Method and system for measuring speed of target by vehicle based on multiple radars

Abstract

The invention discloses a multi-radar-based target speed measurement method and system for a vehicle, and relates to the technical field of target speed measurement methods, and the method comprises the steps: determining a corresponding radial speed based on the combination of two sets of radar data; the X-direction speed and the Y-direction speed of the front target are determined according to the included angle corresponding to the two radial speeds, it is guaranteed that the same front target is subjected to speed measurement by the two radars at the same time, and the accuracy of the X-direction speed and the Y-direction speed of the front target is guaranteed; furthermore, if the measurement error of the X-direction speed or the measurement error of the Y-direction speed is larger than a preset measurement error threshold value, the measurement error is gradually reduced based on further increase of the number of the radars, so that the optimized measurement error is output until the optimized measurement error is lower than the preset measurement error threshold value. The increase of the radar and the optimization of the measurement error are realized, and the accuracy of the X-direction speed and the Y-direction speed of the front target is further ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of target speed measurement methods, and particularly to a vehicle speed measurement method and system for a target based on multiple radars. Background Art

[0002] With the development of technology, the autonomous driving technology of vehicles has gradually matured. Vehicles drive on the road and measure the speed of the target ahead. In the prior art, the radar is a millimeter-wave radar, which is located in front of the vehicle and there is only one. The radar in front of the vehicle measures the speed of the target ahead in a single dimension. Currently, the on-vehicle millimeter-wave radar measures the horizontal angle and pitch angle of the target, but it is impossible to calculate the X-direction speed and Y-direction speed of the target ahead. Summary of the Invention

[0003] The purpose of the present invention is to overcome the deficiencies of the prior art. The present invention provides a vehicle speed measurement method and system for a target based on multiple radars.

[0004] An embodiment of the present invention provides a vehicle speed measurement method for a target based on multiple radars, including: both radars collect the position coordinates and angles of the target ahead of the vehicle to form two sets of radar data combinations;

[0005] Determine the corresponding radial velocity based on the two sets of radar data combinations;

[0006] Determine the X-direction speed and Y-direction speed of the target ahead according to the angle corresponding to the two radial velocities;

[0007] If the measurement error of the X-direction speed or the measurement error of the Y-direction speed is greater than the preset measurement error threshold, then gradually reduce the measurement error based on the further increase in the number of radars until the optimized measurement error is lower than the preset measurement error threshold;

[0008] If the target ahead is not detected by multiple radars at the same time, then multiple radars perform separate speed measurement relative to the target ahead, and determine the X-direction speed and Y-direction speed of the target ahead based on the angle corresponding to the linear velocity detected by multiple radars for the target ahead.

[0009] An embodiment of the present invention provides a vehicle speed measurement system for a target based on multiple radars. The vehicle speed measurement system for a target based on multiple radars is applied to the above-mentioned vehicle speed measurement method for a target based on multiple radars. The vehicle speed measurement system for a target based on multiple radars includes:

[0010] A radar data combination module, configured to collect the position coordinates and angles of the target ahead of the vehicle by both radars to form two sets of radar data combinations;

[0011] A radial velocity module for determining a corresponding radial velocity based on a combination of two sets of radar data;

[0012] A first velocity module for determining the X-direction velocity and Y-direction velocity of a front target based on the angle between two corresponding radial velocities;

[0013] An error module for, if the measurement error of the X-direction velocity or the measurement error of the Y-direction velocity is greater than a preset measurement error threshold, gradually reducing the measurement error based on a further increase in the number of radars to output an optimized measurement error until the optimized measurement error is lower than the preset measurement error threshold;

[0014] A second velocity module for, if the front target is not detected by multiple radars simultaneously, performing independent speed measurement on the front target by multiple radars, and determining the X-direction velocity and Y-direction velocity of the front target based on the angle between the linear velocities detected by multiple radars for the front target.

[0015] Compared with the prior art, the beneficial effects of the present invention are:

[0016] In an embodiment of the present invention, by the method in the embodiment of the present invention, two radars both collect the position coordinates and angles of a front target of a vehicle to form two sets of radar data combinations; determine the corresponding radial velocity based on the two sets of radar data combinations; determine the X-direction velocity and Y-direction velocity of the front target according to the angle between the two corresponding radial velocities, ensuring that the same front target is speed-measured by two radars simultaneously and ensuring the accuracy of the X-direction velocity and Y-direction velocity of the front target;

[0017] Furthermore, if the measurement error of the X-direction velocity or the measurement error of the Y-direction velocity is greater than a preset measurement error threshold, gradually reduce the measurement error based on a further increase in the number of radars to output an optimized measurement error until the optimized measurement error is lower than the preset measurement error threshold, realizing the optimization of the measurement error with the increase of radars and further ensuring the accuracy of the X-direction velocity and Y-direction velocity of the front target.

