An intelligent sensor group for autonomous vehicles

Abstract

The present invention relates to an intelligent sensor group for autonomous driving vehicles applied in the field of electromagnetic waves. By arranging a calibration component in the detection direction of a sensor body, during normal driving of the vehicle, the calibration component is in an unfolded state, and does not block the electromagnetic waves emitted by the sensor body. The sensor body is in a normal detection working state. When the vehicle is in an idle state, the driver can independently perform calibration operations according to needs. At this time, the calibration component is unfolded, and the electromagnetic waves of the sensor body illuminate the reflection detection area on the calibration component to obtain detection data. The data is compared with the normal detection data to preliminarily determine whether there is a possibility of deviation in the detection angle of the sensor body. In addition, by comparing with another calibration component, it is effectively determined whether the calibration component itself is loosely installed, thereby further improving the accuracy of determining the deviation of the sensor body angle.

Description

Technical Field

[0001] The present invention relates to an intelligent sensor group, and in particular to an intelligent sensor group for an autonomous driving vehicle applied in the field of electromagnetic waves. Background Art

[0002] Self-driving cars rely on artificial intelligence, visual computing, radar, monitoring devices and global positioning systems to work together, allowing computers to automatically and safely operate motor vehicles without any active human operation. There are four stages of unmanned driving according to the level of automation: driving assistance, partial automation, high automation, and full automation. The first three categories require the cooperation of drivers for assisted operation.

[0003] Self-driving cars have multiple radars, such as LiDAR and millimeter-wave radar, installed in multiple locations on the vehicle body to detect surrounding vehicles, pedestrians, and obstacles. During driving, the radar's installation angle may change due to factors such as vehicle body vibration and loose mounting structure, resulting in a deviation between the target position detected by the radar and the actual target position, causing false triggering of alarms, braking, and other functions.

[0004] To solve the above problems, the specification of Chinese patent CN111226127B discloses a method for correcting the horizontal installation angle of a radar, a radar, and a vehicle. The method obtains detection data of longitudinal obstacles detected by the radar when the vehicle is in motion; based on the detection data, the offset of the horizontal installation angle of the radar is calculated; and based on the offset, the horizontal installation angle of the radar is corrected. For example, the specification of Chinese patent CN113625234B discloses a method for correcting the installation angle of a vehicle radar and a vehicle radar. The method corrects the radar installation error angle by combining the radar's detection capability of road boundaries, and can maximize the utilization of the FOV range that can be achieved by the design.

[0005] The radar monitoring angle is an important factor affecting vehicle driving safety. Detecting whether the radar angle is offset is also a crucial part of ensuring safe vehicle driving. However, most existing radar calibration methods require setting up detection targets in external locations. This method is complex to operate and difficult to perform irregular autonomous calibration as needed. Some methods also use road boundaries or longitudinal obstacles for detection while the vehicle is driving. This method is affected by external objects and the vehicle's own position, and is prone to inaccuracies. Summary of the Invention

[0006] In view of the above-mentioned existing technologies, the technical problem to be solved by the present invention is that the various existing detection methods for vehicle radar detection angles all have certain defects, such as complex operation, low result accuracy, and difficulty in independent detection.

[0007] To solve the above problems, the present invention provides an intelligent sensor group for an autonomous vehicle, comprising a sensor body and a calibration assembly mounted inside the vehicle body, wherein the calibration assembly is located directly in front of the detection direction of the sensor body. The calibration assembly includes a mounting frame, wherein a pair of electric rotating shafts are fixedly connected to the interior of the mounting frame, and the pair of electric rotating shafts are respectively located on the left and right sides of the sensor body. The outer end of one of the electric rotating shafts is fixedly connected to one end of a calibration curtain, and the calibration curtain is wound around the outer end of the electric rotating shaft. The end of the calibration curtain away from the electric rotating shaft is fixedly connected to a pair of pull ropes, and the end of the pull rope away from the calibration curtain is fixedly connected to the outer end of the other electric rotating shaft.

