Radar sensor alignment module for vehicle

Abstract

A sensor alignment module according to an embodiment of the present invention comprises: a processing unit for receiving radar data sensed by a radar and configuring a correction angle for the radar data by using the radar data; and a communication unit for communicating with a cloud server, wherein the processing unit adjusts the correction angle through the cloud server when it is determined that the correction angle for the radar data needs to be adjusted.

Description

Automotive radar sensor alignment module

[0001] The present invention relates to a power conversion device, and more specifically, to a sensor alignment module that aligns the angle of a radar sensor using a cloud, and a vehicle including the same.

[0002] Radar sensors mounted on vehicles transmit radio waves via a transmitting antenna and receive the reflected waves that bounce off the target, allowing them to detect factors such as the distance to the target and its speed. Radar sensors are mounted on the exterior of the vehicle, and vibrations generated when the vehicle moves can cause them to shift from their initial mounting location, potentially shifting the sensor's position. When the sensor's position shifts, the sensor's angle must be aligned to compensate for the sensor data errors. However, calculating the compensation angle to compensate for the errors can be resource-intensive.

[0003] The technical problem to be solved by the present invention is to provide a sensor alignment module that aligns the angle of a radar sensor using a cloud and a vehicle including the same.

[0004] In order to solve the above technical problem, a sensor alignment module according to one embodiment of the present invention includes a processing unit that receives radar data sensed by a radar and sets a correction angle for the radar data using the radar data; and a communication unit that performs communication with the cloud server, wherein the processing unit adjusts the correction angle through the cloud server when it is determined that adjustment of the correction angle for the radar data is necessary.

[0005] In addition, the processing unit can determine whether the radar data satisfies a correction angle adjustment condition, and if the correction angle adjustment condition is satisfied, calculate a first deformation angle of the radar from the radar data, determine whether a first correction reflection angle that reflects the currently set correction angle to the first deformation angle is within a first threshold angle, and determine whether to adjust the correction angle based on the result.

[0006] Additionally, the processing unit can transmit the radar data to the cloud server and receive a correction angle to be changed from the cloud server when the first correction reflection angle deviates from the first threshold angle.

[0007] In addition, the radar data is data according to the coordinate system of the radar, and the processing unit, when the first correction reflection angle is within the critical angle, converts the radar data into second data according to the coordinate system of the vehicle, calculates a second deformation angle from the second data, determines whether a second correction reflection angle that reflects the currently set correction angle to the second deformation angle is within the second critical angle, and determines whether to adjust the correction angle based on whether the result is the same as the result of determining whether the first correction reflection angle is within the first critical angle.

[0008] In addition, the processing unit may change the result of determining whether to adjust the correction angle using the first result when the second result of determining whether the second correction reflection angle is within the second critical angle and the first result of determining whether the first correction reflection angle is within the first critical angle are different from each other when the process is repeated a preset number of times or more.

[0009] In addition, the processing unit may adjust the correction angle through the cloud server when the first correction reflection angle is within the first critical angle and the first result and the second result are different from each other for a preset number of times or more, and may maintain the current correction angle when the first correction reflection angle is outside the first critical angle and the first result and the second result are repeated for a preset number of times or more.

[0010] Additionally, the processing unit can receive vehicle information from the vehicle and determine whether the radar data satisfies the correction angle adjustment condition using the vehicle information.

[0011] Additionally, the vehicle information may include at least one of yaw data, acceleration data, and speed data.

[0012] In addition, the communication unit may transmit the radar data, unique identification information of the radar data, and a timestamp to the cloud server, receive data, unique identification information, a timestamp, and a correction angle from the cloud server, and change the correction angle according to radar data having a timestamp later than a timestamp corresponding to the radar data of the earliest point in time among the radar data transmitted to the cloud server to a new correction angle.

[0013] In addition, it includes a storage unit that stores the radar data and initial sensor position information of the radar, and the processing unit can adjust the correction angle of the radar data using the radar data stored in the storage unit.

[0014] In order to solve the above technical problem, a vehicle according to an embodiment of the present invention includes a sensor unit including a radar; a storage unit for storing data of the sensor unit; and a processing module for correcting the data of the sensor unit, wherein the processing module includes a processing unit for receiving radar data sensed by the radar and setting a correction angle for the radar data using the radar data; and a communication unit for performing communication with the cloud server, wherein the processing unit adjusts the correction angle through the cloud server when it is determined that adjustment of the correction angle for the radar data is necessary.

[0015] According to embodiments of the present invention, sensor misalignment can be corrected without using vehicle resources by calculating the sensor correction angle in a cloud environment rather than on the vehicle. Furthermore, accurate correction angles can be calculated using a large amount of data from a cloud server.

[0016] FIG. 1 is a block diagram of a sensor alignment module according to one embodiment of the present invention.

[0017] FIGS. 2 to 8 are drawings for explaining a sensor alignment module according to an embodiment of the present invention.

[0018] FIGS. 9 to 14 are drawings for explaining a sensor alignment module according to another embodiment of the present invention.

