristics of strong suddenness, short duration, and inaccurate prediction and early warning. Therefore, how to accurately identify hail weather is a hot issue that technicians in the field are studying. SUMMARY

[0004] The present application provides a hail detection method and related device, which can accurately identify hail impact at night or in bad weather, ensure the continuity and reliability of detection, and improve the accuracy of hail detection.

[0005] In a first aspect, the present application provides a hail detection method, which can be executed by a device with data processing capability. For ease of description, the following will be described by taking a detection device as an example.

[0006] The hail detection method includes: a detection device acquiring an elastic wave signal from an elastic wave sensor, the elastic wave sensor being arranged on a vehicle, and the elastic wave signal propagating on the surface of the vehicle body. The detection device compares the elastic wave signal with a signal feature in a hail scene to obtain a first detection result, the first detection result being used to indicate whether a hail scene is detected.

[0007] Compared with detecting hail by using an image sensor, the above method detects hail by using an elastic wave sensor, which is completely unaffected by light conditions, rain and fog blockage, or visual interference, can accurately identify hail impact at night or in bad weather, and ensures the continuity and reliability of detection. Moreover, the elastic wave generated when hail impacts the vehicle body has a specific signal feature (such as pulse width, amplitude and rising slope). By comparing the elastic wave signal propagating on the surface of the vehicle body with the signal feature in the hail scene, the detection device can effectively distinguish hail, raindrops, tree branch scratches, stone splashes, and human knocking, etc., and can improve the accuracy of hail detection. In addition, the elastic wave propagates at a very fast speed in the vehicle body metal or composite material, and the initial collision of hail can be perceived earlier by detecting hail by using the elastic wave, so as to protect the vehicle from hail as early as possible and reduce the loss of the user as much as possible.

[0008] In a possible implementation of the first aspect, the first detection result is used to indicate whether the hail scene is detected based on the elastic wave signal, and the method further includes: obtaining, by the detection device, acceleration data from an acceleration sensor, the acceleration sensor being arranged on the vehicle; and comparing, by the detection device, the acceleration data with acceleration data features in the hail scene to obtain a second detection result, the second detection result being used to indicate whether the hail scene is detected based on the acceleration data.

[0009] In the above implementation, the acceleration sensor is a relatively common component in current vehicles, and detecting hail by using the acceleration sensor can reduce hardware costs and facilitate rapid popularization in mass-produced vehicles. Moreover, the acceleration sensor can perceive the overall deformation and vibration pattern of the whole vehicle or local structure caused by hail impact, and can detect the impact caused by large-particle hail or dense hail.

[0010] In another possible implementation of the first aspect, the method further includes: generating, by the detection device, a first instruction for waking up at least one sensor including one or more of a microphone, a visible light image sensor, an infrared image sensor, and a radar sensor, when the first detection result indicates that the hail scene is detected based on the elastic wave signal and / or the second detection result indicates that the hail scene is detected based on the acceleration data. The detection device obtains at least one perception detection result based on perception data from the at least one sensor, each perception detection result being used to indicate whether the hail scene is detected based on the perception data of the corresponding one or more sensors.

[0011] Since the visible light image sensor, the infrared image sensor, and the radar sensor belong to high-power-consumption and high-computing-power devices, while the elastic wave sensor and the acceleration sensor have extremely low power consumption and can be in a low-power-consumption long-term working state, the detection device can significantly reduce system power consumption and computing load by only temporarily waking up the high-power-consumption sensors when the low-power-consumption sensors detect the hail scene. By using multiple sensors to detect the hail scene, i.e., by complementing multiple sources of information, a more reliable judgment can be achieved than a single sensor, and the robustness and accuracy of hail detection can be improved.

[0012] In another possible implementation of the first aspect, the method further includes: controlling, by the detection device, one or more components in the vehicle to perform a hail protection operation based on at least the first detection result.

[0013] The above implementation can minimize the physical damage to the vehicle, and the hail protection operation can be automatically completed when the user is not in the vehicle, thereby improving the user's trust in the intelligence and reliability of the vehicle. Even if the user is in the vehicle, the user does not need to remember complex protection steps, and the detection device can automatically complete the hail protection operation, thereby improving the user experience.

[0014] In a further possible implementation form of the first aspect, the detecting device controlling one or more components in the vehicle to perform the hail protection operation comprises one or more of: sending prompt information to a terminal associated with the vehicle, the prompt information being used to prompt that hail occurs at a location where the vehicle is located; obtaining an environmental image around the vehicle and sending the environmental image around the vehicle to the terminal; closing a door of the vehicle; closing a window of the vehicle; folding a rearview mirror of the vehicle.

[0015] In a further possible implementation form of the first aspect, the detecting device controlling one or more components in the vehicle to perform the hail protection operation further comprises one or more of: controlling the vehicle to drive to a safe parking position; or adjusting a pose of the vehicle. The above examples illustrate a plurality of hail protection operations, which can reduce physical damage to the vehicle from multiple aspects.

[0016] In a further possible implementation form of the first aspect, the detecting device controlling one or more components in the vehicle to perform the hail protection operation based on at least the first detection result comprises: the detecting device controlling one or more components in the vehicle to perform the hail protection operation based on at least the first detection result and a hail level of a location where the vehicle is located, there being different hail protection operations corresponding to at least two hail levels.

[0017] The hail protection operations corresponding to different hail levels can be different, for example, providing strong enough protection in high risk and avoiding excessive intervention in low risk, so as to make the protection effect of the vehicle correspond to the actual hail level as much as possible, to achieve fine control, to improve user experience, and to reduce unnecessary interference.

[0018] In a further possible implementation form of the first aspect, the at least two hail levels comprise a first hail level and a second hail level. The hail protection operation corresponding to the first hail level comprises one or more of: sending prompt information to a terminal associated with the vehicle; obtaining an environmental image around the vehicle and sending the environmental image around the vehicle to the terminal; closing a door of the vehicle; closing a window of the vehicle; or folding a rearview mirror of the vehicle. The hail protection operation corresponding to the second hail level comprises the hail protection operation corresponding to the first hail level, and further comprises one or more of: controlling the vehicle to drive to a safe parking position; or adjusting a pose of the vehicle. Illustratively, the first hail level is a low hail level, which causes less damage to the vehicle, and the vehicle is not moved at this time. The second hail level is a high hail level, which causes greater damage to the vehicle, and the vehicle needs to be moved to reduce damage to the vehicle as much as possible.

[0019] In another possible implementation of the first aspect, the detection device controls one or more components in the vehicle to perform hail protection operations based at least on a first detection result, including: the detection device controlling one or more components in the vehicle to perform hail protection operations when the first detection result indicates that a hail scene has been detected based on an elastic wave signal, and the confidence level of the first detection result is greater than or equal to a first threshold.

[0020] Hail typically forms suddenly during severe convective weather, and the time from the start of its fall to its concentrated impact on vehicles can be extremely short. In the above implementation method, if the confidence level of the first detection result is high, the hail protection operation can be executed directly, which can greatly reduce the response time. By executing the hail protection operation in advance before the hailstones hit the vehicle body on a large scale and in concentrated form, the time window can be maximized, reducing the actual number of impacts. Moreover, when the confidence level is high, unnecessary wake-ups can be avoided, minimizing the computational load and energy consumption.

[0021] In another possible implementation of the first aspect, the detection device controls one or more components in the vehicle to perform hail protection operations based at least on a first detection result, including: the detection device controls one or more components in the vehicle to perform hail protection operations when the first detection result indicates that a hail scene has been detected based on an elastic wave signal, and the confidence level of the first detection result is less than a first threshold, and at least one of the second detection result and at least one sensing detection result indicates that a hail scene has been detected.

[0022] A single sensor has a higher probability of false detection. When the confidence level is low, combining the detection results of other sensors to perform hail protection operations can overcome the limitations of a single sensor, achieve more reliable judgment than a single sensor, significantly reduce the false detection rate, reduce unnecessary interference, and improve the user experience.

[0023] In another possible implementation of the first aspect, the detection device compares the elastic wave signal with the signal characteristics under the hail scene to obtain a first detection result, including: the detection device compares the elastic wave signal with the signal characteristics under the hail scene to obtain the first detection result and the confidence level of the first detection result.

[0024] In the above embodiments, the confidence level of the detection results is beneficial for the subsequent detection device to adopt different strategies according to the confidence level, thereby achieving refined control and improving the user experience.

[0025] In another possible implementation of the first aspect, the detection device controls one or more components in the vehicle to perform hail protection operations based at least on a first detection result, including: if at least two of the first detection result, a second detection result, and at least one sensing detection result indicate that a hail scene has been detected, the detection device controls one or more components in the vehicle to perform hail protection operations.

[0026] A single sensor has a high probability of false detection. Combining the detection results of multiple sensors to perform hail protection operations can overcome the limitations of a single sensor, achieve more reliable judgment than a single sensor, significantly reduce the false detection rate, reduce unnecessary interference, and improve the user experience.

[0027] In another possible implementation of the first aspect, the signal characteristics under the hail scenario include hail time-domain characteristics, which are the time-domain characteristics of the elastic wave signal propagating on the vehicle surface when hail collides with the vehicle. The detection device compares the elastic wave signal with the signal characteristics under the hail scenario to obtain a first detection result, including: the detection device obtaining the time-domain characteristics of the elastic wave signal based on the elastic wave signal, the time-domain characteristics including one or more of pulse width, amplitude, and rise slope. The detection device compares the time-domain characteristics of the elastic wave signal with the hail time-domain characteristics to obtain the first detection result.

[0028] The elastic waves generated when hail hits a vehicle have specific time-domain characteristics (such as pulse width, amplitude, and rise slope). By comparing the time-domain characteristics of hail, the detection device described above can effectively distinguish between hail, raindrops, tree branch scratches, flying pebbles, and human impacts, thereby improving the accuracy of hail detection.

[0029] In another possible implementation of the first aspect, the detection device compares the time-domain characteristics of the elastic wave signal with the time-domain characteristics of hail to obtain a first detection result, including: the detection device filtering out elastic wave signals whose time-domain characteristics do not meet the detection conditions to obtain a filtered elastic wave signal, wherein the detection conditions include one or more of pulse width conditions, amplitude conditions, and rise slope conditions. The detection device obtains the first detection result based on the time-domain characteristics of the filtered elastic wave signal and the time-domain characteristics of hail.

[0030] The above implementation method, by setting detection conditions, directly filters out "weak signals" that obviously do not conform to the time domain characteristics of hail, and only retains potential target signals for subsequent complex feature comparison. This can significantly reduce the number of signals that need to be compared with complex features, thereby reducing the computational load, improving hail detection speed, and improving real-time performance.

[0031] In another possible implementation of the first aspect, the detection conditions include an amplitude greater than an amplitude detection threshold and / or an ascent rate greater than an ascent rate detection threshold, and the method further includes: the detection device increasing the amplitude detection threshold and / or the ascent rate detection threshold when the weather forecast information at the vehicle's location indicates the presence of hail, or when the rainfall at the vehicle's location is greater than a rainfall threshold.

[0032] In extreme weather conditions (such as heavy rain or hail), the energy of background noise and interference signals generally increases. If the detection threshold remains low, these background noise and interference signals will be misjudged as "hail impacts," leading to frequent false triggers. The above-described implementation method can significantly reduce the false detection rate by increasing the detection threshold and also enhance the robustness of the detection method under high noise / high interference conditions.

[0033] In another possible implementation of the first aspect, the filtered elastic wave signal includes multiple pulses. The detection device obtains a first detection result based on the time-domain characteristics of the filtered elastic wave signal and the time-domain characteristics of hail. This includes: the detection device determining the number of matching pulses, where the matching pulses are pulses whose time-domain characteristics match the hail time-domain characteristics among the multiple pulses. If the number of matching pulses within a first duration is greater than or equal to a quantity threshold, the first detection result indicates that a hail scene has been detected based on the elastic wave signal. Alternatively, if the number of matching pulses within the first duration is less than the quantity threshold, the first detection result indicates that a hail scene has not been detected based on the elastic wave signal.

[0034] Random noise usually occurs in isolation or at long intervals, while hail is a continuous fall that generates multiple consecutive impact signals in a short period of time. The above implementation method utilizes the statistical characteristics of hail impacts (continuity, density, etc.) and requires "multiple matching pulses to appear within the first time period". Random noise is almost impossible to meet this condition, thereby greatly reducing false detections caused by random noise.