[0018] Therefore, if the front target is not detected by multiple radars simultaneously, multiple radars perform independent speed measurement on the front target, and determine the X-direction velocity and Y-direction velocity of the front target based on the angle between the linear velocities detected by multiple radars for the front target, accommodating the consideration that the front target is not detected by multiple radars simultaneously and realizing the precise control of the X-direction velocity and Y-direction velocity of the front target. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 is a schematic flowchart of a method for a vehicle in an embodiment of the present invention to measure the speed of a target based on multiple radars;

[0020] Figure 2 It is a schematic flowchart of step S11 in the method for measuring the speed of a target by a vehicle based on multiple radars in an embodiment of the present invention;

[0021] Figure 3 It is a schematic flowchart of step S12 in the method for measuring the speed of a target by a vehicle based on multiple radars in an embodiment of the present invention;

[0022] Figure 4 It is a schematic flowchart of step S13 in the method for measuring the speed of a target by a vehicle based on multiple radars in an embodiment of the present invention;

[0023] Figure 5 It is a schematic flowchart of step S14 in the method for measuring the speed of a target by a vehicle based on multiple radars in an embodiment of the present invention;

[0024] Figure 6 It is a schematic flowchart of step S15 in the method for measuring the speed of a target by a vehicle based on multiple radars in an embodiment of the present invention;

[0025] Figure 7 It is a schematic diagram of the structural composition of the system for measuring the speed of a target by a vehicle based on multiple radars in an embodiment of the present invention. Detailed implementation manners

[0026] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention.

[0027] Please refer to Figures 1 to 7 , a method for measuring the speed of a target by a vehicle based on multiple radars, which is applied to the scenario of measuring the speed of a target by a vehicle based on multiple radars with an in-built induction cooker; the method for measuring the speed of a target by a vehicle based on multiple radars includes:

[0028] Step S11: Both radars collect the position coordinates and angles of the front target of the vehicle to form two sets of radar data combinations;

[0029] Step S12: Determine the corresponding radial velocities based on the two sets of radar data combinations;

[0030] Step S13: Determine the X-direction speed and Y-direction speed of the front target according to the angles corresponding to the two radial velocities;

[0031] Step S14: If the measurement error of the X-direction speed or the measurement error of the Y-direction speed is greater than the preset measurement error threshold, then gradually reduce the measurement error based on the further increase in the number of radars until the optimized measurement error is lower than the preset measurement error threshold;

[0032] Step S15: If the front target is not detected by multiple radars simultaneously, the multiple radars perform independent speed measurement relative to the front target, and determine the speed of the front target in the X direction and the speed in the Y direction based on the angles corresponding to the linear speeds detected by the multiple radars for the front target.

[0033] Reference Figure 2 , in step S11, both radars collect the position coordinates and angles of the front target of the vehicle to form two sets of radar data combinations.

[0034] In the specific implementation process of the present invention, the specific steps are as follows:

[0035] S111: Two radars are arranged in front of the vehicle and detect the front target of the vehicle along the forward direction. At this time, there is an overlapping area in the detection ranges of the two radars.

[0036] S112: When the front target is in the overlapping area, the two radars detect the same front target and collect the position coordinates and angles of the front target of the vehicle. At this time, the two radars are the first radar and the second radar respectively.

[0037] S113: In the detection data of the first radar, the position coordinates and angles of the front target of the vehicle are used as the first set of radar data combination; in the detection data of the second radar, the position coordinates and angles of the front target of the vehicle are used as the second set of radar data combination.

[0038] In the embodiment of the present application, two radars are arranged in front of the vehicle and detect the front target of the vehicle along the forward direction. At this time, there is an overlapping area in the detection ranges of the two radars, so that the two radars can detect the same front target.

[0039] At this time, the two radars are carefully placed at the front of the vehicle, and their main task is to detect the front target in the forward direction of the vehicle; the detection ranges of these two radars are not completely independent, but there is a part of the overlapping area; this overlapping area is the key in the design because it allows the two radars to detect the same target simultaneously, thereby providing redundant data, which helps subsequent data fusion and error correction.

[0040] The overlapping area of the detection range should be large enough to cover the main traffic area in front of the vehicle; the size of the overlapping area depends on factors such as the detection range of the radar, the installation position and angle, etc.; the radar should have a certain environmental adaptability and be able to maintain stable detection performance under different weather conditions (such as rain, fog, snow, etc.) and road conditions (such as curves, slopes, etc.).

[0041] Specifically, assume there is an autonomous vehicle equipped with two millimeter-wave radars, which are respectively installed on the left and right sides of the front of the vehicle. The detection ranges of both radars are areas within a 60-degree angle and a distance of 100 meters in front. Since they are installed on the left and right sides of the vehicle, their detection ranges will overlap in an area within an approximately 30-degree angle and a distance of 50 meters in front of the vehicle.

[0042] When the vehicle is driving on the highway, an obstacle suddenly appears ahead (such as a dropped cargo box). Since there is an overlapping area in the detection ranges of the two radars, both of them can detect this obstacle. The first radar (installed on the left) detects the position of the obstacle as 45 meters away from the vehicle and at an azimuth angle of 25 degrees. The second radar (installed on the right) detects the position as 47 meters away from the vehicle and at an azimuth angle of 32 degrees. These two data points will be used in subsequent speed calculation and error correction steps. From this example, it can be seen that the careful setting of the radars and the overlap of the detection ranges provide a reliable data basis for subsequent speed measurement methods. These data will be used to calculate the radial speed, speed components of the target, and perform error correction, so as to ensure that the autonomous vehicle can accurately and safely identify and respond to obstacles ahead.

[0043] Furthermore, when the front target is in the overlapping area, the two radars detect the same front target and collect the position coordinates and included angles of the front target of the vehicle. At this time, the two radars are the first radar and the second radar respectively, and the position coordinates and included angles of the front target are introduced.

[0044] At this time, when the front target is in the detection overlapping area of the two radars, the two radars will simultaneously detect this target. They will respectively collect the position coordinates of the target (polar coordinates relative to the radar, including distance and angle) and the included angle with the vehicle's traveling direction. To distinguish the data of the two radars, they are respectively called the first radar data and the second radar data.