[0019] Figure 15 is a block diagram of a vehicle according to one embodiment of the present invention.

[0020] Hereinafter, a preferred embodiment of the present invention will be described in detail with reference to the attached drawings.

[0021] However, the technical idea of ​​the present invention is not limited to some of the embodiments described, but can be implemented in various different forms, and within the scope of the technical idea of ​​the present invention, one or more of the components between the embodiments can be selectively combined or substituted for use.

[0022] In addition, terms (including technical and scientific terms) used in the embodiments of the present invention may be interpreted as having a meaning that can be generally understood by a person of ordinary skill in the technical field to which the present invention belongs, unless explicitly and specifically defined and described, and terms that are commonly used, such as terms defined in a dictionary, may be interpreted in consideration of the contextual meaning of the relevant technology.

[0023] Additionally, the terms used in the embodiments of the present invention are intended to describe the embodiments and are not intended to limit the present invention.

[0024] In this specification, the singular may also include the plural unless specifically stated otherwise in the phrase, and when it is described as “A and / or at least one (or more) of B, C”, it may include one or more of all combinations that can be combined with A, B, C.

[0025] Additionally, in describing components of embodiments of the present invention, terms such as first, second, A, B, (a), (b), etc. may be used. These terms are only intended to distinguish the components from other components, and are not intended to limit the nature, order, or sequence of the components.

[0026] And, when a component is described as being 'connected', 'coupled', or 'connected' to another component, it may include not only cases where the component is 'connected', 'coupled', or 'connected' directly to the other component, but also cases where the component is 'connected', 'coupled', or 'connected' by another component between the component and the other component.

[0027] Additionally, when described as being formed or arranged "above" or "below" each component, "above" or "below" includes not only cases where the two components are in direct contact with each other, but also cases where one or more other components are formed or arranged between the two components. Furthermore, when expressed as "above" or "below," the meaning may include not only the upward direction but also the downward direction based on one component.

[0028] A variation according to the present embodiment may include some components of each embodiment and some components of other embodiments. That is, a variation may include one embodiment among various embodiments, but may omit some components and include some components of the corresponding other embodiment. Or, the opposite may be true. The features, structures, effects, etc. to be described in the embodiments are included in at least one embodiment, and are not necessarily limited to just one embodiment. Furthermore, the features, structures, effects, etc. exemplified in each embodiment can be combined or modified in other embodiments by a person having ordinary skill in the art to which the embodiments belong. Therefore, the contents related to such combinations and modifications should be interpreted as being included within the scope of the embodiments.

[0029] FIG. 1 is a block diagram of a sensor alignment module according to one embodiment of the present invention. FIGS. 2 to 8 are drawings for explaining a sensor alignment module according to an embodiment of the present invention.

[0030] The present invention comprises a processing unit (110) and a communication unit (120). The sensor alignment module (100) according to the present invention can be mounted on a vehicle (600). The vehicle (600) can include a sensor unit (200) including a radar (210), and the sensor alignment module (100) can correct and align the misaligned angles of the sensor units (200). The sensor unit (200) can include a radar (210), a LiDAR (LiDAR, 220), a position sensor (GPS, 240), etc., and can include a camera (230), navigation information (HP Map, 250), etc. The vehicle (600) can be an autonomous vehicle or an autonomous driving robot, and can generate an autonomous driving path using the data of the sensor unit (200) and control the vehicle accordingly. The vehicle (600) can be an SDV (Software Defined Vehicle), which is a vehicle that controls and manages hardware using software.

[0031] The radar (210) can be mounted on a vehicle (600), and may be misaligned from its installed position due to vibration caused by the movement of the vehicle (600). To ensure accurate measurement of the radar (210), a correction angle must be applied to correct the misaligned angle. The misaligned angle may continuously change depending on the vibration, and the correction angle must be adjusted accordingly.

[0032] To this end, the processing unit (110) receives radar data sensed by the radar (210) and uses the radar data to set a correction angle for the radar data. If the processing unit (110) determines that correction angle adjustment is necessary, it uses the cloud server (400) to adjust the correction angle. The process of adjusting the correction angle will be described in detail below.

[0033] The communication unit (120) can communicate with the cloud server (400). The communication unit (120) can communicate with the cloud server (400) via wireless communication. The communication unit (120) can transmit a correction angle request to the cloud server (400) and receive the correction angle calculated by the cloud server (400).

[0034] The processing unit (110) can send a correction angle request to the cloud server (400) when a correction angle adjustment is required, and can use radar data to determine whether a correction angle adjustment is required. The radar data generated by the radar (210) is stored in the storage unit (300), and the processing unit (110) can determine whether a correction angle adjustment is required using the radar data stored in the storage unit (300). The storage unit (300) can receive and store data from sensors included in the sensor unit (200). Since the measurement cycles of each sensor may be different, the storage unit (300) can store data as an integrated storage that integrates the data of the sensors. The storage unit (300) can also store a timestamp, which is the time at which each piece of data was collected.