[0035] In another possible implementation of the first aspect, the acceleration data features under hail scenarios include the frequency domain features of the acceleration data when hail collides with a vehicle. The detection device compares the acceleration data with the acceleration data features under hail scenarios to obtain a second detection result, including: the detection device obtaining the frequency domain features of the acceleration data based on the acceleration data, the frequency domain features including one or more of the following: dominant frequency, high-frequency energy ratio, and decay time constant. The detection device compares the frequency domain features of the acceleration data with the frequency domain features of the acceleration data when hail collides with a vehicle to obtain the second detection result.

[0036] The vibrations generated when hail hits a vehicle have specific frequency domain characteristics (such as dominant frequency, high-frequency energy ratio, and decay time constant). By comparing the frequency domain characteristics of hail, the detection device described above can effectively distinguish between hail, raindrops, tire noise / driving vibration, etc., thereby improving the accuracy of hail detection.

[0037] In another possible implementation of the first aspect, the elastic wave signal includes multiple matched pulses, each of which is a pulse whose time-domain characteristics match the time-domain characteristics of hail. The method further includes: a detection device determining the time-domain characteristic range to which the time-domain characteristics of each matched pulse belong, with different time-domain characteristic ranges corresponding to different hail levels. The detection device determines the hail level at the vehicle's location based on the time-domain characteristic range to which the time-domain characteristics of each matched pulse belong.

[0038] The temporal characteristics (such as pulse width, amplitude, and rise slope) of the elastic waves generated when hailstones impact a vehicle body at different hail levels belong to different temporal characteristic ranges. For example, large hailstones impacting a vehicle body have a short contact time but extremely high impact force, resulting in elastic waves with large amplitude, high rise slope, and narrow pulse width. Small hailstones impacting a vehicle body have a longer contact time and lower impact force, resulting in elastic waves with smaller amplitude, lower rise slope, and wider pulse width. By determining the temporal characteristic range to which the above-described implementation method belongs, the detection device can effectively distinguish hail levels, thereby improving the accuracy of hail level detection.

[0039] In another possible implementation of the first aspect, the elastic wave signal includes a plurality of matched pulses, wherein the matched pulses are pulses whose time-domain features match the time-domain features of hail; the method further includes: a detection device determining the time-domain feature range to which the time-domain features of each of the plurality of matched pulses belong, with different time-domain feature ranges corresponding to different hail levels. The detection device determines a first hail level based on the time-domain feature range to which the time-domain features of each matched pulse belong. The detection device determines the hail size at the vehicle's location based on first sensing data, wherein the first sensing data includes sensing data from one or more sensors selected from visible light image sensors, infrared image sensors, and radar sensors. The detection device determines the hail level at the vehicle's location based on the first hail level and the hail size at the vehicle's location.

[0040] The possibility of error in detecting hail levels using a single sensor is relatively high. By using image sensors and / or radar sensors to detect hail size and combining the hail size with the hail level detected by a single sensor to obtain a comprehensive hail level, the limitations of a single sensor can be overcome, resulting in a more reliable judgment and improving the accuracy of hail level detection.

[0041] In another possible implementation of the first aspect, the detection device controlling one or more components of the vehicle to perform hail protection operations includes controlling the vehicle to a safe parking position. The detection device controlling one or more components of the vehicle to perform hail protection operations, based at least on a first detection result, includes: the detection device detecting whether a sheltered safe parking position exists around the vehicle. If a sheltered safe parking position exists around the vehicle, the detection device controls the vehicle to drive to the safe parking position, based at least on the first detection result.

[0042] In the above embodiments, the detection device automatically detects whether there is a safe parking location that can be sheltered around the vehicle, so that it can automatically control the vehicle to drive to the detected safe parking location in hail scenarios, thereby minimizing physical damage to the vehicle.

[0043] In another possible implementation of the first aspect, the detection device controlling one or more components in the vehicle to perform hail protection operations includes adjusting the vehicle's orientation. The detection device, based at least on a first detection result, controls one or more components in the vehicle to perform hail protection operations, including: the detection device determining the wind direction at the vehicle's location based on second sensing data, the second sensing data including sensing data from one or more sensors selected from visible light image sensors, infrared image sensors, radar-type sensors, and inertial measurement units. The detection device, based at least on the first detection result, adjusts the vehicle's orientation so that the side of the vehicle body faces the wind direction at the vehicle's location.

[0044] Because vehicle side windows are generally more impact-resistant than windshields, the curved design of car doors can deflect hail to some extent, and car doors are relatively small, with door dent repairs typically costing less than replacing a windshield or repairing a hood. For example, in a safe parking position with no cover, detection devices can steer the vehicle's sides towards the wind in hail scenarios, allowing the sides to absorb more hail impact and minimizing vehicle repair costs.

[0045] In another possible implementation of the first aspect, the detection device compares acceleration data with acceleration data characteristics under a hail scenario to obtain a second detection result, including: the detection device comparing acceleration data with acceleration data characteristics under a hail scenario to obtain a second detection result and a confidence level of the second detection result. The detection device obtains at least one perception detection result based on perception data from at least one sensor, including: the detection device obtains at least one perception detection result and a confidence level of at least one perception detection result based on perception data from at least one sensor. The method further includes: the detection device controlling one or more components in the vehicle to perform hail protection operations based on a first detection result, a confidence level of the first detection result, a second detection result, a confidence level of the second detection result, at least one perception detection result, and a confidence level of at least one perception detection result.

[0046] Any single sensor has a limited coverage area. Combining multiple sensors for hail detection can eliminate single-point blind spots and increase coverage. Moreover, combining the detection results and confidence levels of multiple sensors to perform hail protection operations can achieve more reliable judgments than with a single sensor, significantly reducing false detection rates, minimizing unnecessary interference, and improving the user experience.

[0047] Secondly, this application provides a detection device. The detection device includes units for implementing the method described in the first aspect or any possible embodiment of the first aspect. Optionally, the detection device includes a transceiver unit and a processing unit. The transceiver unit is used to acquire and transmit information, such as receiving, reading, and transmitting information, while the processing unit is used to process data (e.g., processing the data acquired by the transceiver unit). For example, the transceiver unit is used to acquire elastic wave signals from an elastic wave sensor disposed on a vehicle, the elastic wave signal propagating on the surface of the vehicle body. The processing unit is used to compare the elastic wave signal with signal characteristics under a hail scene to obtain a first detection result, the first detection result indicating whether a hail scene has been detected.

[0048] In one possible implementation of the second aspect, the first detection result is used to indicate whether a hail scene has been detected based on the elastic wave signal. The transceiver unit is further used to acquire acceleration data from an acceleration sensor installed on the vehicle. The processing unit is further used to compare the acceleration data with the acceleration data characteristics under a hail scene to obtain a second detection result, which is used to indicate whether a hail scene has been detected based on the acceleration data.

[0049] In another possible implementation of the second aspect, the processing unit is further configured to generate a first instruction if, in the case that a first detection result indicates that a hail scene has been detected based on elastic wave signals, and / or a second detection result indicates that a hail scene has been detected based on acceleration data, the first instruction being used to wake up at least one sensor, the at least one sensor including one or more of a microphone, a visible light image sensor, an infrared image sensor, and a radar-type sensor. The processing unit is further configured to obtain at least one perception detection result based on perception data from the at least one sensor, each perception detection result being used to indicate whether a hail scene has been detected based on perception data from one or more corresponding sensors.

[0050] In another possible implementation of the second aspect, the processing unit is further configured to control one or more components in the vehicle to perform hail protection operations, based at least on the first detection result.

[0051] In another possible implementation of the second aspect, controlling one or more components in the vehicle to perform hail protection operations includes one or more of the following: sending a notification message to a terminal associated with the vehicle, the notification message indicating that hail has occurred at the location of the vehicle; acquiring an image of the environment around the vehicle and sending the image of the environment around the vehicle to the terminal; closing the vehicle doors; closing the vehicle windows; and retracting the vehicle's rearview mirrors.

[0052] In another possible implementation of the second aspect, controlling one or more components in the vehicle to perform hail protection operations further includes one or more of the following: controlling the vehicle to drive to a safe parking position; or adjusting the vehicle's position.

[0053] In another possible implementation of the second aspect, the processing unit is further configured to control one or more components in the vehicle to perform hail protection operations based at least on the first detection result and the hail level at the vehicle's location, wherein there are at least two hail levels for which the hail protection operations are different.

[0054] In another possible implementation of the second aspect, at least two hail levels include a first hail level and a second hail level. The hail protection operation corresponding to the first hail level includes one or more of the following: sending a warning message to a terminal associated with the vehicle; acquiring an image of the environment around the vehicle and sending the image to the terminal; closing the vehicle doors; closing the vehicle windows; or rolling up the vehicle's rearview mirrors. The hail protection operation corresponding to the second hail level includes the hail protection operation corresponding to the first hail level, and further includes one or more of the following: controlling the vehicle to a safe parking position; or adjusting the vehicle's position.

[0055] In another possible implementation of the second aspect, the processing unit is further configured to control one or more components in the vehicle to perform hail protection operations when the first detection result indicates that a hail scene has been detected based on the elastic wave signal and the confidence level of the first detection result is greater than or equal to a first threshold.

[0056] In another possible implementation of the second aspect, the processing unit is further configured to control one or more components in the vehicle to perform hail protection operations when the first detection result indicates that a hail scene has been detected based on the elastic wave signal, and the confidence level of the first detection result is less than a first threshold, and at least one of the second detection result and at least one sensing detection result indicates that a hail scene has been detected.

[0057] In another possible implementation of the second aspect, the processing unit is further configured to compare the elastic wave signal with the signal characteristics under the hail scene to obtain a first detection result and a confidence level of the first detection result.

[0058] In another possible implementation of the second aspect, the processing unit is further configured to control one or more components in the vehicle to perform hail protection operations when at least two of the first detection result, the second detection result, and at least one sensing detection result indicate that a hail scene has been detected.

[0059] In another possible implementation of the second aspect, the signal characteristics in the hail scenario include hail time-domain characteristics, which are the time-domain characteristics of the elastic wave signal propagating on the vehicle surface when hail collides with the vehicle. The processing unit is further configured to obtain the time-domain characteristics of the elastic wave signal based on the elastic wave signal, the time-domain characteristics including one or more of pulse width, amplitude, and rise slope. The processing unit is further configured to compare the time-domain characteristics of the elastic wave signal with the hail time-domain characteristics to obtain a first detection result.

[0060] In another possible implementation of the second aspect, the processing unit is further configured to filter out elastic wave signals whose time-domain characteristics do not meet the detection conditions, thereby obtaining a filtered elastic wave signal. The detection conditions include one or more of pulse width conditions, amplitude conditions, and rise slope conditions. The processing unit is further configured to obtain a first detection result based on the time-domain characteristics of the filtered elastic wave signal and the hail time-domain characteristics.

[0061] In another possible implementation of the second aspect, the detection conditions include an amplitude greater than an amplitude detection threshold and / or an upward slope greater than an upward slope detection threshold. The processing unit is further configured to increase the amplitude detection threshold and / or the upward slope detection threshold when the weather forecast information at the vehicle's location indicates the presence of hail, or when the rainfall at the vehicle's location is greater than a rainfall threshold.

[0062] In another possible implementation of the second aspect, the filtered elastic wave signal includes multiple pulses, and the processing unit is further configured to determine the number of matching pulses, wherein the matching pulses are pulses whose time-domain characteristics match the time-domain characteristics of hail among the multiple pulses. If the number of matching pulses within a first duration is greater than or equal to a quantity threshold, a first detection result indicates that a hail scene has been detected based on the elastic wave signal. Alternatively, if the number of matching pulses within the first duration is less than the quantity threshold, a first detection result indicates that a hail scene has not been detected based on the elastic wave signal.

[0063] In another possible implementation of the second aspect, the acceleration data features in the hail scenario include the frequency domain features of the acceleration data when hail collides with a vehicle. The processing unit is further configured to obtain the frequency domain features of the acceleration data based on the acceleration data, the frequency domain features including one or more of the following: dominant frequency, high-frequency energy ratio, and decay time constant. The processing unit is further configured to compare the frequency domain features of the acceleration data with the frequency domain features of the acceleration data when hail collides with a vehicle to obtain a second detection result.