[0008] It also includes a verification system, which includes an activation module, a drive module and a judgment module. The activation module is used to start the sensor body for detection when the vehicle is in an idling state. The electric shaft is electrically connected to the drive module. The judgment module is connected to an initial database. The initial database stores normal detection data of the sensor body detecting the verification curtain when no angle offset occurs. The judgment module compares the actual detection data of the sensor body on the verification curtain with the normal detection data to determine whether there is an angle offset in the sensor body.

[0009] As a further supplement to the present application, the verification system also includes a vibration monitoring module and a prompt module. The vibration monitoring module is connected to a vibration sensor installed in the vehicle body and is used to record the vibration data of the vehicle during driving. The prompt module is used to prompt the driver to perform verification operations when the vehicle is idling in place.

[0010] As a further supplement to the present application, the verification curtain includes a soft cloth, and a plurality of dispersed reflectors are fixedly connected to one end surface of the soft cloth facing the sensor body, and the plurality of reflectors form a reflection detection area on the soft cloth.

[0011] As a further supplement to the present application, the upper and lower inner walls of the mounting frame are fixedly connected to limit plates, and a pair of limit plates are provided with limit grooves at their ends close to each other, and the pull rope runs through the interior of the limit grooves.

[0012] A method for using an intelligent sensor group for an autonomous driving vehicle includes the following steps:

[0013] S1. When the vehicle is idling, the driver initiates the calibration operation on the sensor body. The activation module turns on the sensor body, enabling it to perform detection. Simultaneously, the drive module activates a pair of electric shafts, causing them to rotate a fixed number of revolutions M in a set direction, driving the calibration curtain to fully unfold directly in front of the sensor body.

[0014] S2. The sensor emits electromagnetic waves to illuminate the reflection detection area on the soft cloth and receives the echo, thereby obtaining detection data of the reflection detection area by the sensor. The detection data includes the heights of multiple reflectors and the distances from the reflectors to the sensor.

[0015] S3. Compare the detection data obtained in step S2 with the normal detection data. When the similarity between the two is not lower than the calibration value, it is determined that the detection angle of the sensor body has not been significantly offset. Conversely, when the similarity between the two is lower than the calibration value, it is determined that the detection angle of the sensor body has been significantly offset.

[0016] As another improvement of the present application, the number of verification components is a pair, and there is no contact between each other. The pair of verification components are both located directly in front of the detection direction of the sensor body, and one of the verification components is located between the sensor body and the other verification component. In the initial state, the verification curtains on the pair of verification components are respectively located on the left and right sides of the sensor body.

[0017] As another improvement and supplement to the present application, assuming that the calibration component activated in step S1 is the first calibration component, and the other calibration component is the second calibration component, in step S3, when the similarity is lower than the calibration value, no direct judgment is performed first, and the following secondary calibration operation is performed:

[0018] S3-1. First, the drive module activates the electric shaft on the first verification component, causing it to rotate a fixed number of turns M, retracting the verification curtain to its initial position. Subsequently, the pair of verification components are simultaneously activated, causing the multiple electric shafts to rotate a fixed number of turns N in a set direction, driving the pair of verification curtains to a semi-expanded position directly in front of the sensor body.

[0019] S3-2, the electromagnetic waves emitted by the sensor body simultaneously illuminate the reflection detection areas on the pair of soft cloths, and receive echoes from both, thereby obtaining detection data of the sensor body on the pair of reflection detection areas;

[0020] S3-3. Compare the detection data obtained in step S3-2 with the normal detection data and make the following determinations:

[0021] When the similarity between the detection data of the reflection detection area on a pair of calibration components and the normal detection data is lower than the calibration value, it is determined that there is a significant deviation in the detection angle of the sensor body;

[0022] When the similarity between the detection data of the reflective detection area on the first verification piece and the normal detection data is lower than the calibration value, and the similarity between the detection data of the reflective detection area on the second verification piece and the normal detection data is not lower than the calibration value, it is determined that the detection angle of the sensor body has not deviated significantly and the first verification piece is loose and deviated.