[0035] The processing unit (110) determines whether the radar data satisfies the correction angle adjustment condition, and when the correction angle adjustment condition is satisfied, calculates the first deformation angle of the radar (210) from the radar data, determines whether the first correction reflection angle that reflects the currently set correction angle to the first deformation angle is within the first critical angle, and determines whether to adjust the correction angle based on the result.

[0036] The processing unit (110) can first determine whether the radar data satisfies the correction angle adjustment conditions. If the radar data includes information that can determine the degree to which the radar (210) is misaligned, the radar data can be used to determine whether or not the correction angle needs to be adjusted. Therefore, first, it can be determined whether the radar data satisfies the correction angle adjustment conditions.

[0037] Radar data is data collected from the radar (210), and may include data obtained by post-processing signals reflected from an object from the radar (210), and installation information when the sensor is initially installed. The results of post-processing the reflected signals from the radar data are expressed as points on a coordinate system, and these points may be referred to as a point cloud or a detection list. Each point may include range, azimuth, elevation, power, radial velocity, and initial sensor position information. Here, the range is the distance to the detected point based on the radar (210), the azimuth is the angle of the detected point on the horizontal plane based on the front of the radar, the altitude is the angle of the detected point on the vertical plane based on the front of the radar, the power is the reflected signal intensity of the detected point, the radial velocity is the relative velocity of the detected point, and the initial sensor position information is the mounting angle for correcting the error that occurs when mounting each sensor at the time of shipment, and the angle can be used as a relative standard for calculating the coordinates of point clouds thereafter.

[0038] In the process of converting the coordinate plane of the point cloud of the radar (210) into the coordinate plane of the vehicle, the existing mounting position of the radar (210) is used as a reference. However, the position of the installed radar (210) may be different from the existing mounting position due to an accident or a large impact. In this way, if the mounting position of the radar (210) is shifted from the existing position, the coordinate plane of the point cloud may be converted to an incorrect coordinate position because the mounting position is shifted during the process of converting the coordinate plane.

[0039] To eliminate these errors, auto-alignment or angle correction can be performed, and the current radar misalignment angle can be calculated using a detection list generated under specific conditions, such as correction angle adjustment conditions. For example, radar data measured on a road environment with a straight line of highly reflective objects, such as a straight guardrail, can be utilized.

[0040] This process can be called a condition check process. The processing unit (110) can determine that radar data containing information indicating that a preset number of objects with high reflectivity, such as guardrails, are listed satisfies the condition.

[0041] At this time, the processing unit (110) can perform a condition verification process using vehicle information. At this time, the processing unit (110) can use yaw rate, acceleration, and vehicle speed as vehicle information. The yaw rate is information indicating the degree to which the front of the vehicle (600) turns.

[0042] The processing unit (110) can determine that the compensation angle adjustment conditions are satisfied when the yaw rate, acceleration, and speed are under preset conditions. The processing unit (110) can determine that the compensation angle adjustment conditions are satisfied when the conditions of the yaw rate being less than ±4 deg / s, the acceleration being less than ±0.5 m / s, and the speed being faster than 30 kmph are satisfied. That is, the processing unit (110) can determine whether to adjust the compensation angle using radar data under the corresponding conditions when the condition of moving straight at a constant speed in a road environment where objects with strong reflectivity are lined up in a straight line is satisfied.

[0043] The processing unit (110) can check (111) the compensation angle adjustment conditions and, if the conditions are satisfied, calculate the first deformation angle of the radar (210) from the radar data. Using the results obtained by measuring the radar data and the results obtained by vehicle information, the first deformation angle, which is the deformation angle at which the radar sensor is misaligned, can be calculated (114) using the difference between the measured angle and the actual angle.

[0044] The processing unit (110) can calculate the first correction reflection angle by reflecting the currently set correction angle to the calculated first deformation angle, determine whether the first correction reflection angle is within the first critical angle, and determine whether to adjust the correction angle based on the result. The first correction reflection angle can be calculated by reflecting the initially or previously set correction angle to the first deformation angle, and the first critical angle can be used to determine (115) whether the first correction reflection angle needs to be adjusted. The first critical angle can be an allowable error range, and can be set for the safety of the vehicle or can be set by the user. Here, the first correction reflection angle can be calculated as follows.

[0045] Angle Offset=Raw Angle Offset + Correction Angle

[0046] Here, Angle Offset is the first correction reflection angle, Raw Angle Offset is the first deformation angle, and Correction Angle is the current correction angle. The first correction reflection angle can indicate the degree of misalignment compared to the initial radar sensor position, and if the first correction reflection angle is within the first critical angle, it can be determined that the radar (210) is located in the on-position, which is a position that is not misaligned, and if the first correction reflection angle is outside the first critical angle, it can be determined that the radar (210) is located in the off-position, which is a misaligned position. If the current state is determined to be the off-position, correction angle adjustment can be performed through the cloud server (400).

[0047] If the first correction reflection angle deviates from the first critical angle, the processing unit (110) determines that correction angle adjustment is necessary, transmits radar data to the cloud server (400), and receives a correction angle to be changed from the cloud server (400). If the first correction reflection angle deviates from the first critical angle, a correction angle request can be transmitted to the cloud server (400) together with radar data through communication with the cloud server (400).