[0064] In another possible implementation of the second aspect, the elastic wave signal includes multiple matched pulses, where each matched pulse's time-domain feature matches the time-domain feature of the hail. The processing unit is further configured to determine the time-domain feature range to which the time-domain feature of each matched pulse belongs, with different hail levels corresponding to different time-domain feature ranges. The processing unit is also configured to determine the hail level at the vehicle's location based on the time-domain feature range to which the time-domain feature of each matched pulse belongs.

[0065] In another possible implementation of the second aspect, the elastic wave signal includes multiple matched pulses, which are pulses whose time-domain characteristics match the time-domain characteristics of hail. The processing unit is further configured to determine the time-domain feature range to which the time-domain characteristics of each matched pulse belong, with different time-domain feature ranges corresponding to different hail levels. The processing unit is further configured to determine a first hail level based on the time-domain feature range to which the time-domain characteristics of each matched pulse belong. The processing unit is further configured to determine the hail size at the vehicle's location based on first sensing data, which includes sensing data from one or more sensors selected from visible light image sensors, infrared image sensors, and radar sensors. The processing unit is further configured to determine the hail level at the vehicle's location based on the first hail level and the hail size at the vehicle's location.

[0066] In another possible implementation of the second aspect, controlling one or more components in the vehicle to perform hail protection operations includes controlling the vehicle to a safe parking position, and the processing unit is further configured to detect whether a sheltered safe parking position exists around the vehicle. If a sheltered safe parking position exists around the vehicle, the processing unit is further configured to control the vehicle to the safe parking position based at least on the first detection result.

[0067] In another possible implementation of the second aspect, controlling one or more components in the vehicle to perform hail protection operations includes adjusting the vehicle's orientation. The processing unit is further configured to determine the wind direction at the vehicle's location based on second sensing data, which includes sensing data from one or more sensors selected from visible light image sensors, infrared image sensors, radar-type sensors, and inertial measurement units. The processing unit is also configured to adjust the vehicle's orientation so that the vehicle's side faces the wind direction at the vehicle's location, based at least on a first detection result.

[0068] In another possible implementation of the second aspect, the processing unit is further configured to compare the acceleration data with acceleration data characteristics under a hail scenario to obtain a second detection result and a confidence level of the second detection result. The processing unit is further configured to obtain at least one perception detection result and a confidence level of at least one perception detection result based on perception data from at least one sensor. The processing unit is further configured to control one or more components in the vehicle to perform hail protection operations based on the first detection result, the confidence level of the first detection result, the second detection result, the confidence level of the second detection result, at least one perception detection result, and the confidence level of at least one perception detection result.

[0069] Thirdly, embodiments of this application provide a computing device, the computing device including a processor and a memory, the memory storing a program, the processor executing the program stored in the memory to enable the computing device to implement the method described in any of the first aspects above.

[0070] Fourthly, this application provides a chip including a processor for executing computer execution instructions to cause a device on which the chip is mounted to perform the method described in any of the first aspects above.

[0071] Fifthly, this application provides a vehicle that includes the detection device described in the second aspect, the computing device described in the third aspect, or the chip described in the fourth aspect, and the vehicle is used to implement the method described in any of the first aspects.

[0072] In a sixth aspect, embodiments of this application provide a computer-readable storage medium for storing a computer program, the computer program including instructions for performing the methods described in any of the first aspects above.

[0073] In a seventh aspect, this application provides a computer program product including computer instructions that, when executed by a detection device, computing device, or processor, cause the method described in any of the first aspects to be implemented.

[0074] The solutions provided in the second to seventh aspects above are used to implement or cooperate with the methods provided in the first aspect above, and therefore can achieve the same or corresponding beneficial effects as the first aspect, which will not be elaborated here. Attached Figure Description

[0075] The accompanying drawings used in the description of the embodiments will be briefly introduced below.

[0076] Figure 1 This is a schematic diagram of the architecture of a terminal provided in an embodiment of this application; Figure 2 This is a schematic diagram of a sensor in a vehicle provided in an embodiment of this application; Figure 3 This is a schematic flowchart of a hail detection method provided in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of a detection device provided in an embodiment of this application; Figure 5 This is a schematic diagram of the structure of a computing device provided in an embodiment of this application. Detailed Implementation

[0077] The following section provides an exemplary description of the systems and scenarios in which this application may be applied. It should be noted that the system architecture and business scenarios described in this application are for the purpose of more clearly illustrating the technical solutions of this application and do not constitute a limitation on the technical solutions provided in this application. Those skilled in the art will understand that, with the evolution of system architecture and the emergence of new business scenarios, the technical solutions provided in this application are equally applicable to similar technical problems.

[0078] Figure 1 This is a schematic diagram of the architecture of a terminal provided in an embodiment of this application. In this embodiment, terminal 100 can be any of the following: meteorological monitoring equipment, intelligent transportation equipment, intelligent manufacturing equipment, or smart city equipment, etc. Intelligent transportation equipment includes, but is not limited to, automobiles, two-wheeled vehicles, ships, drones, trains, freight cars, trucks, or logistics robots. Intelligent manufacturing equipment includes, but is not limited to, robots, intelligent industrial equipment, intelligent logistics, and smart factories. In short, the embodiments of this application are applicable to various types of terminals. For ease of description, some embodiments use a vehicle as an example for terminal 100, but this application is equally applicable to other types of terminals.

[0079] like Figure 1 As shown, the terminal 100 includes a sensor system 10 and a detection device 20.

[0080] The sensor system 10 may include several types of sensors for sensing information about the environment surrounding the terminal 100. For example, combined with Figure 1The sensor system 10 includes an elastic wave sensor 101. The sensor system 10 may also include one or more of the following: an acceleration sensor 102, a microphone 103, a visible light image sensor 104, an infrared image sensor 105, and a radar sensor 106.

[0081] Elastic wave sensor 101 is a tactile sensor used to collect elastic wave signals (also referred to as elastic waves in some embodiments) propagating on the surface of terminal 100. Exemplarily, elastic wave sensor 101, based on the principle of elastic waves and piezoelectric ceramics, is capable of collecting elastic wave signals that can be used for impact detection and / or collision detection. An elastic wave is a complex mechanical wave that can be generated and propagated within hard materials. Elastic waves are classified as longitudinal waves, transverse waves, and surface waves (a combination of longitudinal and transverse waves), and are waves that propagate within or on the surface of a solid or fluid medium (such as stress waves). For example, elastic wave signals include high-frequency pulses propagating on the surface of terminal 100. Elastic wave sensor 101 supports point-to-surface sensing and is passive sensing, with low power consumption, is unaffected by electromagnetic interference, and supports detection on complex curved / complex structural surfaces. Elastic waves contain rich information that can be used to detect the location of the touch point, analyze the force of the touch, and the characteristics of the elastic wave can be used to filter out interference such as rain, and to distinguish the material of the contact object (such as the fingertip or the back of the finger). For example, the elastic wave sensor 101 can detect high-frequency signals (such as 500Hz to 5kHz and above), with waveform characteristics of sharp pulses, steep rise edges, short duration (such as less than 10ms), high energy concentration (high instantaneous power), high sensitivity to installation location (needs to be close to the impact point), and is easily affected by localized impact interference.

[0082] Accelerometer 102 is used to collect acceleration data from terminal 100. Exemplarily, the acceleration data includes low-frequency micro-vibration acceleration caused by forces acting on the structure of terminal 100. That is, accelerometer 102 is a sensor based on structural vibration (or vibration) detection, with a long transmission path and susceptibility to external interference. Exemplarily, accelerometer 102 can detect low-frequency signals (e.g., less than 500Hz), with waveform characteristics of damped oscillation (or oscillation), a slow rise time, a long duration (e.g., 10ms to 100ms), low energy concentration (energy dispersed throughout the vibration cycle), low sensitivity to installation location (able to sense overall vibration), and is easily affected by road surface vibration.

[0083] Microphone 103 is used to collect sound data around terminal 100, and supports sound detection and voiceprint recognition. The sound data can be used to detect the sound of hail falling and voiceprints.

[0084] The visible light image sensor 104 is used to acquire visible light image data around the terminal 100. The visible light image data can be used to detect one or more of the following: hail particles, shape, and size.

[0085] Infrared image sensor 105 is used to collect infrared image data around terminal 100, such as heat source data, thereby realizing functions such as heat source detection, ranging, and temperature measurement. Infrared image data can be used to detect temperature differences between hail and the environment, forming multiple distinct temperature difference zones.

[0086] Radar sensor 106 is used to detect objects using electromagnetic waves and obtain point cloud data. Based on the different detection signals emitted, radar sensor 106 can be classified into millimeter-wave radar, centimeter-wave radar, ultrasonic radar, lidar, etc. Point cloud data can be used to detect one or more of the following: hailstone particles, shape, and size.

[0087] Taking terminal 100 as an example, combined with Figure 2 The vehicle includes two elastic wave sensors 101 (sensors that can reuse the vehicle's existing knock-knock (KK) function) arranged at the front and rear of the vehicle, capable of detecting elastic wave signals (or surface wave signals) propagating on the vehicle's surface when the vehicle is struck (or colliding) from the outside. The vehicle also includes an acceleration sensor 102 installed inside the vehicle. The acceleration sensor 102 can be a multi-axis acceleration sensor installed inside an inertial measurement unit (IMU), capable of supporting XYZ axis vehicle body acceleration detection. For ease of description, the fore-and-aft direction of the normally installed seats is defined as the X direction, the left-and-right direction of the seats as the Y direction, and the direction parallel to gravity as the Z direction. The X, Y, and Z directions are mutually perpendicular. It is understood that in other embodiments, the coordinate system can be established using other reference methods, which are not limited here. The vehicle also includes multiple microphones 103, which can be located at the front, rear, left, and right of the vehicle. The vehicle also includes multiple visible light image sensors 104, which can be located at the front, rear, left, and right of the vehicle, and are capable of acquiring visible light image data of the surroundings. The vehicle also includes multiple infrared image sensors 105, which can be located at the front, rear, left, and right of the vehicle, and are capable of acquiring infrared image data of the surroundings. The vehicle also includes radar sensors 106. Optionally, the radar sensors 106 can be installed on the roof of the vehicle's cabin, or on top of the seats, etc., etc., which will not be listed here. It should be noted that... Figure 2 The number and placement of various sensors shown are merely examples. In actual use, there may be many more designs for the number and placement of various sensors, which will not be listed here and do not constitute a limitation on the embodiments of this application.

[0088] The detection device 20 has data processing capabilities. In some possible implementations, the detection device 20 may also include a hardware module with data processing capabilities and / or a software module with data processing capabilities. Examples are given below based on hardware and software implementations respectively.

[0089] As an example of hardware implementation, the detection device 20 may include at least one processor, which is a module with processing capabilities. In one implementation, the processor may be a circuit with instruction read and execute capabilities, such as a central processing unit (CPU), microprocessor, microcontroller unit (MCU), graphics processing unit (GPU), or digital signal processor (DSP). In another implementation, the processor can implement certain functions through the logical relationships of hardware circuits, which may be fixed or reconfigurable. For example, the processor may be a hardware circuit implemented as an application-specific integrated circuit (ASIC) or a programmable logic device (PLD), such as a field-programmable gate array (FPGA). In reconfigurable hardware circuits, the process of the processor loading a configuration document and configuring the hardware circuit can be understood as the process of the processor loading instructions to implement the corresponding function. Furthermore, the processor can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as a neural network processing unit (NPU), tensor processing unit (TPU), deep learning processing unit (DPU), etc. In some implementations, the detection device 20 includes at least one processor integrated as a system-on-chip (SOC), which is commonly referred to as an SOC by those skilled in the art. The SOC may include at least one processor, and when the SOC includes multiple processors, the types of processors can be different, such as including a CPU and an MCU.

[0090] For example, the detection device 20 includes, but is not limited to, a domain controller (DC), a mobile data center (MDC), an electronic control unit (ECU), and a vehicle integrated / integration unit (VIU). Among them, the DC may include a cockpit domain controller (CDC).

[0091] As an example of software implementation, the detection device 20 may include software functional units. As another example of a software functional unit, the detection device 20 may include one or more of the following: an executable computer program, computer code, or computer instructions, where "executable" means capable of running on a processor or computing instance. As yet another example of a software functional unit, the detection device 20 may include computing instances, including virtual machines, containers, etc. A virtual machine is a computer system simulated by software, possessing complete hardware system functionality and running in an isolated environment. A container is an isolated environment obtained by packaging applications and their dependencies. Figure 1 The following is an example of the detection device 20 being installed inside the terminal 100. In some cases, the detection device 20 can also be installed outside the terminal 100, such as when the detection device 20 is a server, cloud, or something similar, and is connected to the terminal 100.