[0045] The radar detects the target by emitting and receiving electromagnetic waves. When the target is within the detection range of the radar, the radar will receive the reflected electromagnetic waves, thus identifying the existence of the target. The radar processes the received signal and extracts the position coordinates of the target, which usually include the distance and azimuth angle of the target relative to the radar.

[0046] The included angle refers to the angle between the line connecting the target and the radar and the vehicle's traveling direction; this included angle is calculated based on the azimuth angle of the target measured by the radar and the vehicle's own heading angle; the information of the included angle is crucial for subsequent speed decomposition and error correction steps, as it provides the spatial position information of the target relative to the vehicle's moving direction. Since the two radars detect the target simultaneously, it is necessary to ensure that the data they collect is synchronized; this is achieved by introducing a time synchronization mechanism in the radar system; data synchronization is crucial for subsequent data fusion and error correction steps, as it ensures that the two sets of data are collected at the same time point and thus comparable.

[0047] Specifically, assume that an autonomous vehicle is driving on a highway and there is a pedestrian in front; this pedestrian is exactly within the detection overlapping area of the two radars in front of the vehicle; the first radar (installed on the left side of the vehicle) detects the position of this pedestrian as 30 meters away from the vehicle and at an azimuth angle of 20 degrees; at the same time, based on the vehicle's heading angle (assumed to be 0 degrees, i.e., directly in front of the vehicle), the first radar calculates that the included angle between this pedestrian and the vehicle's traveling direction is 20 degrees.

[0048] The second radar (installed on the right side of the vehicle) also detects the position of this pedestrian, and the data obtained is 32 meters away from the vehicle and at an azimuth angle of 25 degrees; similarly, based on the vehicle's heading angle, the second radar calculates that the included angle between this pedestrian and the vehicle's traveling direction is 25 degrees; these two sets of data (the first radar data and the second radar data) will be used for subsequent speed calculation and error correction steps; by comparing and analyzing these two sets of data, the autonomous vehicle can more accurately determine the speed and movement trajectory of the pedestrian, and thus make a safer driving decision; through this example, it can be seen that when the target is in the radar detection overlapping area, the two radars can simultaneously detect the target and collect data; these data provide a reliable basis for subsequent speed calculation and error correction.

[0049] Therefore, in the detection data of the first radar, the position coordinates and included angle of the target in front of the vehicle are used as the first set of radar data combination; in the detection data of the second radar, the position coordinates and included angle of the target in front of the vehicle are used as the second set of radar data combination.

[0050] At this time, the data detected by the first radar and the second radar are combined respectively to form two sets of radar data combinations; for the first radar, the target position coordinates (distance and azimuth angle) and the included angle with the vehicle's traveling direction in its detection data are combined into the first set of radar data combination; similarly, the detection data of the second radar is also combined into the second set of radar data combination.

[0051] Extract the position coordinates (distance and azimuth angle) of the target and the angle with the vehicle's traveling direction from the detection data of the first radar and the second radar; ensure that the extracted data is accurate and complete, without omission or error; combine the target position coordinates and the angle of the first radar to form the first set of radar data combinations; this combination usually includes three parameters: the distance of the target relative to the radar, the azimuth angle, and the angle with the vehicle's traveling direction; similarly, combine the target position coordinates and the angle of the second radar to form the second set of radar data combinations.

[0052] To distinguish between the two sets of data, assign different identifiers or labels to them, such as "the first set of radar data" and "the second set of radar data"; these identifiers or labels will play an important role in subsequent data processing and analysis; store the two sets of radar data combinations for use in subsequent speed calculation and error correction steps; consider using an appropriate data structure (such as an array, a list, or a database) to organize and manage this data when storing.

[0053] Specifically, assume that an autonomous vehicle is driving on an urban road and there is a cyclist in front; this person is exactly within the detection overlap area of the two radars in front of the vehicle; the first radar (installed on the left side of the vehicle) detects the position of this cyclist as 25 meters away from the vehicle and at an azimuth angle of 15 degrees; at the same time, based on the vehicle's heading angle (assuming the vehicle is driving straight and the heading angle is 0 degrees), the first radar calculates that the angle between this cyclist and the vehicle's traveling direction is 15 degrees; therefore, the first set of radar data combinations is (25 meters, 15 degrees, 15 degrees).

[0054] The second radar (installed on the right side of the vehicle) also detects the position of this cyclist, and the data obtained is 27 meters away from the vehicle and at an azimuth angle of 20 degrees; similarly, based on the vehicle's heading angle, the second radar calculates that the angle between this cyclist and the vehicle's traveling direction is 20 degrees; therefore, the second set of radar data combinations is (27 meters, 20 degrees, 20 degrees); these two sets of data combinations will be used in subsequent speed decomposition, error correction, and final speed calculation steps; by comparing and analyzing these two sets of data combinations, the autonomous vehicle can more accurately determine the speed and movement trajectory of the cyclist, and thus make safer driving decisions.

[0055] Reference Figure 3 , in step S12, determine the corresponding radial velocity based on the two sets of radar data combinations;

[0056] In the specific implementation process of the present invention, the specific steps are as follows:

[0057] S121: Determine the position coordinates of the vehicle's front target corresponding to the combination of each group of radar data, and perform directional detection on the position coordinates of the vehicle's front target to collect the current speed of the vehicle's front target.

[0058] S122: Determine the corresponding radial velocity according to the operation of the current speed of the vehicle's front target and the corresponding included angle. At this time, there are two radial velocities, corresponding to two radars.