[0048] Radar data from autonomous vehicles contains a wealth of information, and calculating a correction angle from this data requires significant resources. Furthermore, the more data is available, the more accurate the correction angle can be calculated. The sensor alignment module (100) according to an embodiment of the present invention can calculate the correction angle using a cloud server (400) with extensive resources, rather than within the module or vehicle (600).

[0049] The cloud server (400) may be a module including software used in a cloud environment outside the vehicle (600). The cloud server (400) may include various algorithms for calculating the correction angle of the radar (210) and may include a neural network trained to calculate the correction angle from radar data through machine learning or the like. In addition, the cloud server (400) may utilize various methods for calculating the correction angle from radar data.

[0050] When communicating with a cloud server (400), the communication unit (120) can transmit the radar data, unique identification information of the radar data, and a timestamp to the cloud server (400), and receive the data, unique identification information, timestamp, and correction angle from the cloud server (400).

[0051] A timestamp is stored together with radar data generated by the radar (210) and stored in the storage unit (300), and when the radar data satisfies the correction angle adjustment conditions, unique identification information may be assigned and stored. The unique identification information (Unique ID) is information that can distinguish data that satisfies the condition verification, and may be assigned using a sequentially assigned frame number or a hash value of the incoming data. Here, the hashing function is a function that maps data of an arbitrary length to data of a fixed length, and the hash may mean a value obtained by the hashing function. When transmitting radar data to the cloud server (400), the communication unit (120) may transmit the timestamp and unique identification information of the corresponding radar data together.

[0052] The communication unit (120) can store sensor data including radar data in a data queue (121) and transmit a plurality of stored sensor data to a cloud server (400). At this time, the number of sensor data stored in the data queue (121) may be limited. Therefore, in order to confirm which sensor data is transmitted when transmitting to the cloud server (400), a start timestamp may be set and stored. If the current start timestamp is null, the timestamp of the first sensor data to be stored can be set as the start timestamp and stored (122). When the sensor data stored in the data queue (121) is transmitted to the cloud server (400), the validity can be determined by comparing it with the timestamp of data received from the cloud server (400) thereafter using the currently set start timestamp.

[0053] The cloud server (400) can calculate the correction angle of the radar (210) using sensor data including radar data received through the communication unit (120). The cloud server (400) can include various algorithms for calculating the correction angle of the radar (210) and can include a neural network trained to calculate the correction angle from radar data through machine learning or the like. In addition, the cloud server (400) can utilize various methods for calculating the correction angle from radar data.

[0054] The cloud server (400) can transmit sensor data, unique identification information and timestamp of the sensor data, and a correction angle calculated from the sensor data to the communication unit (120).

[0055] The communication unit (120) can change the correction angle according to the radar data having a timestamp later than the start timestamp corresponding to the radar data of the earliest point in time among the radar data transmitted to the cloud server (400) to a new correction angle. The communication unit (120) compares (123) the timestamp of the received sensor data with the currently set start timestamp. If the timestamp of the currently received sensor data is a timestamp before the currently set start timestamp, it is not data for adjusting the correction angle, and in this case, the procedure can be terminated without adjusting the correction angle. If the timestamp of the currently received sensor data is a timestamp before the currently set start timestamp, it is data for adjusting the correction angle, and the processing unit (110) can adjust the correction angle by changing the correction angle using the correction angle calculated from the corresponding sensor data.

[0056] The processing unit (110) can convert radar data into second data according to the vehicle's coordinate system when the first correction reflection angle is within the first critical angle. The processing unit (110) can determine that the radar (210) is located in the on-position, which is a position where the radar is not twisted. The radar data is data according to the coordinate system of the radar (210), and when it is determined that the radar data corresponds to the on-position in the radar's coordinate system, the processing unit (110) can go through a process of once again checking whether correction angle adjustment is unnecessary in the vehicle's coordinate system. To this end, a coordinate transformation (112) can be performed to convert the radar coordinate system into the vehicle's coordinate system.

[0057] Thereafter, the processing unit (110) converts the radar data into second data according to the vehicle's coordinate system when the first correction reflection angle is within the critical angle, calculates a second deformation angle from the second data, determines whether the second correction reflection angle reflecting the currently set correction angle to the second deformation angle is within the second critical angle, and determines whether to adjust the correction angle based on whether the result is the same as the result of determining whether the first correction reflection angle is within the first critical angle.

[0058] A second deformation angle can be calculated in the vehicle coordinate system, a second correction reflection angle reflecting the currently set correction angle can be calculated, and whether the second correction reflection angle is within the second critical angle can be determined, and based on the result, it can be confirmed whether it is the same as the previously determined current state.