[0092] The above describes the various modules in terminal 100. Combined with... Figure 1 It can be seen that there may be connections between multiple modules in terminal 100. Figure 1 The example illustrates a module connection method. It should be understood that for two modules that are connected, such as sensor system 10 and detection device 20, the connection method can be wired, wireless, or a combination of wired and wireless connections. This connection includes direct and / or indirect connections; for example, sensor system 10 and detection device 20 may be indirectly connected through intermediate nodes, which may be one or more.

[0093] Optionally, the multiple modules included in the terminal 100 can be set up independently, or they can be integrated together, or some can be designed independently and some can be integrated together. For example, the sensor system 10 and the detection device 20 can be integrated together.

[0094] The architecture of the terminal has been described above. It should be noted that due to technological advancements or changes in application scenarios, some terminals may include only some of the aforementioned components, or they may include even more components.

[0095] The apparatus of the present application has been described above. The method of the present application is described below.

[0096] Please see Figure 3 , Figure 3 This is a schematic flowchart of a hail detection method provided in an embodiment of this application. Optionally, this method is applied to a device with data processing capabilities, such as... Figure 1 The detection device 20 shown may refer to the software and / or hardware modules within the detection device 20. For ease of description, the following description will use the detection device as the executing entity.

[0097] like Figure 3 The hail detection methods shown include S301 and S302. S301 and S302 are detailed below: S301, the detection device acquires the elastic wave signal from the elastic wave sensor.

[0098] Among them, the detection device is a device with data processing capabilities, such as... Figure 1 The detection device 20 shown.

[0099] Elastic wave sensors (such as) Figure 1 The elastic wave sensor 101 shown is installed in the terminal (e.g., Figure 1 The terminal 100 shown propagates the elastic wave signal on its surface. It should be noted that this embodiment does not limit the number or location of the elastic wave sensors, as long as they can detect the elastic wave signal propagating on the terminal's surface.

[0100] The following description primarily uses a vehicle as an example to illustrate the embodiments of this application. For instance, an elastic wave sensor is installed in the vehicle, and the elastic wave signal propagates on the surface of the vehicle body. Combined with... Figure 2 The vehicle includes two elastic wave sensors, located at the front and rear of the vehicle, respectively.

[0101] For details on the elastic wave signal, please refer to the relevant description of the aforementioned elastic wave sensor 101.

[0102] In some possible implementations, the elastic wave sensor can collect elastic wave signals propagating from the vehicle's surface and send these signals to the detection device via a connection between the sensor and the device. This allows the detection device to acquire the elastic wave signals from the sensor. Optionally, the elastic wave sensor converts the elastic wave signals propagating from the vehicle's surface into electrical signals and sends these converted signals to the detection device. In this case, the elastic wave signals acquired by the detection device are the converted electrical signals.

[0103] In some possible implementations, the elastic wave sensor can be continuously operational, constantly collecting elastic wave signals propagating from the vehicle's surface. Alternatively, the elastic wave sensor may be briefly operational under certain conditions (such as weather forecasts indicating hail or hail protection being enabled), and in sleep mode when these conditions are not met to reduce power consumption. For example, under certain conditions, the detection device can send a command to the elastic wave sensor to wake it up. Exemplarily, the elastic wave sensor can send the collected elastic wave signals to the detection device in real time, or it can send the signals periodically (i.e., at regular intervals). The transmission period of the elastic wave sensor can be flexibly set based on actual usage requirements.

[0104] S302, the detection device compares the elastic wave signal with the signal characteristics under the hail scene to obtain the first detection result.

[0105] The elastic waves generated when hail hits a vehicle have specific signal characteristics. The signal characteristics in a hail scenario can be the signal characteristics of the elastic wave signal that propagates on the surface of the vehicle when hail hits it in a pre-collected (or calibrated or tested) hail scenario.

[0106] The first detection result is used to indicate whether a hail scene has been detected. For example, the first detection result is used to indicate whether a hail scene has been detected at the vehicle's location. Alternatively, the first detection result is used to indicate whether a hail scene has been detected based on an elastic wave signal.

[0107] In some possible implementations, the detection device compares the elastic wave signal with the signal characteristics of a hail scene to obtain a first detection result. There are multiple possible implementations, such as threshold judgment, network model, image detection, and similarity judgment, among others. These implementations can distinguish the difference between hail impacts and other scenes (such as user knocking or heavy rain scenes). For example, in a rain scene, the elastic wave signal waveform is a continuous random fluctuation with low (dispersed) energy density, a dominant frequency less than 500Hz, and a duration of several minutes to several hours. The false detection probability is moderate (requiring multiple accumulations), and it can be filtered out by the force and signal characteristics of a rain scene. In a hail scene, the elastic wave signal waveform is a discrete high-amplitude pulse with high (concentrated) energy density, a dominant frequency between 500Hz and 5kHz, and a duration of several minutes to tens of minutes. The false detection probability is extremely high (it can be triggered in a single instance).

[0108] The following implementation requires the use of hail time-domain features and elastic wave signal time-domain features. We will first introduce the hail time-domain features and elastic wave signal time-domain features. For example, the signal features in a hail scenario include hail time-domain features, which are the time-domain features of the elastic wave signal propagating on the vehicle surface when hail collides with (or impacts) a vehicle. The detection device can obtain the time-domain features of the elastic wave signal based on the elastic wave signal from the elastic wave sensor. Further, the detection device compares the time-domain features of the elastic wave signal with the hail time-domain features to obtain a first detection result. The elastic wave signal includes at least one elastic wave pulse (also simply referred to as a pulse). Time-domain features can include one or more of pulse width, amplitude, and rise slope. Pulse width refers to the duration of the pulse in the time domain, such as the time taken for the pulse to travel from its start point to its end point. Amplitude refers to the maximum displacement of medium particles from their equilibrium position during the propagation of the elastic wave. The amplitude of the elastic wave signal acquired by the elastic wave sensor can be understood as the peak voltage or signal height read by the elastic wave sensor. The rise slope is the rate at which a pulse rises from a low amplitude to a high amplitude, and it can characterize the steepness of the rise in amplitude of an elastic wave signal.

[0109] The following sections will introduce these implementation methods respectively.

[0110] Implementation Method 1: Threshold Judgment. The detection device can obtain one or more judgment thresholds based on the time-domain characteristics of hail, such as one or more of the following: amplitude judgment threshold, rise slope judgment threshold, and pulse width judgment threshold. When the time-domain characteristics of the elastic wave signal satisfy one or more of the following: amplitude greater than the amplitude judgment threshold, rise slope greater than the rise slope judgment threshold, and pulse width less than the pulse width judgment threshold, the detection device considers hail to have been detected, and the detection result obtained by the detection device (such as the first detection result) indicates that a hail scene has been detected. Any one of the amplitude judgment threshold, rise slope judgment threshold, and pulse width judgment threshold can be flexibly set according to actual usage requirements.

[0111] The second implementation method involves similarity assessment. The detection device compares the time-domain features of the elastic wave signal with the time-domain features of the hail to obtain the similarity between the two. When the similarity is greater than or equal to a similarity threshold, the detection device considers it highly likely that hail has been detected, and the detection result obtained by the device (such as the first detection result) indicates that a hail scene has been detected. The similarity threshold can be flexibly set based on actual usage requirements, such as 80%, 85%, etc.

[0112] The third implementation method involves a network model (or recognition algorithm). We pre-learn the differences between "hail elastic wave signals" and "non-hail elastic wave signals" using a network model (such as a machine learning model, deep learning model, or recognition algorithm). For example, we pre-label the time-domain features of some elastic wave signals and the corresponding hail detection results. These labeled time-domain features, the corresponding hail detection results, and the hail time-domain features are used as training samples for the network model (or recognition algorithm) to train a hail detection model (or hail recognition algorithm). For instance, in actual use, the detection device inputs the time-domain features of the elastic wave signal into the trained hail detection model, and the hail detection model can directly output the hail detection result corresponding to the time-domain features of the elastic wave signal, such as outputting the first detection result. Optionally, in actual use, the detection device directly inputs the elastic wave signal into the trained hail detection model, and the hail detection model can also output the hail detection result corresponding to the elastic wave signal, such as outputting the first detection result. Optionally, the elastic wave sensor can analyze the contact point location and force based on the collected elastic wave signal. The contact point location and force can be input into a trained hail detection model, thereby enabling the trained hail detection model to distinguish the difference between hail impact and other scenarios (such as user knocking or rainstorm scenarios).

[0113] Implementation Method Four: Graphical Detection. The detection device obtains the waveform of the elastic wave signal propagating on the vehicle surface when hail impacts a vehicle from historical data (or test data or calibration data), and derives the hail waveform based on these waveforms. The detection device measures the waveform similarity between the real-time acquired elastic wave signal and the hail waveform by using geometric overlap or distance, etc. When the waveform similarity is greater than or equal to the waveform similarity threshold, the detection device considers hail to have been detected, and the detection result obtained by the detection device (such as the first detection result) indicates that a hail scene has been detected. The waveform similarity threshold can be flexibly set based on actual usage requirements, such as 80%, 85%, etc.

[0114] It should be noted that the above implementation methods can be combined with each other. For example, multiple implementation methods can be used simultaneously to detect whether hail has occurred. When the detection results of most (or all) of the multiple implementation methods indicate that a hail scene has been detected, the first detection result indicates that a hail scene has been detected. The combination schemes will not be illustrated one by one here.

[0115] The above mainly uses the signal characteristics of hail scenarios, including the time-domain characteristics of hail, as an example to introduce the hail detection process. Similarly, the signal characteristics of hail scenarios can also include one or more of the frequency domain characteristics of elastic wave signals under hail scenarios (such as center frequency, spectral centroid, frequency band energy ratio, peak frequency distribution, etc.), time-frequency domain characteristics (such as time-frequency entropy, instantaneous frequency change, wavelet coefficient modulus maxima, etc.), and energy and statistical characteristics (such as zero-crossing rate, root mean square energy change rate, sparsity, etc.). These characteristics can also be used to determine whether a hail scenario has been detected, and the detection process is similar to that described above, so they will not be illustrated one by one here.

[0116] In some possible implementations, the detection device can filter out elastic wave signals whose time-domain characteristics do not meet the detection conditions, obtaining a filtered elastic wave signal. The detection device then obtains a first detection result based on the time-domain characteristics of the filtered elastic wave signal and the time-domain characteristics of hail. The detection conditions include one or more of the following: pulse width condition, amplitude condition, and rise slope condition. Elastic wave signals that do not meet the detection conditions can be understood as "weak signals" that clearly do not conform to the time-domain characteristics of hail, or elastic wave signals that are highly unlikely to conform to the time-domain characteristics of hail. For example, the pulse width condition includes a pulse width less than a pulse width detection threshold, the amplitude condition includes an amplitude greater than an amplitude detection threshold, and the rise slope condition includes a rise slope greater than a rise slope detection threshold. Any one of the amplitude detection threshold, rise slope detection threshold, and pulse width detection threshold can be flexibly set based on actual usage requirements.

[0117] In some possible implementations, the detection device can use a time-accumulation method to compare the elastic wave signal with the signal characteristics of a hail scene to obtain a first detection result. For example, the time-accumulation method does not rely on a single signal occurrence, but rather on whether the number of times (or cumulative score) a "suspected hail event" occurs within a time window meets a condition. For instance, the elastic wave signal includes multiple pulses, and the detection device determines the number of matching pulses among these pulses. Matching pulses are those whose temporal characteristics match the temporal characteristics of hail; these matching pulses can be understood as "suspected hail events." Similarly, the detection device can determine the number of matching pulses among the multiple pulses based on one or more of the aforementioned threshold judgment, network model, image detection, and similarity judgment methods, which will not be exemplified here. When the number of matching pulses within a first duration is greater than or equal to a quantity threshold, the first detection result indicates that a hail scene was detected based on the elastic wave signal. Alternatively, when the number of matching pulses within a first duration is less than a quantity threshold, the first detection result indicates that a hail scene was not detected based on the elastic wave signal. The first duration is a preset time window, such as 500ms to 2000ms, which can be flexibly set according to actual usage requirements. Optionally, the detection device first filters the elastic wave signal, and then uses a time accumulation method to obtain a first detection result based on the filtered pulses. Similarly, the filtered elastic wave signal includes multiple pulses, and the detection device determines the number of matching pulses among the multiple filtered pulses, thereby obtaining the first detection result.