[0059] In the embodiment of the present application, the position coordinates of the vehicle's front target are determined based on the screening of each group of radar data combinations, and directional detection is performed on the position coordinates of the vehicle's front target to collect the current speed of the vehicle's front target, introducing the current speed of the front target.

[0060] At this time, screen each group of radar data combinations obtained in the previous steps (such as S113) to determine the most accurate or reliable position coordinates of the vehicle's front target; the screening process involves multiple factors, such as data consistency, signal strength, radar reliability, etc. Check whether there are obvious inconsistencies between each group of radar data, such as whether the differences in distance and azimuth are too large.

[0061] After screening out the optimal radar data combination, use this data to determine the position coordinates of the vehicle's front target; this position coordinate usually includes the distance and azimuth of the target (relative to the radar); based on the propagation time and speed of the radar signal (such as the speed of light), calculate the distance of the target relative to the radar; according to the direction of the radar antenna and the direction of the received signal, calculate the azimuth of the target relative to the radar.

[0062] Perform directional detection on the position coordinates of the vehicle's front target. Directional detection means using a radar or other sensors to continuously track the target to obtain its dynamic information; in this step, directional detection will be performed on the determined position coordinates to collect the current speed of the target; use the continuous scanning or Doppler effect of the radar to track the target; calculate the current speed of the target by processing the change of the radar echo signal (such as frequency shift).

[0063] Specifically, assume that there is an autonomous vehicle driving on a highway, and there is a moving truck in front as the target; there are two radars (Radar A and Radar B) installed on the left and right sides of the vehicle respectively, and they can both detect this truck.

[0064] Both Radar A and Radar B detected the truck, but Radar A had a higher signal strength and its data was more consistent with the known map information. Therefore, it was decided to use the data from Radar A to determine the position coordinates of the truck. Radar A measured the truck to be 30 meters away from the vehicle at an azimuth angle of 10 degrees (relative to the front of the vehicle). Therefore, the position coordinates of the truck were determined to be (30 meters, 10 degrees).

[0065] Radar A began to continuously track the truck and collect its speed information. By processing the changes in the radar echo signals, Radar A measured the current speed of the truck to be 80 km / h (relative to Radar A). In this example, by screening the optimal radar data combination, the position coordinates of the truck were determined, and its current speed was collected through directional detection. This information will provide the basis for subsequent speed decomposition, error correction, and final driving decisions.

[0066] Furthermore, the corresponding radial speed is determined according to the operation of the current speed of the front target of the vehicle and the corresponding included angle. At this time, there are two radial speeds, corresponding to two radars, which takes into account the overall operation of the current speed of the front target of the vehicle and the corresponding included angle, ensuring the accuracy of the corresponding radial speed.

[0067] At this time, two key pieces of information need to be clarified: one is the current speed of the front target of the vehicle, which is collected through the previous steps (such as S121); the other is the included angle between the target and the vehicle's traveling direction, which is usually calculated in the previous steps (such as S112). These two pieces of information are crucial for subsequent calculation of the radial speed. The current speed refers to the linear motion speed of the target relative to the radar (or the vehicle). In an autonomous driving scenario, this is usually measured by sensors such as radar or lidar (LiDAR). The included angle refers to the angle between the target and the vehicle's traveling direction. This angle is calculated from the azimuth angle measured by the radar and the vehicle's heading angle.

[0068] After obtaining the information of the current speed and the included angle, the calculation of the radial speed begins. The radial speed refers to the component of the relative speed between the target and the vehicle in the direction of the line connecting the target and the vehicle. It is calculated by multiplying the current speed by the cosine value of the included angle. Application of the cosine function: The cosine function is used to decompose the current speed into the radial component. In a two-dimensional plane, this is calculated by the formula vradial = vcurrent·cos(θ), where vradial is the radial speed, vcurrent is the current speed, and θ is the included angle. Note the directionality: When calculating the radial speed, the direction of the speed needs to be noted. If the target and the vehicle are approaching each other, the radial speed is positive; if they are moving away from each other, the radial speed is negative.

[0069] Since there are two radars (such as Radar A and Radar B), two radial velocities will be calculated separately; these two velocities should be consistent to a certain extent (although there are minor differences), and this consistency serves as a verification of the data accuracy and reliability; compare the radial velocities calculated by the two radars to evaluate the data accuracy and consistency; if there is a significant difference between the two velocities, error correction is required; this is achieved by considering factors such as the installation position of the radar, detection angle, environmental noise, etc.

[0070] Specifically, assume there is an autonomous vehicle driving on a highway, and there is a moving truck ahead as the target; two radars (Radar A and Radar B) are installed on the left and right sides of the vehicle respectively, and both can detect this truck.

[0071] Through the measurement of Radar A, it is known that the current speed of the truck is 80 km / h (relative to Radar A); by calculating the azimuth angle of Radar A and the heading angle of the vehicle, the angle between the truck and the vehicle's traveling direction is obtained as 10 degrees.

[0072] Using the cosine function and the formula vradial = vcurrent·cos(θ), the radial velocity corresponding to Radar A is calculated as 80·cos(10°) ≈ 79.4 km / h; similarly, if there is also data from Radar B (assuming the measured angle is slightly different, such as 12 degrees), the radial velocity corresponding to Radar B is also calculated. Compare the radial velocities calculated by Radar A and Radar B, and it is found that there is a minor difference between them (which is caused by factors such as the installation position and detection angle of the radar); considering that this difference is within the acceptable range, it is decided not to perform further error correction on the data; finally, the radial velocities calculated by the two radars are obtained respectively, and this velocity information will provide a basis for subsequent driving decisions; in this example, by understanding the information of the current speed and angle, the radial velocities corresponding to the two radars are calculated, and these velocities are compared and processed; this information is crucial for the safe driving of the autonomous vehicle.