[0059] Angle Offset_2=Raw Angle Offset_2 + Correction Angle_2

[0060] Here, Angle Offset_2 is the second correction reflection angle, Raw Angle Offset_2 is the second deformation angle, and Correction Angle_2 is the current correction angle. The second correction reflection angle can indicate the degree of distortion in the vehicle's coordinate system, and if the second correction reflection angle is within the second critical angle, it is determined that it is located in the on-position, which is a position that is not distorted in the vehicle's coordinate system, and the Correct Angle flag can be set. If the second correction reflection angle is outside the second critical angle, the Correct Angle flag can be cleared.

[0061] The processing unit (110) can change the result of determining whether to adjust the correction angle using the first result when the second result of determining whether the second correction reflection angle is within the second critical angle and the first result of determining whether the first correction reflection angle is within the first critical angle are different for a preset number of repetitions or more. By comparing the first result of determining whether the first correction reflection angle is within the first critical angle and the second result of determining whether the second correction reflection angle is within the second critical angle, it is possible to confirm whether the first result was an accurate judgment.

[0062] At this time, it is possible to determine whether the first result, which determines whether the first correction reflection angle is within the first critical angle, is a result derived from normal radar data or a result derived from radar data that is incorrectly measured or contains an error. To this end, the first result and the second result are compared multiple times, and if the first result and the second result are different from each other for a preset number of repetitions, the result of determining whether to adjust the correction angle using the first result can be changed. That is, even if the first result is determined to be an on-position, if the first result and the second result are different from each other for a preset number of repetitions, the current state can be changed to an off-position, and the correction angle can be adjusted via the cloud server (400) depending on the off-position. Conversely, even if the first result is determined to be an off-position, if the first result and the second result are different from each other for a preset number of repetitions, the current state can be changed to an on-position, and the correction angle can be not adjusted via the cloud server (400) depending on the off-position.

[0063] That is, state determination (113) can be performed using coordinate transformed data in coordinate transformation (112).

[0064] As shown in Fig. 7, the first result and the second result can be compared to see if they are the same, and if they are different, the Different State Counter value can be increased, and if the first result and the second result are the same, the Different State Counter value can be initialized. After the Different State Counter value is increased because the first result and the second result are different, if the number of different state counts exceeds the number of state changes, the current state can be changed depending on whether or not to adjust the compensation angle.

[0065] The processing unit (110) can adjust the correction angle via the cloud server (400) if the first correction reflection angle is within the first critical angle and if the first result and the second result are different from each other more than a preset number of times. That is, if the first result is on-positioned or changed to off-position through repeated comparisons of the first result and the second result, the correction angle can be adjusted via the cloud server (400).

[0066] The processing unit (110) can maintain the current correction angle if the first correction reflection angle exceeds the first threshold angle and the first result and the second result are repeated a preset number of times or more. In other words, if the first result is in an off-position or is changed to an on-position through repeated comparisons of the first result and the second result, the correction angle may not be adjusted.

[0067] State Determination (113) is a process that prevents malfunctions due to values ​​jumping or fluctuating using Hit or Miss, and various State Decision methods can be used. For example, if the sequence is Hit, Miss, Hit, Hit, then changes can be made if there are two consecutive Hits. Since radar data collected in an actual environment may be false positives, if a specific number of consecutive hits is exceeded, whether or not a fault has occurred can be determined.

[0068] In addition, the Temporary Current State and the Current State can be distinguished by comparing the unique identification information (Unique ID) of the data used for coordinate transformation (112) and state determination (113) with the current unique identification information (Recent Unique ID). The unique identification information (Unique ID) comparison can be performed to confirm that all data stored in the cloud server (400) while the mounting state is off-position have been reflected up to the latest data. If the mounting state is on-position, the current state can also be updated immediately because the unique identification information of the corresponding data is the current unique identification information.

[0069] For example, when the current state is on-position, the current unique identifier is 5, the number of state changes (State Change Threshold) is 3, the number of mismatch counts is 0, and the Correct Angle and unique identifier (Unique ID) values ​​are consecutively (True, 1), (False, 2), (False, 3), (False, 4), (False, 5), when (False, 5) comes in, the mismatch count exceeds the number of state changes, so the temporary current state is changed to off-position, the number of mismatch counts is initialized, and since the current unique identifier and the unique identifier of the last processed data are the same as 5, the current state can also be changed to off-position.

[0070] As another example, when the current state is off-position, the current unique identifier is 6, the number of state changes (State Change Threshold) is 2, the number of mismatch counts is 0, and the Correct Angle and unique identifier (Unique ID) values ​​are consecutively (True, 1), (True, 2), (True, 3), (False, 4), (False, 5), (False, 6), when (True, 3) is entered, the mismatch count exceeds the number of state changes, so the temporary current state is changed to on-position, the number of mismatch counts is initialized, and the current unique identifier and the unique identifier of the last processed data are different as 6,3, so the current state can be maintained as off-position. After (False, 6) is entered, the inverse and mismatch counts exceed the number of state changes, so the temporary current state is changed to an off-position, the Different State Counter is initialized to 0, and the current unique identification information and the unique identification information of the last processed data are the same as 6, so the current state can be maintained as an off-position according to the temporary current state, and the correction angle can be adjusted through the cloud server (400) according to the off-position.