[0118] In some possible implementations, under extreme weather conditions (such as heavy rain or hail), the detection device can dynamically adjust the detection thresholds (such as amplitude detection thresholds and / or rise slope detection thresholds) in the detection conditions. For example, detection conditions may include amplitude exceeding the amplitude detection threshold and / or rise slope exceeding the rise slope detection threshold. If the weather forecast information at the vehicle's location indicates hail, or if the rainfall at the vehicle's location exceeds the rainfall threshold, the detection device may increase the amplitude detection threshold and / or rise slope detection threshold, such as by 3 to 5 times (which can be flexibly designed based on actual usage requirements). The rainfall at the vehicle's location can be provided by a rain sensor (or a solar rain sensor) in the vehicle. The rainfall threshold can be flexibly designed based on actual usage requirements.

[0119] The preceding text described how the detection device detects whether a hail scene is detected based on elastic wave signals. In practical applications, the detection device can also detect whether a hail scene is detected based on sensing data from other types of sensors. In this case, the first detection result can be used to indicate whether a hail scene is detected based on the elastic wave signal. For example, the detection device obtains at least one sensing detection result based on sensing data from at least one sensor. Each sensing detection result indicates whether a hail scene is detected based on sensing data from one or more corresponding sensors. Optionally, sensing data from one sensor is used to obtain one sensing detection result. Alternatively, sensing data from multiple sensors can be used to obtain one sensing detection result.

[0120] The following describes the sensing data of several other types of sensors that can be used to detect whether hail has been detected, and how the detection device obtains the sensing detection result from each type of sensing data.

[0121] Sensing data one: acceleration data. The detection device acquires data from an acceleration sensor (such as...). Figure 1 The acceleration data is obtained from the acceleration sensor 102 shown, which is installed in the vehicle. The acceleration data can be a low-frequency micro-vibration acceleration signal (or simply acceleration) generated by the vehicle body structure under stress. The detection device compares the acceleration data with the acceleration data characteristics under a hail scene to obtain a second detection result. This second detection result indicates whether a hail scene has been detected based on the acceleration data. Similarly, the detection device can compare the acceleration data with the acceleration data characteristics under a hail scene using one or more of the following methods: threshold judgment, network model, image detection, and similarity judgment, to obtain a second detection result. These methods are not illustrated here. The acceleration data can verify whether the low-frequency vibration is consistent with the vehicle body structure response, thus reducing the false detection rate. For example, the acceleration data characteristics under a hail scene include the frequency domain characteristics of the acceleration data when hail collides with the vehicle. Based on the acceleration data, the detection device obtains the frequency domain characteristics of the acceleration data, which include one or more of the following: dominant frequency, high-frequency energy ratio, and decay time constant. The detection device compares the frequency domain characteristics of the acceleration data with those of the acceleration data during a hail collision with a vehicle to obtain a second detection result. This is similar to the first detection result obtained by comparing the time domain characteristics of the elastic wave signal with those of the hail, as described earlier; examples will not be repeated here. Here, the dominant frequency refers to the frequency with the highest energy (amplitude) in the acceleration spectrum. The high-frequency energy ratio refers to the percentage of high-frequency energy in the acceleration that constitutes the total energy. The decay time constant is a parameter indicating how quickly the acceleration energy dissipates after the impact.

[0122] Perceptual data two, sound data. The detection device acquires data from the microphone (e.g., Figure 1The sound data from microphone 103 (shown) is located on the vehicle. The sound data can be the sound data surrounding the vehicle. The detection device can obtain a first perception detection result based on the sound data, which indicates whether a hail scene has been detected based on the sound data. For example, the detection device compares the sound data with the sound data features under a hail scene to obtain the first perception detection result. Similarly, the detection device can compare the sound data with the sound data features under a hail scene using one or more of the following methods: threshold judgment, network model, image detection, and similarity judgment, to obtain the first perception detection result. These methods will not be illustrated here. The sound data features under a hail scene include one or more of the following: the sound of hail falling (or impacting) and voiceprints.

[0123] The third type of sensory data is visible light image data. The detection device acquires data from a visible light image sensor (such as...). Figure 1 The visible light image data shown is from a visible light image sensor 104, which is installed on the vehicle. In this case, the visible light image data can be the visible light image data surrounding the vehicle. The detection device can obtain a second perception detection result based on the visible light image data, which indicates whether a hail scene has been detected based on the visible light image data. For example, the detection device compares the visible light image data with the features of visible light image data under a hail scene to obtain the second perception detection result. Similarly, the detection device can compare the visible light image data with the features of visible light image data under a hail scene through one or more of the following methods: threshold judgment, network model, image detection, and similarity judgment, to obtain the second perception detection result. These methods will not be exemplified here. The features of visible light image data under a hail scene include one or more of the following: hail geometric shape features (high roundness), color features (semi-transparent, milky white, white, or grayish white, etc.), and optical and illumination features (highlights / specular reflections).

[0124] Sensing data four: Infrared image data. The detection device acquires data from an infrared image sensor (such as...). Figure 1The infrared image data shown is from an infrared image sensor 105, which is installed on the vehicle. In this case, the infrared image data can be infrared image data of the area surrounding the vehicle. The detection device can obtain a third-sensor detection result based on the infrared image data. This third-sensor detection result indicates whether a hail scene has been detected based on the infrared image data. For example, the detection device compares the infrared image data with the features of infrared image data under a hail scene to obtain the third-sensor detection result. Similarly, the detection device can compare the infrared image data with the features of infrared image data under a hail scene through one or more of the following methods: threshold judgment, network model, image detection, and similarity judgment, to obtain the third-sensor detection result. These methods will not be exemplified here. The features of infrared image data under a hail scene include one or more of the following: thermal contrast features (low-temperature spots, thermal differences with the background, etc.) and spatial distribution features (discrete low-temperature particles).

[0125] Sensing data five, point cloud data. The detection device acquires data from radar-type sensors (such as...). Figure 1 The point cloud data shown is from a radar-type sensor 106, which is installed on the vehicle. In this case, the point cloud data can be the point cloud data surrounding the vehicle. The detection device can obtain a fourth perception detection result based on the point cloud data, which indicates whether a hail scene has been detected based on the point cloud data. For example, the detection device compares the point cloud data with the point cloud data features under a hail scene to obtain the fourth perception detection result. Similarly, the detection device can compare the point cloud data with the point cloud data features under a hail scene through one or more of the following methods: threshold judgment, network model, image detection, and similarity judgment, to obtain the fourth perception detection result. These methods will not be illustrated here. The point cloud data features under a hail scene include one or more of the following: geometric features under a hail scene (sphericity / compactness, cluster density, or size estimation, etc.), reflection intensity features (high reflection intensity, intensity consistency), and surface smoothness (relatively smooth).

[0126] The above mainly uses the sensing data from one sensor to obtain a single sensing detection result as an example. It should be noted that the various sensing data can be combined. For example, multiple sensing data can be used simultaneously to detect whether hail has occurred; that is, sensing data from multiple sensors can be used to obtain a single sensing detection result. For instance, the detection device can combine point cloud data, visible light image data, and infrared image data to obtain a sensing detection result. Of course, there are many other ways to combine data, but these will not be illustrated here.

[0127] In some possible implementations, the elastic wave sensor and the accelerometer can be continuously operational. When the detection device detects a hail scene based on the elastic wave signal, and / or when a second detection result indicates that a hail scene is detected based on acceleration data, it generates a first instruction to wake up at least one sensor, which includes one or more of a microphone, a visible light image sensor, an infrared image sensor, and a radar-type sensor. In this case, the elastic wave sensor and the accelerometer can be the primary sensors for hail detection and also the wake-up source for at least one sensor to perform hail detection; the at least one sensor can also be an auxiliary sensor for hail detection. The detection device can wake up the auxiliary sensor (i.e., the sensor used to detect whether a hail scene is detected) through primary sensing (i.e., using the elastic wave sensor and the accelerometer to detect whether a hail scene is detected).

[0128] The above examples illustrate various detection methods for a detection device to determine whether a hail scene has been detected, namely, the process for determining the aforementioned hail detection results (such as the first detection result, second detection result, first perception detection result, second perception detection result, third perception detection result, and fourth perception detection result, which are also referred to as detection results in some embodiments). Further, the detection device can control one or more components in the vehicle to perform hail protection operations based on these hail detection results. For example, if one or two (the number can be flexibly designed based on actual usage needs) hail detection results indicate that a hail scene has been detected, the device controls one or more components in the vehicle to perform hail protection operations. For instance, if at least two (or one, the number can be flexibly designed based on actual usage needs) of the first detection result, second detection result, and at least one perception detection result indicate that a hail scene has been detected, the device controls one or more components in the vehicle to perform hail protection operations.

[0129] For example, the primary sensor has a higher priority, and the detection device controls one or more components in the vehicle to perform hail protection operations based at least on the first detection result. For instance, if the first detection result indicates that a hail scene has been detected based on elastic wave signals, and / or if the second detection result indicates that a hail scene has been detected based on acceleration data, the detection device controls one or more components in the vehicle to perform hail protection operations. As another example, if the first detection result indicates that a hail scene has been detected based on elastic wave signals, and the confidence level of the first detection result is greater than or equal to a first threshold, the detection device controls one or more components in the vehicle to perform hail protection operations. The confidence level of the first detection result is used to indicate the reliability of the first detection result. The first threshold is a pre-set confidence threshold. When the confidence level is greater than the first threshold, the first detection result is considered to have a high degree of reliability and a low probability of false detection. The first threshold can be flexibly set based on actual usage requirements, such as 85%, 90%, etc. For example, if the detection device controls one or more components in the vehicle to perform hail protection operations when the first detection result indicates that a hail scene has been detected based on the elastic wave signal, and the confidence level of the first detection result is less than a first threshold, and at least one of the second detection result and at least one sensing detection result indicates that a hail scene has been detected; or if the detection device controls one or more components in the vehicle to perform hail protection operations when the first detection result indicates that a hail scene has not been detected based on the elastic wave signal, and at least two of the second detection result and at least one sensing detection result indicate that a hail scene has been detected.

[0130] The detection device controls one or more components in the vehicle to perform hail protection operations, including controlling one or more components in the vehicle to perform hail protection operations that do not require vehicle movement and controlling one or more components in the vehicle to perform hail protection operations that require vehicle movement. For example, hail protection operations that do not require vehicle movement include sending a notification message to a terminal associated with the vehicle (such as the owner's mobile phone) to indicate that hail has occurred at the vehicle's location, acquiring an image (or video) of the environment around the vehicle and sending the image to the terminal (i.e., allowing the user to remotely view the vehicle's surroundings via images or videos), closing the vehicle doors, closing the vehicle windows, or rolling up the vehicle's rearview mirrors, etc. Hail protection operations that require vehicle movement include controlling the vehicle to move to a safe parking position, or adjusting the vehicle's position, etc. Optionally, there are many ways to send notification messages to the terminal associated with the vehicle, such as sending notification messages via SMS, applications, telephone, or one or more other methods.

[0131] In some possible implementations, the detection device compares the elastic wave signal with signal characteristics under a hail scenario to obtain a first detection result and a confidence level of the first detection result; that is, the detection device can obtain the confidence level of the first detection result simultaneously when it obtains the first detection result. Similarly, the detection device compares acceleration data with acceleration data characteristics under a hail scenario to obtain a second detection result and a confidence level of the second detection result; that is, the detection device can obtain the confidence level of the second detection result simultaneously when it obtains the second detection result. Similarly, the detection device obtains at least one sensing detection result and a confidence level of at least one sensing detection result based on sensing data from at least one sensor; that is, the detection device can obtain the confidence level of the sensing detection result simultaneously when it obtains the sensing detection result.