[0073] In an embodiment of the present application, assume there is an autonomous vehicle, and there is a moving truck ahead as the target; two radars (Radar A and Radar B) are installed on the vehicle, and they detect the information of the truck respectively; the current speed detected by Radar A is 80 km / h, and the angle is 10 degrees; the current speed detected by Radar B is 78 km / h, and the angle is 12 degrees; the radial velocity of Radar A: v radial,A = 80·cos(10°) ≈ 79.4 km / h; the radial velocity of Radar B: v radial,B = 78·cos(12°) ≈ 77.1 km / h.

[0074] The radial velocity matching table is shown in Table 1 as follows:

[0075] Table 1 Radial Velocity Matching Table

[0076] Radar ID Current speed (km / h) Angle (°) Radial speed (km / h) A 80 10 79.4 B 78 12 77.1

[0077] Reference Figure 4 , in step S13, determine the X-direction velocity and Y-direction velocity of the target ahead according to the included angle corresponding to the two radial velocities;

[0078] In the specific implementation process of the present invention, the specific steps are as follows:

[0079] S131: Collect two radial velocities and the corresponding included angle, and construct a system of binary linear equations according to the two radial velocities and the corresponding included angle;

[0080] S132: Determine the X-direction velocity and Y-direction velocity of the target ahead according to the solution of the system of binary linear equations;

[0081] S133: Collect the current speed of the vehicle, and determine the speed difference between the vehicle and the target ahead according to the comparison of the current speed of the vehicle, the X-direction velocity and Y-direction velocity of the target ahead; Determine the position change amount between the vehicle and the target ahead according to the speed difference and time, and determine the real-time distance between the vehicle and the target ahead according to the position change amount between the vehicle and the target ahead and the relative distance between the vehicle and the target ahead. Based on the regulation of the real-time distance between the vehicle and the target ahead, dynamically trigger the optimization of the dynamic detection mode of multiple radars for the same target ahead, which takes into account the position change amount between the vehicle and the target ahead and the relative distance between the vehicle and the target ahead as a whole, and ensures the accuracy of the real-time distance between the vehicle and the target ahead.

[0082] In the system of binary linear equations, the system of binary linear equations is:

[0083]

[0084] At this time, v 1x , v 1y are respectively the X-direction velocity and Y-direction velocity of the target ahead; the included angle between the line connecting the second radar and the target ahead and the X-axis is α2; the included angle between the line connecting the first radar and the target ahead and the X-axis is α1; the radial velocity of the target ahead measured by the second radar is v 12L ; the radial velocity of the target ahead measured by the second radar is v 11L .

[0085] In the embodiment of the present application, collect two radial velocities and the corresponding included angle, and construct a system of binary linear equations according to the two radial velocities and the corresponding included angle;

[0086] At this time, data needs to be collected from two radars (assumed to be Radar A and Radar B); each radar will provide a radial velocity value (i.e., the component of the relative velocity between the target and the radar in the direction of the radar line of sight) and an included angle value (i.e., the angle between the direction of the line connecting the target and the radar and the direction of vehicle travel).

[0087] The radar calculates the radial velocity of the target by measuring the time difference or frequency difference between the transmitted signal and the received signal; the included angle is usually jointly determined by the azimuth angle measurement of the radar and the heading angle measurement of the vehicle; the azimuth angle is the angle between the direction pointed by the radar antenna and the due north direction, and the heading angle is the angle between the direction of vehicle travel and the due north direction; by taking the difference between these two angles, the included angle between the target and the direction of vehicle travel is obtained.

[0088] After collecting the radial velocities and included angles of the two radars, a system of linear equations with two variables is constructed using this information; this system of equations is used to solve for the velocity components of the target in front in the X and Y directions in the vehicle coordinate system; Principle of constructing the system of equations: According to the principle of velocity decomposition, the radial velocity of each radar is expressed as the projection of the velocity component of the target in front in the direction of the radar line of sight; therefore, for each radar, an equation about the velocity component of the target in front is obtained; Form of the system of equations: Assume the radial velocity of Radar A is v radial,A , the included angle is ; the radial velocity of Radar B is vradial,B, and the included angle is θ B ; then the system of equations is expressed as:

[0089] v radial,A =v x ·cos(θ A )+v y ·sin(θ A )

[0090] v radial,B =v x ·cos(θ B )+v y ·sin(θ B )

[0091] where v x and v y are the velocities of the target in front in the X and Y directions respectively.

[0092] Specifically, assume there is an autonomous vehicle driving on a highway, and there is a moving truck in front as the target; two radars (Radar A and Radar B) are installed on the vehicle, and they respectively detect the information of the truck; the radial velocity detected by Radar A is 79.4 km / h, and the included angle is 10°; the radial velocity detected by Radar B is 77.1 km / h, and the included angle is 12°.

[0093] Substitute the collected data into the system of equations to obtain:

[0094] 79.4 = v x ·cos(10°) + v y ·sin(10°)

[0095] 77.1 = v x ·cos(12°) + v y ·sin(12°)

[0096] Now, with this system of linear equations in two variables, next use algebraic methods (such as the substitution method, the elimination method, or the matrix solution method) to solve for v x and v y ; these velocity components will be used in subsequent autonomous driving decision-making and planning processes.