[0071] FIGS. 9 to 14 are drawings for explaining a sensor alignment module according to another embodiment of the present invention. Detailed descriptions of each component of FIGS. 9 to 14 correspond to the detailed descriptions of each component of FIGS. 1 to 8, and thus, redundant descriptions will be omitted below.

[0072] The sensor alignment module according to an embodiment of the present invention may be an auto-alignment module (Auto-Alignment Module, 510). Vehicle information and radar data received from the sensor unit (200) including the vehicle (260) and the radar (210) may be stored in the storage (Intermediate Storage, 310). Here, the sensor collects data, the storage (310) is an integrated data storage of sensors collected at different intervals, the auto-alignment module (510) is included in a software module used in the vehicle, and the cloud (410) may be a software module used in a cloud environment rather than in the vehicle. Since the data collection cycle may be different for each sensor, the storage (310) may store the latest information collected for each sensor and store the collection time together.

[0073] Vehicle information (Vehicle Info) is information provided by default in the vehicle (600) and may include yaw rate, acceleration, and speed collected from each sensor in the vehicle.

[0074] Radar sensor information (Radar Info) is information collected from the radar (210), and may include information obtained by post-processing signals reflected from an object from the radar (210), and installation information when the sensor is first installed. The radar sensor information includes a Detection List in which the results of post-processing the reflected signals are expressed as points on a coordinate system, and each point may include Range, Azimuth, Elevation, Power, Radial Velocity, and Initial Sensor Position.

[0075] As the trend changes to Software Defined Vehicle (SDV), Software Defined Components (SDC) are also increasing, which creates a constraint that a single component / module cannot use the limited vehicle memory / storage space alone, and there is a need to increase efficiency. The auto-alignment module (510) according to an embodiment of the present invention applies data and algorithms required for auto-alignment according to the SDV concept to the cloud (cloud, 410), and as computing resources that increase using the cloud (410) grow to an incomparable extent compared to before, a complex but more accurate auto-alignment algorithm can be used, and data that can be used for the complex algorithm (e.g., LiDAR, HDMap, etc. for Radar Auto Alignment) can be sent to the cloud (410) and used together.

[0076] At this time, data can be acquired (Condition Check, 511) only under specific conditions (e.g., a road with a straight guardrail or objects with strong reflectivity lined up in a straight line), and while using the cloud (410) as a data storage device, data can be avoided by not wasting data, and a method of using Unique ID & Timestamp to distinguish data (State Decision (513) & Cloud Communication (520)) can be applied.

[0077] The sensors update each collected data to the storage (310), and the automatic alignment module (510) performs a condition check (511) on the radar information and vehicle information among the data updated to the storage (310), thereby determining whether the environment satisfies the conditions. The environment satisfies the conditions includes an environment in which a straight road has reflective objects such as guardrails on the left and right, and if the condition is met, Valid Data = True can be returned, and if the condition is not met, Valid Data = False can be returned. If the environment satisfies the conditions, the Raw Angle Offset is calculated (514), a unique identification information (Unique ID) is assigned to the data, the unique identification number of the recently collected data is stored (Recent Unique ID, 517), and depending on the current mounting state of the radar, Angle Adjustment (512) or Cloud Communication (520) is performed.

[0078] 'Raw Angle Offset' is the misalignment angle of the radar before adding the correction angle, and the unique identification information (Unique ID) can be a value that can distinguish data that has passed the condition check (511). For example, it can be a sequentially assigned Frame Number or a hash value of the incoming data. Here, the hashing function is a function that maps data of an arbitrary length to data of a fixed length, and the hash can mean a value obtained by the hashing function. The mounting state (Current State) is called On-Position when the misalignment degree (Angle Offset = Raw Angle Offset + Correction Angle) compared to the Initial Sensor Position, which is the standard when the radar is shipped, is within the specified misalignment allowable threshold (Offset Limit), and it can be called Off-Position when it is outside the specified misalignment allowable threshold.

[0079] In the case of on-position, the coordinates of objects detected by the radar are calculated as coordinates based on the vehicle through the process of converting the radar coordinate system into the vehicle coordinate system in Angle Adjustment (512), and using these values, the presence or absence of misalignment (Correct Angle Flag) based on the corresponding data can be returned. The presence or absence of misalignment (Correct Angle Flag) indicates whether the sum of the misalignment degree calculated from the corresponding data (Raw Angle Offset) and the currently used correction angle (Correction Angle) is within the misalignment allowable threshold (Offset Limit) and can be True or False. The presence or absence of misalignment (Correct Angle Flag) is a value derived based on one currently processed data, and the mounting state is a value derived by accumulating the misalignment values ​​of consecutive data in State Decision (513), and can mean different values. The mounting state decision (State Decision, 513) can be performed using the returned Current Angle Flag.

[0080] In the case of an off-position, Auto-alignment (angle correction, 411) is executed in the cloud (410) to receive a correction value (Correction Angle) through cloud communication (520), and after receiving the returned value from the cloud (410) through cloud communication, it is checked again through Angle Adjustment (512) and State Decision (153) whether the returned correction angle is properly improved for misalignment, and the mounting state (Current State) can be updated through the final result when all data accumulated during the off-position are processed up to the most recent data.