[0132] Furthermore, the detection device can control one or more components in the vehicle to perform hail protection operations based on a first detection result, the confidence level of the first detection result, a second detection result, the confidence level of the second detection result, at least one sensing detection result, and the confidence level of at least one sensing detection result. Optionally, the confidence level value is between 0 and 1, with a higher value indicating a higher degree of confidence. For example, when the detection result indicates that a hail scene has been detected, the value of the detection result can be recorded as 1, and when the detection result indicates that a hail scene has not been detected, the value of the detection result can be recorded as 0. The detection device adds the product of the value of each detection result and the confidence level of each detection result, and when the sum is greater than a first preset threshold, it controls one or more components in the vehicle to perform hail protection operations. This first preset threshold can be flexibly set based on actual usage requirements. Optionally, different detection results can have different weights; for example, the weight of the detection result obtained by the primary sensing can be designed to be greater than the weight of the detection result obtained by the auxiliary sensing. The detection device can control one or more components in a vehicle to perform hail protection operations based on a first detection result, the confidence level of the first detection result, the weight of the first detection result, a second detection result, the confidence level of the second detection result, the weight of the second detection result, at least one sensing detection result, the confidence level of at least one sensing detection result, and the weight of at least one sensing detection result. Optionally, the weight value is between 0 and 1, with a larger value indicating a larger weight. For example, the detection device adds the products of each detection result value, the confidence level of each detection result, and the weight of each detection result, and when the sum is greater than a second preset threshold, it controls one or more components in the vehicle to perform hail protection operations. This second preset threshold can be flexibly set based on actual usage requirements.

[0133] The following describes the specific implementation process of several detection devices controlling one or more components in a vehicle to perform hail protection operations.

[0134] In some possible implementations, the detection device controls one or more components in the vehicle to perform hail protection operations, including controlling the vehicle to a safe parking location. The detection device can detect whether a sheltered, safe parking location exists around the vehicle. If such a location exists, the detection device can control the vehicle to a safe parking location based on the aforementioned hail detection result (e.g., an indication that hail has been detected), such as if a user remotely controls the vehicle to a safe parking location via a terminal associated with the vehicle. For example, if a sheltered, safe parking location exists around the vehicle, the detection device controls the vehicle to a safe parking location based at least on a first detection result (e.g., an indication that hail has been detected). A sheltered, safe parking location includes one or more of the following: a carport with a sturdy roof, an underground parking lot, an indoor / underground garage, or a parking space under a bridge.

[0135] In some other possible implementations, the detection device controls one or more components in the vehicle to perform hail protection operations, including adjusting the vehicle's orientation. The detection device may determine the wind direction at the vehicle's location based on second sensing data, which may include sensing data from one or more sensors such as visible light image sensors, infrared image sensors, radar sensors, and inertial measurement units. Based on the aforementioned hail detection results (e.g., an indication that a hail scene has been detected), the detection device may adjust the vehicle's orientation so that the vehicle's side faces the wind direction at the vehicle's location, such as if a user remotely adjusts the vehicle's orientation via a terminal associated with the vehicle. For example, the detection device may adjust the vehicle's orientation so that the vehicle's side faces the wind direction at least based on a first detection result (e.g., an indication that a hail scene has been detected). Optionally, the detection device may determine a wind-blocking wall based on the wind direction at the vehicle's location, and control the vehicle to drive to the wind-blocking wall, so that the wind-blocking wall can block some of the wind and mitigate the impact of the hail. Optionally, if a safe, sheltered parking location exists around the vehicle, the detection device prioritizes guiding the vehicle to that location. If no safe, sheltered parking location exists around the vehicle, the detection device adjusts the vehicle's position so that the side of the vehicle faces the wind direction, or guides the vehicle to a wall that provides shelter from the wind, in order to minimize vehicle maintenance costs.

[0136] In some possible implementations, the detection device can obtain a final detection result based on the aforementioned hail detection results. This final detection result indicates whether hail has occurred at the vehicle's location. Similar to how the detection device controls one or more components in the vehicle to perform hail protection operations based on these hail detection results, the methods for determining the final detection result will not be described in detail here. If the final detection result indicates that hail has occurred at the vehicle's location, the detection device controls one or more components in the vehicle to perform hail protection operations. Optionally, the detection device sends the final detection result to a terminal associated with the vehicle and supports remote user confirmation (such as user-inputted information), or the detection device outputs the final detection result through the vehicle's interactive devices (such as display devices and / or audio devices, etc.) and supports user on-vehicle confirmation (such as receiving user-inputted information). Display devices include one or more of the following: physical screens (such as central control screens), instrument panels, projection systems, smart entities, or button panels. Projection systems include, for example, light field screens, head-up displays (HUDs), or other projection systems. When the detection device receives input information from the user (such as an instruction to perform hail protection operation or activate hail protection mode), the detection device can control one or more components in the vehicle to perform hail protection operation.

[0137] In other possible implementations, the hail protection operations corresponding to different hail levels may differ. For example, there may be at least two hail levels with different hail protection operations, thereby enabling different hail protection operations to be formulated based on the degree of hail damage. For instance, the at least two hail levels include a first hail level and a second hail level. The hail protection operations corresponding to the first hail level include the aforementioned hail protection operations that do not require vehicle movement. The hail protection operations corresponding to the second hail level include the aforementioned hail protection operations that require vehicle movement. For example, the hail protection operations corresponding to the first hail level include one or more of the following: sending a warning message to a terminal associated with the vehicle; acquiring an environmental image of the vehicle's surroundings and sending the image to the terminal; closing the vehicle doors; closing the vehicle windows; or retracting the vehicle's rearview mirrors. The hail protection operations corresponding to the second hail level include the hail protection operations corresponding to the first hail level, and also include one or more of the following: controlling the vehicle to a safe parking position; or adjusting the vehicle's position. For instance, the first hail level is a low hail level, causing minimal damage to the vehicle, in which case the vehicle is not moved. The second level of hail is a high hail level, which can cause significant damage to vehicles. Vehicles need to be moved to minimize the damage.

[0138] For example, the detection device controls one or more components in the vehicle to perform hail protection operations based on the aforementioned hail detection results and the hail level at the vehicle's location. For instance, if the final detection result indicates hail at the vehicle's location, the detection device controls one or more components in the vehicle to perform hail protection operations corresponding to the hail level at the vehicle's location. Alternatively, the detection device controls one or more components in the vehicle to perform hail protection operations based at least on the first detection result and the hail level at the vehicle's location. For example, hail levels include small hail, medium hail, large hail, and extra-large hail. For example, small hail causes minimal damage to the vehicle, possibly leaving only minor marks. In this case, the hail protection operation corresponding to small hail includes the hail protection operation corresponding to the aforementioned first hail level, such as sending a warning message to a terminal associated with the vehicle (e.g., via SMS or an application). The hail protection operation corresponding to small hail may also include controlling the vehicle to move to a nearby location, such as under a tree, to mitigate the damage. For example, moderate hail can cause significant damage to vehicles, potentially resulting in dents or scratches on the paint, or slight deformation of thin metal panels. In this case, hail protection measures include sending alerts to the vehicle's associated terminals (e.g., via SMS or an app, while simultaneously reminding the user by phone), and directing the vehicle to a nearby location that provides shelter (such as a wall to block the wind), under a tree, or other suitable location for slowing down (or the aforementioned safe parking location). Similarly, large hail can cause substantial damage to vehicles, potentially resulting in noticeable dents or cracks in the body panels and shattered windshields. In this case, hail protection measures include sending alerts to the vehicle's associated terminals (e.g., via SMS or an app, while simultaneously reminding the user by phone), and immediately directing the vehicle to a sheltered location such as a carport, underground garage, or under a bridge (or the aforementioned safe parking location) upon detection of hail. For example, extremely large hailstones have immense destructive power, capable of penetrating vehicle body panels, shattering windshields, damaging sunroofs, and damaging the vehicle's internal structure. In such cases, hail protection measures for extremely large hailstones include sending alerts to terminals associated with the vehicle (e.g., via SMS or an app, and simultaneously notifying the user by phone), and immediately moving the vehicle to a sheltered location such as a carport, underground garage, or under a bridge (or the aforementioned safe parking location) upon detection of hail. It can be seen that the hail protection measures for small hailstones differ from those for large hailstones; that is, there are at least two different hailstone levels for which different hail protection measures are required.

[0139] The following describes how the detection device determines the hail level at the vehicle's location.

[0140] In some possible implementations, the detection device can determine the hail level at the vehicle's location based on the elastic wave signal. For example, different hail levels correspond to different signal feature ranges, and the detection device can determine the hail level at the vehicle's location based on the feature range to which the elastic wave signal's features belong. For instance, the signal features include time-domain features; the elastic wave signal includes multiple matched pulses, and the detection device determines the time-domain feature range to which the time-domain features of each matched pulse belong. The detection device determines the hail level at the vehicle's location based on the time-domain feature range to which the time-domain features of each matched pulse belong. For example, the detection device divides the time-domain feature range (such as intervals and / or thresholds) according to one or more of the pulse width, amplitude, and rise slope of the elastic wave signal; different time-domain feature ranges correspond to different hail levels, thereby obtaining the hail level at the vehicle's location. In some possible implementations, the first detection result may include the hail level, such as no hail, small hail, medium hail, large hail, and extra-large hail. When the hail level is no hail, the first detection result is used to indicate that no hail was detected. When the hail level is small, medium, large, or very large, the first detection result is used to indicate that a hail scene has been detected. Only temporal characteristics are illustrated here; the detection device can also determine the hail level based on other signal characteristics mentioned above, which will not be illustrated here.

[0141] In other possible implementations, the detection device determines the hail size at the vehicle's location based on first sensing data, which includes sensing data from one or more sensors selected from visible light image sensors, infrared image sensors, and radar sensors. The detection device then determines the hail severity level at the vehicle's location based on the hail size. For example, a hail size (e.g., diameter) less than 5mm is classified as small hail; a hail size (e.g., diameter) between 5mm and 20mm is classified as medium hail; a hail size (e.g., diameter) between 20mm and 50mm is classified as large hail; and a hail size (e.g., diameter) greater than 50mm is classified as extra-large hail. Optionally, in scenarios with good ambient light, the detection device (e.g., using a network model or recognition algorithm) can detect the hail size based on visible light image data and point cloud data with a small error (e.g., within 5mm). In low-light conditions such as at night, the detection device uses temperature differences in different areas of infrared image data to detect hail size. By using point cloud data for size verification, the accuracy of hail size detection can be improved.

[0142] In other possible implementations, the detection device determines the temporal feature range to which the temporal features of each of the multiple matching pulses belong, with different hail levels corresponding to different temporal feature ranges. The detection device determines a first hail level based on the temporal feature range to which the temporal features of each matching pulse belong. The detection device determines the hail size at the vehicle's location based on first sensing data, which includes sensing data from one or more sensors selected from visible light image sensors, infrared image sensors, and radar sensors. The detection device determines the hail level at the vehicle's location based on the first hail level and the hail size at the vehicle's location. In this case, it can be understood that the hail level obtained by the detection device based on the elastic wave signal (i.e., the first hail level) is a preliminary level classification, and the hail size obtained by the detection device based on the first sensing data is used to verify the first hail level to improve the accuracy of the determined hail level. Optionally, if the hail level obtained by the detection device based on the elastic wave signal is not hail-free, the detection device determines the hail size at the vehicle's location based on the first sensing data. For example, if the hail level obtained by the detection device based on the elastic wave signal is not hail-free, the detection device sends a wake-up command to the visual sensor (i.e., a visible light image sensor and / or an infrared image sensor) and / or a radar sensor (to wake up the visual sensor and / or the radar sensor). Then, based on the first perception data from the woken sensor (such as the visual sensor and / or the radar sensor), the detection device determines the hail size at the vehicle's location.

[0143] exist Figure 3 In the illustrated embodiment, the detection device utilizes an elastic wave sensor to detect hail, completely unaffected by lighting conditions, rain or fog obstruction, or visual interference. It can accurately identify hail impacts at night or in severe weather, ensuring continuous and reliable detection. Furthermore, the elastic waves generated when hail impacts a vehicle body possess specific signal characteristics (such as pulse width, amplitude, and rise slope). By comparing the elastic wave signals propagating on the vehicle's surface with the signal characteristics under hail conditions, the detection device can effectively distinguish between hail, raindrops, branch scratches, flying pebbles, and human impacts, improving the accuracy of hail detection. Additionally, elastic waves propagate extremely quickly within the vehicle's metal or composite materials; detecting hail using elastic waves allows for earlier detection of the initial impact, facilitating earlier hail protection for the vehicle and minimizing potential losses for the user.