[0097] Furthermore, based on the solution of the system of linear equations in two variables, the X-direction velocity and Y-direction velocity of the target ahead are determined. By introducing the X-direction velocity and Y-direction velocity of the target ahead, it is ensured that the same target ahead is measured for velocity by two radars simultaneously, ensuring the accuracy of the X-direction velocity and Y-direction velocity of the target ahead.

[0098] At this time, in step S131, a system of linear equations in two variables has been constructed based on the radial velocities and angles detected by two radars; this system of equations contains two equations, and each equation involves the velocity components (v x and v y ) of the target ahead in the X-direction and Y-direction.

[0099] The system of linear equations in two variables can be solved by various methods, including the substitution method, the elimination method, and the matrix solution method, etc.; in practical applications, an efficient and accurate method is usually selected to solve it;

[0100] Substitution method: If one of the equations can easily solve for the expression of one variable, then substitute this expression into the other equation to solve for the other variable;

[0101] Elimination method: By performing addition and subtraction operations on the equations, eliminate one of the variables, thus obtaining an equation that only contains one variable, then solve this equation, and substitute the solution into the original equation to solve for the other variable;

[0102] Matrix solution method: Represent the system of equations in matrix form, and then use methods of linear algebra (such as Gaussian elimination or LU decomposition) to solve.

[0103] Meanwhile, according to the selected solution method, specific solution steps are executed; this usually involves algebraic operations and equation transformations; after obtaining the solution, its correctness should be verified; this is done by substituting the solution into the original system of equations to check if all equations are satisfied; once the correctness of the solution is verified, determine the velocity components of the target ahead in the X and Y directions; these velocity components will be used in subsequent autonomous driving decision-making and planning processes.

[0104] Therefore, collect the current speed of the vehicle, and determine the speed difference between the vehicle and the target ahead based on the comparison of the current speed of the vehicle, the X-direction speed and Y-direction speed of the target ahead; determine the position change amount between the vehicle and the target ahead according to the speed difference and time, and determine the real-time distance between the vehicle and the target ahead according to the position change amount between the vehicle and the target ahead and the relative distance between the vehicle and the target ahead. Based on the regulation of the real-time distance between the vehicle and the target ahead, dynamically trigger the optimization of the dynamic detection mode of multiple radars for the same target ahead, which takes into account the overall consideration of the position change amount between the vehicle and the target ahead and the relative distance between the vehicle and the target ahead, and ensures the accuracy of the real-time distance between the vehicle and the target ahead.

[0105] At this time, collect the current driving speed from the vehicle's sensors (such as wheel speed sensors, GPS speed sensors, etc.); this speed is the speed of the vehicle relative to the ground and is usually used for path planning, speed control, and obstacle avoidance decision-making in the autonomous driving system. In step S132, the velocity components (vx,target and vy,target) of the target ahead in the X and Y directions have been determined; in this step, compare these speeds with the current speed of the vehicle (v vehicle , usually assumed to be mainly in the X direction) to determine the speed difference in the X direction (Δvx = vvehicle - v x,target ); the speed difference in the Y direction is usually small and has a limited impact on the relative position change between the vehicle and the target ahead, so it is ignored in this step.

[0106] Once the speed difference is determined, calculate the relative position change amount (Δx = Δv x ·Δt) between the vehicle and the target ahead in the X direction during this time according to the time interval (Δt); this time interval is a fixed sampling period and also a dynamic time interval triggered based on specific events.

[0107] In the autonomous driving system, the vehicle usually also obtains the relative distance (drelative) to the target ahead through sensors such as radars, lidars (LiDARs), or cameras; this relative distance is the measured value at the initial moment; to obtain the real-time distance, this initial relative distance needs to be combined with the position change amount calculated in step three (dreal-time = d relative-Δx (assuming the change in position in the Y direction is negligible).

[0108] After obtaining the real-time distance, the autonomous driving system dynamically adjusts the detection mode of the radar based on this distance. For example, when the real-time distance is relatively close, increase the sampling frequency, resolution, or scanning range of the radar to improve the detection accuracy and reliability of the target ahead. Conversely, when the real-time distance is relatively far, appropriately reduce these parameters to save energy and computing resources. During the whole process, both the change in position and the relative distance between the vehicle and the target ahead are considered. This comprehensive consideration ensures the accuracy of the real-time distance and provides more accurate environmental perception information for the autonomous driving system.

[0109] Specifically, assume the following scenario: the current speed of the vehicle: v vehicle = 100 km / h (about 27.8 m / s); the speed of the target ahead in the X direction: v x,target = 90 km / h (about 25.0 m / s); time interval: Δt = 1 second; initial relative distance: d relative = 50 meters; determine the speed difference: Δvx = vvehicle - v x,target = 27.8 - 25.0 = 2.8 m / s;

[0110] Calculate the change in position: Δx = Δv x ·Δt = 2.8×1 = 2.8 meters;

[0111] Determine the real-time distance: dreal-time = drelative - Δx = 50 - 2.8 = 47.2 meters;

[0112] Assume that the autonomous driving system sets a threshold. When the real-time distance is less than 45 meters, increase the sampling frequency and resolution of the radar. In this example, the real-time distance is 47.2 meters, which has not reached the threshold, so the detection mode of the radar remains unchanged. However, if the real-time distance continues to decrease, the system will automatically trigger the optimization of the radar detection mode. Through this example, we can see how step S133 works in the autonomous driving system to ensure that the vehicle can accurately perceive and respond to the changes of the target ahead.