[0023] As another improved supplement of the present application, the number of turns N is smaller than the number of turns M, and the number of turns N is greater than half of the number of turns M.

[0024] To sum up, the present application sets a verification component in the detection direction of the sensor body. During normal driving of the vehicle, the verification component is in an unfolded state, and the electromagnetic waves emitted by the sensor body are not blocked. The sensor body performs normal detection work around the vehicle. When the vehicle is idling in place, the driver can independently perform verification operations according to needs. At this time, the verification component is unfolded, and the electromagnetic waves of the sensor body are irradiated on the reflection detection area on the verification component to obtain detection data. When the similarity with the normal detection data is lower than the calibration value, it is preliminarily determined that the detection angle of the sensor body may be offset. In addition, by comparing with another verification component set, it can be effectively determined whether the verification component itself is loosely installed, thereby further improving the accuracy of the determination of the sensor body angle offset. Therefore, the present application provides a verification method for the radar detection angle that is easy to operate, highly accurate, and can be independently detected on demand. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 The three-dimensional embodiment of the first and second embodiments of this application Figure 1 ;

[0026] Figure 2 The three-dimensional embodiment of the first and second embodiments of this application Figure 2 ;

[0027] Figure 3 This is a schematic diagram of the top surface structure of the sensor body in normal use in the first embodiment of the present application;

[0028] Figure 4 This is a schematic diagram of the top surface structure of the sensor body during the calibration process in the first embodiment of the present application;

[0029] Figure 5 This is a three-dimensional diagram of the sensor body during the calibration process in the first embodiment of the present application;

[0030] Figure 6 This is a schematic diagram of the top surface structure of the sensor body in normal use in the second embodiment of the present application;

[0031] Figure 7 This is a schematic side structural diagram of the sensor body in normal use in the second embodiment of the present application;

[0032] Figure 8 This is a schematic side structural diagram of the sensor body during the calibration process in the second embodiment of the present application;

[0033] Figure 9This is a schematic diagram of the top surface structure of the sensor body during the calibration process in the second embodiment of the present application;

[0034] Figure 10 This is a three-dimensional diagram of the sensor body during the calibration process in the second embodiment of the present application.

[0035] Description of the numbers in the figure:

[0036] 1 sensor body, 2 mounting frame, 3 electric shaft, 4 verification curtain, 41 soft cloth, 42 reflector, 5 pull rope, 6 limit plate, 601 limit slot. DETAILED DESCRIPTION

[0037] Two implementation modes of the present application are described in detail below with reference to the accompanying drawings.

[0038] The first implementation method:

[0039] The present invention provides an intelligent sensor group for autonomous driving vehicles. Figure 1 , including a sensor body 1 and a verification component installed inside the vehicle body, and the verification component is located in front of the detection direction of the sensor body 1. The sensor body 1 in this embodiment can adopt a millimeter wave radar, combined with Figure 2 and Figure 3 As shown, the calibration assembly includes a mounting frame 2, a pair of electric rotating shafts 3 are fixedly connected to the interior of the mounting frame 2, and the pair of electric rotating shafts 3 are respectively located on the left and right sides of the sensor body 1, the outer end of one of the electric rotating shafts 3 is fixedly connected to one end of a calibration curtain 4, and the calibration curtain 4 is wound around the outer end of the electric rotating shaft 3, and the end of the calibration curtain 4 away from the electric rotating shaft 3 is fixedly connected to a pair of pull ropes 5, and the end of the pull rope 5 away from the calibration curtain 4 is fixedly connected to the outer end of the other electric rotating shaft 3;

[0040] Combine Figure 4 and Figure 5 The verification curtain 4 includes a soft cloth 41, and a plurality of dispersed reflectors 42 are fixedly connected to one end surface of the soft cloth 41 facing the sensor body 1. The plurality of reflectors 42 form a reflection detection area on the soft cloth 41. Since metal has a strong reflection ability for electromagnetic waves, the reflector 42 can adopt a metal reflection film, which can not only be detected by the sensor body 1, but also adapt to the winding and unfolding process of the soft cloth 41.