[0081] Condition Check (511) can set a valid data flag using yaw rate, acceleration, and speed, as shown in Fig. 11, and Angle Adjustment (512) can set a correct Angle flag by determining whether Angle Offset = Raw Angle Offset + Correction Angle is within the Offset Limit, as shown in Fig. 12. State Decision (513) can determine whether there is a misalignment if the number of consecutive specified times exceeds a certain number of times, considering that radar data collected in an actual environment may be false detections, as shown in Fig. 13. In addition, the Unique ID and Recent Unique ID of the data used in Angle Adjustment (512) + State Decision (513) can be compared to distinguish between the Temporary Current State and the Current State. In order to confirm that the data stored in the cloud while the mounting status is off-position have been reflected up to the latest data, Unique ID comparison is performed, and if the mounting status is on-position, the Unique ID of the data used for Angel Adjust + State Decision becomes the Recent Unique ID, so the Current State can also be updated immediately. As shown in Fig. 14, in the cloud communication (520), if the mounting status is off-position, among the collected data, data that passed the Condition Check (511) is transferred, and if the Start Timestamp is not stored when the first data is received, the Timestamp of the received data can be stored in the Start Timestamp (522).Here, Start Timestamp refers to storing the collection time (timestamp) of the first data that came in when the mounting status changed from on-position to off-position. Start Timestamp can be initialized to a Null value when the mounting status changes from off-position to on-position in State Decision (513).

[0082] Data received through cloud communication (520) is stored in the data queue (521). Due to the limitations of the in-vehicle memory, data cannot be accumulated indefinitely, so only the most recent N data sets set in advance can be accumulated. When connected to the cloud (410), data that has been in the data queue (521) for the longest time can be transmitted to the cloud (410).

[0083] The data transmitted and received from the cloud (410) may be as follows. Unique ID is required to distinguish data, and is required to check whether the data is the latest in State Decision (513) when returned thereafter. Sensor data includes data required to calculate Correction Angle in the cloud (410), and Raw Angle Offset may be included because it is required for Angle Adjustment. In addition, other data such as Vehicle Info, Radar Info, and additional LiDAR Info, Camera Info, etc. may be changed according to the data required by the Auto-alignment (angle correction) algorithm. The timestamp is a timestamp for the time when the current data was collected, and is required to filter only the data after the Start Timestamp when returned thereafter. A timestamp may be added to prevent the order of data from being pushed or changed due to a communication error or an error on the cloud (410) side. The correction angle is a correction angle value calculated and returned from the cloud (410), and may only be included in the return information. The correction angle calculated and received from the cloud (410) is used to confirm the mounting status through Angle Adjustment (512) and State Decision (513), and a Unique ID and Raw Angle Offset may be required for use at this time.

[0084] As described above, by calculating the sensor correction angle in a cloud environment, rather than on the vehicle, the sensor's misalignment can be corrected without using vehicle resources. Furthermore, accurate correction angles can be calculated by utilizing the large amount of data available on the cloud server.

[0085] Fig. 15 is a block diagram of a vehicle according to one embodiment of the present invention. Detailed descriptions of each component of Fig. 15 correspond to detailed descriptions of each component of Figs. 1 to 8 or each component of Figs. 9 to 14, and thus, any redundant descriptions will be omitted.

[0086] A vehicle (600) according to an embodiment of the present invention is composed of a sensor unit (200), a storage unit (300), and a processing module (100), and the processing module (100) may include a processing unit (110) and a communication unit (120).

[0087] The sensor unit (200) includes a radar (210), the storage unit (300) stores data of the sensor unit (200), and the processing module (100) corrects data of the sensor unit (200).

[0088] The processing module (100) includes a processing unit (110) that receives radar data sensed by the radar (210), sets a correction angle for the radar data using the radar data, and a communication unit (120) that performs communication with a cloud server (400). When the processing unit (110) determines that adjustment of the correction angle for the radar data is necessary, the processing unit (110) adjusts the correction angle through the cloud server (400).

[0089] The processing unit (110) determines whether the radar data satisfies the correction angle adjustment condition, and when the correction angle adjustment condition is satisfied, calculates the first deformation angle of the radar from the radar data, determines whether the first correction reflection angle that reflects the currently set correction angle to the first deformation angle is within the first threshold angle, and determines whether to adjust the correction angle based on the result.

[0090] When the first correction reflection angle deviates from the first threshold angle, the processing unit (110) can transmit the radar data to the cloud server (400) and receive the correction angle to be changed from the cloud server (400).

[0091] The above radar data is data according to the coordinate system of the radar, and the processing unit (110) converts the radar data into second data according to the coordinate system of the vehicle when the first correction reflection angle is within the critical angle, calculates a second deformation angle from the second data, determines whether the second correction reflection angle that reflects the currently set correction angle to the second deformation angle is within the second critical angle, and determines whether to adjust the correction angle based on whether the result is the same as the result of determining whether the first correction reflection angle is within the first critical angle.