[0113] In an embodiment of the present application, assume there is a real-time distance matching table for dynamically adjusting the radar detection mode according to the real-time distance. The real-time distance matching table is shown in Table 2:

[0114] Table 2 Real-time Distance Matching Table

[0115] Real-time distance (meters) Radar detection mode <10 High frequency and high resolution 10-20 Medium frequency and medium resolution 20-50 Low frequency and low resolution >50 Sleep / Low power consumption

[0116] Now, assume that the calculated real-time distance is 30 meters. According to the real-time distance matching table, adjust the radar detection mode to "medium frequency and medium resolution".

[0117] In addition, in the system of binary linear equations, the system of binary linear equations is as follows:

[0118]

[0119] At this time, v 1x and v 1y are the X-direction speed and Y-direction speed of the front target respectively; the angle between the line connecting the second radar and the front target and the X-axis is α2; the angle between the line connecting the first radar and the front target and the X-axis is α1; the radial speed of the front target measured by the second radar is v 12L ; the radial speed of the front target measured by the second radar is v 11L .

[0120] Reference Figure 5 , in step S14, if the measurement error of the X-direction speed or the measurement error of the Y-direction speed is greater than the preset measurement error threshold, the measurement error is gradually reduced based on the further increase of the number of radars until the optimized measurement error is lower than the preset measurement error threshold;

[0121] In the specific implementation process of the present invention, the specific steps are as follows:

[0122] S141: Collect the X-direction speed and Y-direction speed of the front target, and determine the measurement error of the X-direction speed according to the X-direction speed of the front target, the real-time distance of the front target, and the measurement error mapping relationship;

[0123] S142: Determine the measurement error of the Y-direction speed according to the Y-direction speed of the front target, the real-time distance of the front target, and the measurement error mapping relationship; compare the measurement error of the X-direction speed and the measurement error of the Y-direction speed with the preset measurement error threshold respectively;

[0124] S143: If the measurement error of the X-direction speed or the measurement error of the Y-direction speed is greater than the preset measurement error threshold, trigger the detection and control of the front target by multiple radars; in the detection and control of the front target by multiple radars, increase the number of multiple radars one by one to further increase the number of radars, construct a corresponding system of equations based on the further increase of the number of radars, and determine the optimized measurement error according to this system of equations and the least squares method, and the optimized measurement error is lower than the preset measurement error threshold;

[0125] When the number of radars is 3, the system of equations is as follows:

[0126]

[0127] According to the least squares method, the optimized measurement error can be obtained:

[0128]

[0129] At this time, at this time, v 1x and v 1y are respectively the X-direction speed and Y-direction speed of the front target; the optimized measurement error The angle between the line connecting the second radar and the front target and the X-axis is α2; the angle between the line connecting the first radar and the front target and the X-axis is α1; the angle between the line connecting the third radar and the front target and the X-axis is α3; the radial speed of the front target measured by the second radar is v 12L ; the radial speed of the front target measured by the second radar is v 11L .

[0130] In the embodiment of the present application, the X-direction speed and Y-direction speed of the front target are collected, and the measurement error of the X-direction speed is determined according to the X-direction speed of the front target, the real-time distance of the front target, and the measurement error mapping relationship, which takes into account the overall consideration of the X-direction speed of the front target, the real-time distance of the front target, and the measurement error mapping relationship, and ensures the accuracy of the measurement error of the X-direction speed.

[0131] At this time, sensors of the autonomous driving system (such as radar, LiDAR, or camera, etc.) are used to capture the motion information of the front target; these sensors can measure the speed components of the front target in the X-direction (usually the main direction of vehicle travel) and the Y-direction (perpendicular to the X-direction); these speed data are the basis for subsequent calculation of the measurement error.

[0132] A preset measurement error mapping relationship is used; this mapping relationship is obtained based on a large amount of experimental data or simulation models, and it describes the error distribution of sensor-measured speed under different speed and distance conditions; specifically, taking the X-direction speed and real-time distance of the front target (obtained through steps such as S133) as inputs, querying or calculating the corresponding measurement error mapping relationship, so as to obtain the measurement error of the X-direction speed; this error represents the difference between the sensor measurement value and the true value.

[0133] Specifically, assume the following scenario: the X-direction speed of the front target is 50 km / h (about 13.9 m / s); the real-time distance is 80 meters, which is the dynamic distance between the vehicle and the front target calculated through step S133; the measurement error mapping relationship is obtained based on experimental data, and it shows that under the conditions of a speed of 50 km / h and a distance of 80 meters, the error of radar-measured speed is ±2 km / h (about ±0.56 m / s); now, perform specific calculations: the radar sensor measures the X-direction speed of the front target to be 50 km / h.

[0134] According to the measurement error mapping relationship, it is known that under the conditions of a speed of 50 km / h and a distance of 80 meters, the measurement error is ±2 km / h; therefore, the measurement error of the X-direction speed is expressed as ±0.56 m / s (i.e., converting ±2 km / h to meters per second); through this example, we can see how step S141 works in the autonomous driving system; it first collects the X-direction and Y-direction speeds of the target ahead through sensors, and then uses the preset measurement error mapping relationship to determine the measurement errors of these speeds; this error information is crucial for subsequent speed correction, path planning, and obstacle avoidance decisions, etc.; in practical applications, these steps will be more complex, involving more data processing and algorithm optimization, but the basic principle is similar.

[0135] Furthermore, determine the measurement error of the Y-direction speed according to the Y-direction speed of the target ahead, the real-time distance of the target ahead, and the measurement error mapping relationship; compare the measurement error of the X-direction speed and the measurement error of the Y-direction speed with the preset measurement error threshold respectively, taking into account the overall consideration of the Y-direction speed of the target ahead, the real-time distance of the target ahead, and the measurement error mapping relationship, to ensure the accuracy of the measurement error of the Y-direction speed.