[0041] The intelligent sensor group used in this autonomous driving vehicle also includes a matching verification system, which includes an activation module, a drive module and a judgment module. The activation module is used to start the sensor body 1 for detection when the vehicle is in an idle state. The electric shaft 3 is electrically connected to the drive module. The judgment module is connected to an initial database. The initial database stores normal detection data of the sensor body 1 detecting the verification curtain 4 when no angle offset occurs. The judgment module compares the actual detection data of the sensor body 1 on the verification curtain 4 with the normal detection data to determine whether the sensor body 1 has an angle offset.

[0042] The sensor body 1 is fixed to the vehicle body using existing installation technology, and then the mounting frame 2 is fixed to the vehicle body using fasteners (such as bolts) in front of the detection direction of the sensor body 1. The mounting frame 2 is provided with multiple mounting holes (not shown in the figure) that match the fasteners, so that the entire verification assembly is installed on the vehicle body. Under normal circumstances, the verification curtain 4 is in a fully rolled-up state, and the inner side of the mounting frame 2 is in an unobstructed state, which will not block the electromagnetic waves emitted by the sensor body 1 (such as Figure 3 and Figure 7 As shown), the electromagnetic waves generated by the sensor body 1 can completely pass through the inside of the mounting frame 2 and be emitted to the outside world, thereby detecting pedestrians, obstacles, vehicles and other objects around the vehicle and reducing the occurrence of traffic accidents;

[0043] In this embodiment, the number of verification components is single (such as Figure 3 As shown in the figure), after completing the installation of the sensor body 1 and the verification component and confirming that the installation positions and angles of the two are correct, the normal detection data is pre-collected: the sensor body 1 is started by the activation module to enable it to enter the detection work, and the pair of electric shafts 3 are started by the drive module to make the electric shafts 3 rotate a fixed number of circles M in the pre-set direction, driving the verification curtain 4 to be fully unfolded in front of the sensor body 1 (as shown in the figure). Figure 4 As shown), that is, the reflective detection area formed by the reflector 42 is fully expanded. At this time, the sensor body 1 illuminates the reflective detection area and obtains normal detection data, which includes the height and orientation of the reflector 42, the distance from the reflector 42 to the sensor body 1, etc., so as to subsequently compare with the actual detection data to determine whether the sensor body 1 has an angular deviation.

[0044] The verification system also includes a vibration monitoring module and a prompt module. The vibration monitoring module is connected to a vibration sensor installed in the vehicle body and is used to record the vibration data of the vehicle during driving. The prompt module is used to prompt the driver to perform a verification operation when the vehicle is idling. When the driver starts the car but does not shift gears or step on the accelerator pedal, the vehicle is idling. At this time, the prompt module will choose whether to issue an operation prompt based on the interval time since the last verification operation or the vibration condition of the vehicle during driving (the prompt can be displayed on the control panel inside the vehicle. After the driver touches the screen to confirm, the verification system starts , perform calibration operation), such as: 1. When the interval time from the last calibration operation exceeds T days, the driver will be prompted to perform calibration operation in the idle state after the vehicle is started next time; 2. The vibration monitoring module will transmit the obvious vibration signal of the vehicle during driving to the prompt module. Even if the interval time from the last calibration operation has not reached T days, when the vibration data received by the prompt module reaches the set maximum threshold, the driver will be prompted to perform calibration operation in the idle state next time the vehicle is started, thereby realizing timely calibration of the detection angle of the sensor body 1 and effectively improving the safety of vehicle driving.

[0045] Combine Figure 1 As shown, the upper and lower inner walls of the mounting frame 2 are fixedly connected to the limiting plates 6, and a pair of limiting plates 6 are provided with limiting grooves 601 at the ends close to each other. The pull rope 5 moves through the inside of the limiting groove 601, and the limiting plates 6 play a certain limiting role in the movement of the pull rope 5 and the verification curtain 4 during the unwinding process.