[0144] Please see Figure 4 , Figure 4This is a schematic diagram of a detection device provided in an embodiment of this application. Optionally, the detection device 20 can be a standalone device, such as a controller, processor, or SOC. Alternatively, the detection device 20 can be a component within a standalone device, such as a chip, integrated circuit, or software module (cloud service). The detection device 20 is used to implement the aforementioned method, such as... Figure 3 The hail detection method shown, or at least part of the method performed by the detection device, etc.

[0145] The detection device 20 includes a transceiver unit 201 and a processing unit 202. The transceiver unit 201 is used to perform one or more operations such as receiving, acquiring, reading, or sending, and optionally, it is also used to perform other operations involved in the aforementioned method embodiments. The processing unit 202 is used to perform one or more operations such as processing, determining, generating, calculating, encoding, decoding, reasoning, or compressing, and optionally, it is also used to perform other operations involved in the aforementioned method embodiments. It should be understood that the unit division and naming here are only illustrative; in specific implementations, some units may be combined together, or one unit may be split into multiple units.

[0146] In one possible implementation, the transceiver unit 201 is used to acquire elastic wave signals from an elastic wave sensor installed on the vehicle, and the elastic wave signals propagate along the vehicle's surface. The processing unit 202 is used to compare the elastic wave signals with signal characteristics under a hail scene to obtain a first detection result, which indicates whether a hail scene has been detected.

[0147] In another possible implementation, the first detection result is used to indicate whether a hail scene has been detected based on the elastic wave signal. The transceiver unit 201 is also used to acquire acceleration data from an acceleration sensor installed on the vehicle. The processing unit 202 is further used to compare the acceleration data with the acceleration data characteristics under a hail scene to obtain a second detection result, which is used to indicate whether a hail scene has been detected based on the acceleration data.

[0148] In another possible implementation, the processing unit 202 is further configured to generate a first instruction when a first detection result indicates that a hail scene has been detected based on elastic wave signals, and / or a second detection result indicates that a hail scene has been detected based on acceleration data. The first instruction is used to wake up at least one sensor, which includes one or more of a microphone, a visible light image sensor, an infrared image sensor, and a radar-type sensor. The processing unit 202 is also configured to obtain at least one perception detection result based on perception data from the at least one sensor, each perception detection result indicating whether a hail scene has been detected based on perception data from one or more corresponding sensors.

[0149] In another possible implementation, the processing unit 202 is further configured to control one or more components in the vehicle to perform hail protection operations, based at least on the first detection result.

[0150] In another possible implementation, controlling one or more components in the vehicle to perform hail protection operations includes one or more of the following: sending a notification message to a terminal associated with the vehicle to indicate that hail has occurred at the vehicle's location; acquiring an image of the environment around the vehicle and sending the image to the terminal; closing the vehicle's doors; closing the vehicle's windows; and retracting the vehicle's rearview mirrors.

[0151] In another possible implementation, controlling one or more components in the vehicle to perform hail protection operations also includes one or more of the following: controlling the vehicle to a safe parking position; or adjusting the vehicle's position.

[0152] In another possible implementation, the processing unit 202 is further configured to control one or more components in the vehicle to perform hail protection operations based at least on the first detection result and the hail level at the vehicle's location, wherein there are at least two hail levels for which the hail protection operations are different.

[0153] In another possible implementation, at least two hail levels include a first hail level and a second hail level. The hail protection operation corresponding to the first hail level includes one or more of the following: sending a warning message to a terminal associated with the vehicle; acquiring an image of the environment around the vehicle and sending the image to the terminal; closing the vehicle doors; closing the vehicle windows; or rolling up the vehicle's rearview mirrors. The hail protection operation corresponding to the second hail level includes the hail protection operation corresponding to the first hail level, and further includes one or more of the following: controlling the vehicle to a safe parking position; or adjusting the vehicle's position.

[0154] In another possible implementation, the processing unit 202 is further configured to control one or more components in the vehicle to perform hail protection operations when the first detection result indicates that a hail scene has been detected based on the elastic wave signal and the confidence level of the first detection result is greater than or equal to a first threshold.

[0155] In another possible implementation, the processing unit 202 is further configured to control one or more components in the vehicle to perform hail protection operations when the first detection result indicates that a hail scene has been detected based on the elastic wave signal, and the confidence level of the first detection result is less than a first threshold, and at least one of the second detection result and at least one sensing detection result indicates that a hail scene has been detected.

[0156] In another possible implementation, the processing unit 202 is further configured to compare the elastic wave signal with the signal characteristics under the hail scene to obtain a first detection result and the confidence level of the first detection result.

[0157] In another possible implementation, the processing unit 202 is further configured to control one or more components in the vehicle to perform hail protection operations if at least two of the first detection result, the second detection result, and at least one perception detection result indicate that a hail scene has been detected.

[0158] In another possible implementation, the signal characteristics in the hail scenario include hail time-domain characteristics, which are the time-domain characteristics of the elastic wave signal propagating on the vehicle surface when hail collides with the vehicle. The processing unit 202 is further configured to obtain the time-domain characteristics of the elastic wave signal based on the elastic wave signal, the time-domain characteristics including one or more of pulse width, amplitude, and rise slope. The processing unit 202 is also configured to compare the time-domain characteristics of the elastic wave signal with the hail time-domain characteristics to obtain a first detection result.

[0159] In another possible implementation, the processing unit 202 is further configured to filter out elastic wave signals whose time-domain characteristics do not meet the detection conditions, thereby obtaining a filtered elastic wave signal. The detection conditions include one or more of pulse width conditions, amplitude conditions, and rise slope conditions. The processing unit 202 is further configured to obtain a first detection result based on the time-domain characteristics of the filtered elastic wave signal and the hail time-domain characteristics.

[0160] In another possible implementation, the detection conditions include an amplitude greater than an amplitude detection threshold and / or an upward slope greater than an upward slope detection threshold. The processing unit 202 is also used to increase the amplitude detection threshold and / or the upward slope detection threshold when the weather forecast information at the vehicle's location indicates that there is hail, or when the rainfall at the vehicle's location is greater than a rainfall threshold.

[0161] In another possible implementation, the filtered elastic wave signal includes multiple pulses, and the processing unit 202 is further configured to determine the number of matching pulses, where the matching pulses are those whose temporal characteristics match the hail temporal characteristics among the multiple pulses. If the number of matching pulses within a first duration is greater than or equal to a quantity threshold, the first detection result indicates that a hail scene has been detected based on the elastic wave signal. Alternatively, if the number of matching pulses within the first duration is less than the quantity threshold, the first detection result indicates that a hail scene has not been detected based on the elastic wave signal.

[0162] In another possible implementation, the acceleration data features in the hail scenario include the frequency domain features of the acceleration data when hail collides with a vehicle. The processing unit 202 is further configured to obtain the frequency domain features of the acceleration data based on the acceleration data, the frequency domain features including one or more of the following: dominant frequency, high-frequency energy ratio, and decay time constant. The processing unit 202 is also configured to compare the frequency domain features of the acceleration data with the frequency domain features of the acceleration data when hail collides with a vehicle to obtain a second detection result.

[0163] In another possible implementation, the elastic wave signal includes multiple matched pulses, where each matched pulse's time-domain feature matches the time-domain feature of the hail. The processing unit 202 is further configured to determine the time-domain feature range to which the time-domain feature of each matched pulse belongs, with different hail levels corresponding to different time-domain feature ranges. The processing unit 202 is also configured to determine the hail level at the vehicle's location based on the time-domain feature range to which the time-domain feature of each matched pulse belongs.

[0164] In another possible implementation, the elastic wave signal includes multiple matched pulses, where each matched pulse's time-domain feature matches the time-domain feature of the hail. Processing unit 202 is further configured to determine the time-domain feature range to which the time-domain feature of each matched pulse belongs, with different hail levels corresponding to different time-domain feature ranges. Processing unit 202 is further configured to determine a first hail level based on the time-domain feature range to which the time-domain feature of each matched pulse belongs. Processing unit 202 is further configured to determine the hail size at the vehicle's location based on first sensing data, which includes sensing data from one or more sensors selected from visible light image sensors, infrared image sensors, and radar sensors. Processing unit 202 is further configured to determine the hail level at the vehicle's location based on the first hail level and the hail size at the vehicle's location.

[0165] In another possible implementation, controlling one or more components in the vehicle to perform hail protection operations includes controlling the vehicle to a safe parking position, and the processing unit 202 is further configured to detect whether there is a sheltered safe parking position around the vehicle. If a sheltered safe parking position exists around the vehicle, the processing unit 202 is further configured to control the vehicle to the safe parking position based at least on a first detection result.

[0166] In another possible implementation, controlling one or more components in the vehicle to perform hail protection operations includes adjusting the vehicle's orientation. The processing unit 202 is further configured to determine the wind direction at the vehicle's location based on second sensing data, which includes sensing data from one or more sensors selected from visible light image sensors, infrared image sensors, radar-type sensors, and inertial measurement units. The processing unit 202 is also configured to adjust the vehicle's orientation so that the vehicle's side faces the wind direction at the vehicle's location, based at least on a first detection result.

[0167] In another possible implementation, the processing unit 202 is further configured to compare the acceleration data with the acceleration data characteristics under a hail scenario to obtain a second detection result and a confidence level of the second detection result. The processing unit 202 is further configured to obtain at least one perception detection result and a confidence level of at least one perception detection result based on perception data from at least one sensor. The processing unit 202 is further configured to control one or more components in the vehicle to perform hail protection operations based on the first detection result, the confidence level of the first detection result, the second detection result, the confidence level of the second detection result, at least one perception detection result, and the confidence level of at least one perception detection result.

[0168] For details on the operations performed by the detection device 20, please refer to the description of the system architecture and method embodiments above.

[0169] Please see Figure 5 , Figure 5 This is a schematic diagram of the structure of a computing device provided in an embodiment of this application, such as... Figure 5 The computing device 500 shown can be a standalone device, such as an MDC, GPU, or server. Alternatively, the computing device 500 can be a component within a standalone device, such as a chip, integrated circuit, or software module (e.g., cloud service, AI model). The computing device 500 is used to implement the aforementioned methods, such as... Figure 3 The hail detection method shown is, for example, at least a part of the method performed by the detection device.

[0170] The computing device 500 includes at least one processor and at least one memory. Optionally, the computing device 500 also includes a communication interface. Further optionally, the computing device 500 also includes connection lines, wherein the processor, communication interface, and / or memory are connected via the connection lines, and / or communicate with each other via the connection lines to transmit control signals and / or data signals. Wherein: A processor is a module with computing capabilities, including one or more of the following: arithmetic operations, logical operations, image-related operations, and artificial intelligence-related operations. Memory provides storage space, which can store data such as the operating system and computer programs. Memory can be one or a combination of several of the following: random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), or compact disc read-only memory (CD-ROM).

[0171] A communication interface can be used to provide information input or output to at least one processor, and / or to receive and / or send signals to externally transmitted signals. For example, a computing device may be a package containing chips or circuitry, and its communication interface may include interface circuitry. Alternatively, a computing device may be a communication-enabled device, and its communication interface may include data transmission interfaces such as Ethernet interfaces, serial data interfaces, and parallel data interfaces, and / or wireless link interfaces (Wi-Fi, Bluetooth, general wireless transmission, vehicular short-range communication technology, and other short-range wireless communication technologies). In some cases, the functionality of the communication interface is implemented through transceiver circuitry or dedicated transceiver chips.

[0172] The functions and actions of the modules or units in the computing devices listed above are merely illustrative examples.

[0173] The functional units in the computing device can be used to implement the aforementioned method, such as Figure 3 The hail detection method shown.

[0174] Optionally, the processor is a processor specifically designed to perform the aforementioned methods (referred to as a dedicated processor for easy distinction), or a processor that performs the aforementioned methods by invoking a computer program (referred to as a dedicated processor for easy distinction). Optionally, at least one processor may include both dedicated processors and general-purpose processors.

[0175] Optionally, if the computing device includes at least one memory, and the processor implements the aforementioned hail detection method by calling a computer program, the computer program can be stored in the memory.

[0176] This application also provides a chip including a processor, which is used to invoke (or execute) computer instructions to cause a device equipped with the chip to implement the aforementioned method, such as... Figure 3 The hail detection method shown is, for example, at least a part of the method performed by the detection device.