[0092] The processing unit (110) can change the result of determining whether to adjust the correction angle using the first result when the second result of determining whether the second correction reflection angle is within the second critical angle and the first result of determining whether the first correction reflection angle is within the first critical angle are different from each other for a preset number of repetitions.

[0093] The processing unit (110) adjusts the correction angle through the cloud server (400) when the first correction reflection angle is within the first critical angle and the first result and the second result are different from each other more than the preset number of times, and maintains the current correction angle when the first correction reflection angle is outside the first critical angle and the first result and the second result are repeated more than the preset number of times.

[0094] The processing unit (110) receives vehicle information and can use the vehicle information to determine whether the radar data satisfies the correction angle adjustment condition, and the vehicle information can include at least one of yaw data, acceleration data, and speed data.

[0095] The communication unit (120) transmits the radar data, unique identification information of the radar data, and a timestamp to the cloud server (400), receives data, unique identification information, a timestamp, and a correction angle from the cloud server (400), and changes the correction angle according to radar data having a timestamp later than the timestamp corresponding to the radar data at the earliest point in time among the radar data transmitted to the cloud server to a new correction angle.

[0096] Meanwhile, embodiments of the present invention can be implemented as computer-readable code on a computer-readable recording medium. Computer-readable recording media include all types of recording devices that store data that can be read by a computer system.

[0097] Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disk, and optical data storage devices. In addition, the computer-readable recording media can be distributed across network-connected computer systems, so that computer-readable code can be stored and executed in a distributed manner. In addition, functional programs, codes, and code segments for implementing the present invention can be easily inferred by programmers in the technical field to which the present invention pertains.

[0098] As described above, the present invention has been described with specific details such as specific components and limited examples and drawings, but these are provided only to help a more general understanding of the present invention, and the present invention is not limited to the above examples, and those with ordinary knowledge in the field to which the present invention pertains can make various modifications and variations from this description.

[0099] Therefore, the idea of ​​the present invention should not be limited to the described embodiments, and all things that are equivalent or equivalent to the following claims as well as the claims are considered to fall within the scope of the idea of ​​the present invention.

Claims

1. A processing unit that receives radar data sensed by the radar and sets a correction angle for the radar data using the radar data; and Includes a communication unit that performs communication with a cloud server, The above processing unit, A sensor alignment module that adjusts the correction angle via the cloud server when it is determined that the correction angle adjustment for the above radar data is necessary.

2. In paragraph 1, The processing unit is, A sensor alignment module that determines whether the above radar data satisfies a compensation angle adjustment condition, and if the compensation angle adjustment condition is satisfied, calculates a first deformation angle of the radar from the radar data, determines whether a first correction reflection angle that reflects the currently set compensation angle to the first deformation angle is within a first threshold angle, and determines whether to adjust the compensation angle based on the result.

3. In paragraph 2, The above processing unit, A sensor alignment module that transmits the radar data to the cloud server and receives a correction angle to be changed from the cloud server when the first correction reflection angle exceeds the first threshold angle.

4. In paragraph 2, The above radar data is data according to the coordinate system of the above radar, The above processing unit, A sensor alignment module that converts the radar data into second data according to the vehicle's coordinate system when the first correction reflection angle is within the critical angle, calculates a second deformation angle from the second data, determines whether the second correction reflection angle reflecting the currently set correction angle to the second deformation angle is within the second critical angle, and determines whether to adjust the correction angle based on whether the result is the same as the result of determining whether the first correction reflection angle is within the first critical angle.

5. In paragraph 4, The above processing unit, A sensor alignment module that changes the result of determining whether to adjust the compensation angle using the first result when the second result of determining whether the second compensation reflection angle is within the second critical angle and the first result of determining whether the first compensation reflection angle is within the first critical angle are different from each other when repeated a preset number of times or more.

6. In paragraph 5, The above processing unit, If the first correction reflection angle is within the first critical angle and the first result and the second result are different from each other more than the preset number of times, the correction angle is adjusted through the cloud server. A sensor alignment module that maintains the current correction angle when the first correction reflection angle deviates from the first threshold angle and the first result and the second result are repeated more than the preset number of times.

7. In paragraph 2, The above processing unit, A sensor alignment module that receives vehicle information from a vehicle and determines whether the radar data satisfies the correction angle adjustment condition using the vehicle information.

8. In paragraph 7, The above vehicle information is, A sensor alignment module comprising at least one of yaw data, acceleration data, and velocity data.

9. In paragraph 1, The above communication department, Transmitting the above radar data, unique identification information of the above radar data and a timestamp to the cloud server, and receiving the data, unique identification information, timestamp, and correction angle from the cloud server, A sensor alignment module that changes the correction angle according to radar data having a timestamp later than the timestamp corresponding to the earliest point in time among the radar data transmitted to the cloud server to a new correction angle.

10. In paragraph 1, A storage unit for storing the radar data and the initial sensor position information of the radar is included. The above processing unit, A sensor alignment module that adjusts the correction angle of the radar data using the radar data stored in the storage unit.

Images