[0046] A method for using an intelligent sensor group for an autonomous driving vehicle includes the following steps:

[0047] S1. When the vehicle is idling, the driver initiates the calibration operation of the sensor body 1. The activation module turns on the sensor body 1, enabling it to perform detection. Simultaneously, the drive module activates a pair of electric shafts 3, causing them to rotate a fixed number of revolutions M in a set direction, driving the calibration curtain 4 to fully unfold in front of the sensor body 1.

[0048] Supplementary explanation: The idle state in step S1 refers to the idle state of the vehicle from the time the key is inserted to the start until the vehicle officially drives. Generally speaking, the radar on the vehicle does not work in the idle state. When the R gear or D gear is engaged, the radar starts to work (of course, there may be some types of vehicles whose radar also works in the idle state. Therefore, if the sensor body 1 is already in working state, the activation module does not need to start the sensor body 1 a second time). In addition, the idle state in the present application does not include the idle state of the vehicle waiting for the traffic light. Since the waiting time for the traffic light is short, for safety reasons, no verification operation is performed.

[0049] S2. The sensor body 1 emits electromagnetic waves to illuminate the reflection detection area on the soft cloth 41 and receives the echo, thereby obtaining detection data of the reflection detection area by the sensor body 1. The detection data includes the height of the multiple reflectors 42 and the distance from the reflectors 42 to the sensor body 1.

[0050] S3. Compare the detection data obtained in step S2 with the normal detection data. When the similarity between the two is not lower than the calibration value, it is determined that the detection angle of the sensor body 1 has not been significantly offset. Conversely, when the similarity between the two is lower than the calibration value, it is determined that the detection angle of the sensor body 1 has been significantly offset.

[0051] Supplementary explanation: If the vehicle suddenly moves during the verification operation (which can be monitored by means of vehicle speed sensors, vibration sensors, etc.), the verification system will immediately stop the verification operation and return the verification curtain 4 to its initial retracted state, which will not easily affect the normal detection use of the sensor body 1.

[0052] Second implementation method:

[0053] See also Figure 3 and Figure 6 Based on the first embodiment, this embodiment sets the number of verification components to a pair, and there is no contact between them. A pair of verification components are located directly in front of the detection direction of the sensor body 1, and one of the verification components is located between the sensor body 1 and the other verification component. In the initial state, the verification curtains 4 on the pair of verification components are respectively located on the left and right sides of the sensor body 1, and do not block the electromagnetic waves of the sensor body 1.

[0054] Assume that the verification component started in step S1 is the first verification component (i.e., the verification component close to the sensor body 1), and the other verification component is the second verification component (i.e., the verification component far away from the sensor body 1). Correspondingly, the initial database also stores normal detection data when the sensor body 1 performs synchronous detection on the verification curtains 4 on a pair of verification components without angular deviation. The acquisition method is: after completing the pre-collection operation of the first embodiment, first restore the unfolded verification curtain 4 to the initial reeled state, and then start a pair of verification components at the same time, so that the multiple electric shafts 3 rotate a fixed number of circles N in a preset direction, and at the same time unwind a pair of verification curtains 4, wherein the number of circles N is less than the number of circles M, and the number of circles N is greater than half of the number of circles M. Therefore, through the above operation, a pair of verification curtains 4 can be unwound. Figure 9 and Figure 10 In the semi-expanded state shown in FIG, the electromagnetic wave of the sensor body 1 will be irradiated onto a pair of calibration curtains 4 that are staggered on the left and right. Figure 7 and Figure 8As shown, the detection data of the local reflection detection area is obtained and stored in the initial database for data comparison in the following step S3-3.