[0177] This application also provides a computer-readable storage medium storing computer program instructions. When these computer program instructions are executed by at least one processor, they implement the aforementioned method, such as... Figure 3 The hail detection method shown is, for example, at least a part of the method performed by the detection device.

[0178] This application also provides a computer program product, which includes computer program instructions for implementing the aforementioned method, such as... Figure 3 The hail detection method shown is, for example, at least a part of the method performed by the detection device.

[0179] This application embodiment also provides a terminal (such as...) Figure 1 The terminal 100 shown includes the aforementioned detection device 20, computing device 500, or chip. The terminal is used to implement the aforementioned method, such as... Figure 3 The hail detection method shown. Optionally, the terminal includes a vehicle.

[0180] In addition, a few additional points need to be made regarding this application: I. The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. These modifications or substitutions do not cause the essence of the corresponding technical solutions to depart from the protection scope of the technical solutions of the embodiments of this application.

[0181] 2. Unless otherwise stated, “multiple” means two or more.

[0182] 3. Unless otherwise specified or in case of logical conflict, the terms and / or descriptions in different embodiments of this application are consistent and can be referenced by each other. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0183] IV. The various numerical designations used in this application are merely for descriptive convenience and are not intended to limit the scope of protection of this application. The magnitude of the serial numbers used in this application does not imply a sequential order of execution; the execution order of each process should be determined by its function and internal logic. For example, the terms "first," "second," "third," "fourth," and other various terminology (if present) in the specification, claims, and drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. Such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein.

[0184] Furthermore, any embodiment or design described in this application as "exemplary" or "for example" should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner for ease of understanding.

[0185] V. The terms “comprising” and “having” and any variations thereof are intended to cover a non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or modules is not necessarily limited to those steps or modules that are expressly listed, but may include other steps or modules that are not expressly listed or that are inherent to such process, method, product or device.

[0186] VI. The terms “center,” “upper,” “lower,” “left,” “right,” “vertical,” “horizontal,” “inner,” and “outer,” etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0187] 7. Unless otherwise stated, " / " indicates that the objects before and after are in an "or" relationship. For example, A / B can mean A or B. In this application, "and / or" is merely a description of the relationship between the related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. A and B can be singular or plural.

[0188] 8. Unless otherwise stated, the names of devices, systems, modules and other information in the embodiments of this application are merely examples, and devices, modules and modules are used to represent possible entities that implement a certain function, and the meanings of the three can be used interchangeably.

Claims

1. A hail detection method, characterized in that, The method includes: Acquire elastic wave signals from an elastic wave sensor, which is installed in the vehicle, and the elastic wave signals propagate on the surface of the vehicle body. The elastic wave signal is compared with the signal characteristics under a hail scene to obtain a first detection result, which is used to indicate whether a hail scene has been detected.

2. The method according to claim 1, characterized in that, The first detection result is used to indicate whether a hail scene is detected based on the elastic wave signal, and the method further includes: Acquire acceleration data from an acceleration sensor located in the vehicle; The acceleration data is compared with the acceleration data features under a hail scene to obtain a second detection result. The second detection result is used to indicate whether a hail scene has been detected based on the acceleration data.

3. The method according to claim 2, characterized in that, The method further includes: If the first detection result indicates that a hail scene is detected based on the elastic wave signal, and / or the second detection result indicates that a hail scene is detected based on the acceleration data, a first instruction is generated. The first instruction is used to wake up at least one sensor, which includes one or more of a microphone, a visible light image sensor, an infrared image sensor, and a radar-type sensor. Based on the sensing data from the at least one sensor, at least one sensing detection result is obtained, each of the sensing detection results indicating whether a hail scene was detected based on the sensing data from the corresponding one or more sensors.

4. The method according to any one of claims 1-3, characterized in that, The method further includes: Based at least on the first detection result, control one or more components in the vehicle to perform hail protection operations.

5. The method according to any one of claims 4, characterized in that, The control of one or more components in the vehicle to perform hail protection operations includes one or more of the following: Send a notification message to the terminal associated with the vehicle, the notification message being used to indicate that hail has occurred at the location of the vehicle; Acquire an image of the environment surrounding the vehicle and send the image of the environment surrounding the vehicle to the terminal; Close the doors of the vehicle; Close the windows of the vehicle; Or fold up the vehicle's rearview mirrors.

6. The method according to claim 5, characterized in that, The control of one or more components in the vehicle to perform hail protection operations also includes one or more of the following: Control the vehicle to a safe parking position; Or adjust the position of the vehicle.

7. The method according to any one of claims 4-6, characterized in that, The step of controlling one or more components in the vehicle to perform hail protection operations, based at least on the first detection result, includes: Based at least on the first detection result and the hail level at the location of the vehicle, control one or more components in the vehicle to perform hail protection operations, where there are at least two hail levels for which the hail protection operations are different.

8. The method according to claim 7, characterized in that, The at least two hail levels include a first hail level and a second hail level; Hail protection operations corresponding to the first hail level include one or more of the following: Send a notification message to the terminal associated with the vehicle; Acquire an image of the environment surrounding the vehicle and send the image of the environment surrounding the vehicle to the terminal; Close the doors of the vehicle; Close the windows of the vehicle; Or fold up the rearview mirrors of the vehicle; Hail protection operations corresponding to the second hail level include the hail protection operations corresponding to the first hail level, and also include one or more of the following: Control the vehicle to a safe parking position; Or adjust the position of the vehicle.

9. The method according to any one of claims 4-6, characterized in that, The step of controlling one or more components in the vehicle to perform hail protection operations, based at least on the first detection result, includes: If the first detection result indicates that a hail scene has been detected based on the elastic wave signal, and the confidence level of the first detection result is greater than or equal to a first threshold, control one or more components in the vehicle to perform hail protection operations.

10. The method according to any one of claims 4-6, characterized in that, The step of controlling one or more components in the vehicle to perform hail protection operations, based at least on the first detection result, includes: If the first detection result indicates that a hail scene has been detected based on the elastic wave signal, and the confidence level of the first detection result is less than a first threshold, and at least one of the second detection result and at least one sensing detection result indicates that a hail scene has been detected, control one or more components in the vehicle to perform hail protection operations.

11. The method according to any one of claims 1-10, characterized in that, The step of comparing the elastic wave signal with the signal characteristics under a hail scene to obtain a first detection result includes: The elastic wave signal is compared with the signal characteristics under a hail scene to obtain a first detection result and the confidence level of the first detection result.

12. The method according to any one of claims 4-6, characterized in that, The step of controlling one or more components in the vehicle to perform hail protection operations, based at least on the first detection result, includes: If at least two of the first detection result, the second detection result, and at least one perception detection result indicate that a hail scene has been detected, control one or more components in the vehicle to perform hail protection operations.

13. The method according to any one of claims 1-10 or claim 12, characterized in that, The signal characteristics in the hail scenario include hail time-domain characteristics, which are the time-domain characteristics of the elastic wave signal propagating on the vehicle surface when hail collides with the vehicle. The step of comparing the elastic wave signal with the signal characteristics under a hail scene to obtain a first detection result includes: Based on the elastic wave signal, the time-domain characteristics of the elastic wave signal are obtained, and the time-domain characteristics include one or more of the pulse width, amplitude, and rise slope; The first detection result is obtained by comparing the time-domain characteristics of the elastic wave signal and the time-domain characteristics of the hail.

14. The method according to claim 13, characterized in that, The step of comparing the time-domain features of the elastic wave signal and the time-domain features of the hail to obtain the first detection result includes: The elastic wave signals whose time-domain characteristics do not meet the detection conditions are filtered out to obtain the filtered elastic wave signals. The detection conditions include one or more of the pulse width condition, amplitude condition, and rise slope condition. The first detection result is obtained based on the time-domain characteristics of the filtered elastic wave signal and the time-domain characteristics of the hail.

15. The method according to claim 14, characterized in that, The detection conditions include an amplitude greater than an amplitude detection threshold and / or an upward slope greater than an upward slope detection threshold; the method further includes: If the weather forecast information for the vehicle's location indicates hail, or if the rainfall at the vehicle's location exceeds a rainfall threshold, the amplitude detection threshold and / or the rise slope detection threshold shall be increased.

16. The method according to claim 14 or 15, characterized in that, The filtered elastic wave signal includes multiple pulses. The first detection result, obtained based on the time-domain characteristics of the filtered elastic wave signal and the time-domain characteristics of the hail, includes: The number of matching pulses is determined, wherein the matching pulses are the pulses whose time-domain features match the time-domain features of the hail among the plurality of pulses; If the number of matching pulses within a first duration is greater than or equal to a number threshold, the first detection result indicates that a hail scene was detected based on the elastic wave signal. Alternatively, if the number of matching pulses within a first duration is less than a number threshold, the first detection result indicates that no hail scene was detected based on the elastic wave signal.

17. The method according to claim 2 or 3, characterized in that, The acceleration data features in the hail scenario include the frequency domain features of the acceleration data when hail collides with a vehicle; The step of comparing the acceleration data with the acceleration data features under a hail scene to obtain a second detection result includes: Based on the acceleration data, the frequency domain characteristics of the acceleration data are obtained, and the frequency domain characteristics include one or more of the following: dominant frequency, high-frequency energy ratio, and decay time constant. The second detection result is obtained by comparing the frequency domain features of the acceleration data with the frequency domain features of the acceleration data when the hail collided with the vehicle.

18. The method according to any one of claims 1-17, characterized in that, The elastic wave signal includes multiple matched pulses, wherein the matched pulses are pulses whose time-domain characteristics match the time-domain characteristics of the hail; the method further includes: Determine the time-domain feature range to which the time-domain features of each of the multiple matching pulses belong, with different time-domain feature ranges corresponding to different hail levels; The hail level at the vehicle's location is determined based on the time-domain feature range to which the time-domain features of each matched pulse belong.

19. The method according to any one of claims 1-17, characterized in that, The elastic wave signal includes multiple matched pulses, wherein the matched pulses are pulses whose time-domain characteristics match the time-domain characteristics of the hail; the method further includes: Determine the time-domain feature range to which the time-domain features of each of the multiple matching pulses belong, with different time-domain feature ranges corresponding to different hail levels; The first hail level is determined based on the time domain feature range to which the time domain features of each matched pulse belong; Based on the first perception data, the size of the hail at the location of the vehicle is determined. The first perception data includes perception data from one or more of the following sensors: visible light image sensor, infrared image sensor, and radar sensor. The hail level at the vehicle's location is determined based on the first hail level and the hail size at the vehicle's location.

20. The method according to any one of claims 4-6, characterized in that, Controlling one or more components of the vehicle to perform hail protection operations includes controlling the vehicle to a safe parking position, and controlling one or more components of the vehicle to perform hail protection operations based at least on the first detection result includes: Detect whether there is a safe parking location that can be covered around the vehicle; If there is a safe parking location that can be sheltered around the vehicle, the vehicle is controlled to drive to the safe parking location, based at least on the first detection result.

21. The method according to any one of claims 4-6, characterized in that, Controlling one or more components in the vehicle to perform hail protection operations includes adjusting the vehicle's orientation. Controlling one or more components in the vehicle to perform hail protection operations, at least based on the first detection result, includes: Based on the second sensing data, the wind direction at the location of the vehicle is determined. The second sensing data includes sensing data from one or more of the following sensors: visible light image sensor, infrared image sensor, radar sensor, and inertial measurement unit. Based at least on the first detection result, the vehicle's position is adjusted so that the side of the vehicle body faces the wind direction of the vehicle's location.

22. A detection device, characterized in that, The detection device includes: A transceiver unit is used to acquire elastic wave signals from an elastic wave sensor, which is installed in the vehicle and the elastic wave signals propagate on the surface of the vehicle body. The processing unit is used to compare the elastic wave signal with the signal characteristics under the hail scene to obtain a first detection result, which is used to indicate whether a hail scene has been detected.

23. A computing device, characterized in that, The computing device includes a processor and a memory, the memory storing a program, and the processor executing the program to cause the computing device to implement the method as described in any one of claims 1-21.

24. A vehicle, characterized in that, The vehicle includes the detection device as described in claim 22 or the computing device as described in claim 23.

25. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program, the computer program including instructions for performing the method as described in any one of claims 1-21.

26. A computer program product, characterized in that, The computer program product includes instructions that, when executed by a processor, cause the method as described in any one of claims 1-21 to be implemented.

Images