[0055] Through the above settings, in step S3, when the similarity is lower than the calibration value, there is a possibility that the calibration component itself is loose, resulting in abnormal detection data. Therefore, in order to further improve the judgment accuracy, direct judgment is not performed first, and the following secondary calibration operation is performed:

[0056] S3-1. First, the drive module starts the electric shaft 3 on the first verification component, causing it to rotate a fixed number of turns M, and rewinding the verification curtain 4 to its initial state. Then, a pair of verification components are simultaneously started, causing multiple electric shafts 3 to rotate a fixed number of turns N in the set direction, driving the pair of verification curtains 4 to be semi-expanded and located directly in front of the sensor body 1 (as shown in the figure). Figure 9 and Figure 10 shown);

[0057] S3-2, the electromagnetic waves emitted by the sensor body 1 simultaneously illuminate the reflection detection areas on the pair of soft cloths 41, and receive echoes from both, thereby obtaining detection data of the sensor body 1 on the pair of reflection detection areas;

[0058] S3-3. Compare the detection data obtained in step S3-2 with the normal detection data and make the following determinations:

[0059] When the similarity between the detection data of the reflection detection area on the pair of verification components and the normal detection data is lower than the calibration value, that is, the detection data of the sensor body 1 on the partial reflection detection area on the second verification component is also significantly different from the normal detection data, it can be effectively determined that the detection angle of the sensor body 1 is significantly offset;

[0060] Of course, there is an extreme case mentioned above: a pair of detection components are loose and offset, while the sensor body 1 has no obvious offset, but the probability of this case is low and is not considered for the time being; or from another perspective, even if such an extreme case exists, it also indicates to a certain extent that the vehicle may have more vibration or long-term use, which may cause a group of detection components to become loose. At this time, the sensor body 1 may also have a tendency to be offset. Through the above judgment, personnel can check and repair the sensor body 1 in time, which can also effectively improve the detection accuracy of the sensor body 1 in the later stage and reduce driving accidents caused by the offset of the detection angle of the sensor body 1.

[0061] When the similarity between the detection data of the reflection detection area on the first verification piece and the normal detection data is lower than the calibration value, and the similarity between the detection data of the reflection detection area on the second verification piece and the normal detection data is not lower than the calibration value, it is determined that the detection angle of the sensor body 1 has not deviated significantly, and the first verification piece is loose and deviated. Personnel can check and repair the installation of the first verification piece in time.

[0062] In view of current actual needs, the protection scope of the above-mentioned implementation mode adopted in this application is not limited to this. Various changes made within the knowledge scope of technical personnel in this field without departing from the concept of this application still fall within the protection scope of the present invention.

Claims

1. An intelligent sensor assembly for an autonomous vehicle, characterized in that: The invention comprises a sensor body (1) and a calibration component installed inside a vehicle body, wherein the calibration component is located in front of the detection direction of the sensor body (1), wherein the calibration component comprises a mounting frame (2), wherein a pair of electric rotating shafts (3) are fixedly connected inside the mounting frame (2), and the pair of electric rotating shafts (3) are respectively located on the left and right sides of the sensor body (1), wherein the outer end of one of the electric rotating shafts (3) is fixedly connected to one end of a calibration curtain (4), and the calibration curtain (4) is wound around the outer end of the electric rotating shaft (3), and the end of the calibration curtain (4) away from the electric rotating shaft (3) is fixedly connected to a pair of pull ropes (5), and the end of the pull rope (5) away from the calibration curtain (4) is fixedly connected to the outer end of the other electric rotating shaft (3); The vehicle further comprises a calibration system, the calibration system comprising an activation module, a drive module and a judgment module, the activation module being used to start the sensor body (1) to perform detection work when the vehicle is in an idle state, the electric shaft (3) being electrically connected to the drive module, the judgment module being connected to an initial database, the initial database storing normal detection data of the sensor body (1) detecting the calibration curtain (4) when no angle deviation occurs, the judgment module comparing the actual detection data of the sensor body (1) detecting the calibration curtain (4) with the normal detection data, and judging whether the sensor body (1) has an angle deviation; The number of the verification components is a pair and they are not in contact with each other. The verification curtain (4) includes a soft cloth (41). A plurality of dispersed reflectors (42) are fixedly connected to one end surface of the soft cloth (41) facing the sensor body (1). The plurality of reflectors (42) form a reflection detection area on the soft cloth (41); The method for using the above-mentioned intelligent sensor group for autonomous driving vehicles includes the following steps: S1. When the vehicle is in an idle state, the driver starts the calibration operation of the sensor body (1) by himself. At this time, the activation module turns on the sensor body (1) to make it perform detection work. At the same time, the drive module starts a pair of electric shafts (3) to rotate them in a set direction for a fixed number of turns M, driving the calibration curtain (4) to be fully unfolded in front of the sensor body (1); S2, the electromagnetic waves emitted by the sensor body (1) illuminate the reflection detection area on the soft cloth (41) and receive the echo, thereby obtaining detection data of the reflection detection area by the sensor body (1), the detection data including the heights of the multiple reflectors (42) and the distances from the reflectors (42) to the sensor body (1); S3. Compare the detection data obtained in step S2 with the normal detection data. When the similarity between the two is not lower than the calibration value, it is determined that the detection angle of the sensor body (1) has not significantly deviated. On the contrary, when the similarity between the two is lower than the calibration value, the calibration component activated in step S1 is set as the first calibration component, and the other calibration component is set as the second calibration component, and then the following secondary calibration operation is performed: S3-1, first start the electric rotating shaft (3) on the first verification component through the driving module, so that it rotates a fixed number of turns M, and rewinds the verification curtain (4) to the initial state, and then simultaneously start a pair of verification components, so that the multiple electric rotating shafts (3) respectively rotate a fixed number of turns N in the set direction, driving the pair of verification curtains (4) to be in a semi-expanded state and located directly in front of the sensor body (1); S3-2, the electromagnetic waves emitted by the sensor body (1) simultaneously illuminate the reflection detection areas on a pair of soft cloths (41), and receive echoes from both, thereby obtaining detection data of the sensor body (1) on the pair of reflection detection areas; S3-3. Compare the detection data obtained in step S3-2 with the normal detection data and make the following determinations: When the similarity between the detection data of the reflection detection area on a pair of calibration components and the normal detection data is lower than the calibration value, it is determined that the detection angle of the sensor body (1) has a significant deviation; When the similarity between the detection data of the reflection detection area on the first verification piece and the normal detection data is lower than the calibration value, and the similarity between the detection data of the reflection detection area on the second verification piece and the normal detection data is not lower than the calibration value, it is determined that the detection angle of the sensor body (1) has not significantly deviated and the first verification piece has a loose deviation.

2. The intelligent sensor assembly for an autonomous driving vehicle according to claim 1, characterized in that: The verification system also includes a vibration monitoring module and a prompt module. The vibration monitoring module is connected to a vibration sensor installed in the vehicle body and is used to record the vibration data of the vehicle during driving. The prompt module is used to prompt the driver to perform verification operations when the vehicle is idling.

3. The intelligent sensor assembly for an autonomous driving vehicle according to claim 1, characterized in that: The upper and lower inner walls of the installation frame (2) are fixedly connected to the limiting plates (6), and a pair of the limiting plates (6) are provided with limiting grooves (601) at their ends close to each other, and the pull rope (5) is movable through the interior of the limiting grooves (601).

4. The intelligent sensor assembly for an autonomous driving vehicle according to claim 1, characterized in that: In step S3, when the similarity is lower than the calibration value, no secondary calibration operation is performed, and it is directly determined that there is a significant deviation in the detection angle of the sensor body (1).

5. The intelligent sensor group for an autonomous driving vehicle according to claim 1, characterized in that: A pair of alignment components are both located directly in front of the detection direction of the sensor body (1), and one of the alignment components is located between the sensor body (1) and the other alignment component. In an initial state, the alignment curtains (4) on the pair of alignment components are respectively located on the left and right sides of the sensor body (1).

6. The intelligent sensor assembly for an autonomous driving vehicle according to claim 1, characterized in that: The number of turns N is smaller than the number of turns M, and the number of turns N is greater than half of the number of turns M.

Images