[0136] At this time, use a method similar to step S141 to determine the measurement error of the Y-direction speed of the target ahead; the difference is that this time, the focus is on the speed component in the Y-direction; the Y-direction speed of the target ahead, the real-time distance (also obtained through steps such as S133), and the preset measurement error mapping relationship; use the Y-direction speed and the real-time distance as inputs, query or calculate the measurement error mapping relationship, and obtain the measurement error of the Y-direction speed; output the measurement error of the Y-direction speed, indicating the difference between the measured speed and the true speed of the sensor in the Y-direction.

[0137] In this step, compare the measurement errors of the X-direction speed and the Y-direction speed with the preset measurement error threshold respectively; for the measurement errors in the X-direction and Y-direction, check whether they exceed the preset threshold respectively; if the measurement error in any direction exceeds the threshold, it means that the measurement accuracy of the sensor is insufficient, and further measures need to be taken to improve the measurement accuracy; if the measurement error exceeds the threshold, trigger the radar detection control strategy in step S143 to optimize the measurement by increasing the number of radars or adjusting the radar parameters.

[0138] Specifically, assume the following scenario: the Y-direction speed of the target ahead is 10 km / h (about 2.8 m / s); the real-time distance is 80 meters, which is calculated through step S133 before; the measurement error mapping relationship shows that under the conditions of a speed of 10 km / h and a distance of 80 meters, the error of the radar in measuring the Y-direction speed is ±1.5 km / h (about ±0.42 m / s); the preset measurement error threshold is ±0.5 m / s. Now, let's conduct specific calculations and comparisons: according to the measurement error mapping relationship, the measurement error of the Y-direction speed is ±0.42 m / s; the measurement error of the X-direction speed (assuming it has been calculated in step S141 before, here it is assumed to be ±0.3 m / s) does not exceed the preset threshold of ±0.5 m / s; the measurement error of the Y-direction speed of ±0.42 m / s also does not exceed the preset threshold of ±0.5 m / s; in this example, since the measurement errors in both the X-direction and Y-direction do not exceed the preset threshold, the radar detection control strategy in step S143 does not need to be triggered; however, in actual applications, if the measurement error exceeds the threshold, the autonomous driving system will take a series of measures to optimize the measurement, such as increasing the number of radars, adjusting the scanning angle or frequency of the radars, or fusing data from other sensors to improve the measurement accuracy; this example demonstrates the role of step S142 in the autonomous driving system, that is, by comparing the measurement error with the preset threshold, to evaluate the accuracy of the sensor measurement and trigger further optimization measures as needed.

[0139] Therefore, if the measurement error of the X-direction speed or the measurement error of the Y-direction speed is greater than the preset measurement error threshold, the detection and control of the target ahead by multiple radars will be triggered; in the detection and control of the target ahead by multiple radars, the number of multiple radars is increased one by one to further increase the number of radars, a corresponding system of equations is constructed based on the further increase in the number of radars, and the optimized measurement error is determined according to this system of equations and the least squares method. The optimized measurement error is lower than the preset measurement error threshold, realizing the optimization of the measurement error with the increase of the radars, and further ensuring the accuracy of the X-direction speed and Y-direction speed of the target ahead.

[0140] At this time, the system first checks whether the measurement errors of the X-direction speed and Y-direction speed calculated in step S142 exceed the preset measurement error threshold; if the measurement error in any direction exceeds the threshold, the system will enter the next radar detection control process; when the measurement error exceeds the threshold, the system will trigger multiple radars to detect the target ahead simultaneously; the purpose of this is to improve the measurement accuracy and reliability through the data fusion of multiple radars; initially, a basic number of radars are selected for detection, and then the number of radars is gradually increased as needed.

[0141] The system will increase the number of radars one by one according to the preset strategies or algorithms. The purpose of increasing the number of radars is to collect more data points in order to construct a more accurate system of equations in the subsequent steps. At the same time, increasing the number of radars also improves the redundancy and fault tolerance of the system. As the number of radars increases, the system will construct a system of equations based on the data provided by each radar. This system of equations involves multiple unknowns (such as the speed and position of the target ahead), and each equation represents the relationship between the radar measurement data and the true state of the target.

[0142] Apply the least squares method to solve the system of equations. In this step, the system will use mathematical optimization methods such as the least squares method to solve the system of equations. The least squares method is a commonly used data fitting method that finds the optimal solution in the presence of measurement errors. Through the least squares method, the system calculates more accurate forward target speeds and other relevant parameters. After solving the system of equations, the system will calculate the optimized measurement error based on the new measurement data. Then, this optimized measurement error is compared with the preset measurement error threshold to evaluate the optimization effect. If the optimized measurement error still exceeds the threshold, it is necessary to continue to increase the number of radars or adjust other parameters for further optimization.

[0143] When the optimized measurement error is lower than the preset measurement error threshold, the system will terminate the process of increasing the number of radars and solving the system of equations. At this time, the system believes that it has obtained sufficiently accurate forward target speed information and continues with subsequent path planning, obstacle avoidance decision-making, and other operations.

[0144] Specifically, assume the following scenario: The preset measurement error threshold is ±0.3 m / s. The measured velocity error in the X direction calculated in step S142 is ±0.2 m / s (not exceeding the threshold), and the measured velocity error in the Y direction is ±0.4 m / s (exceeding the threshold). Initially, the system selects two radars to detect the target ahead. Now, for the specific operations and calculations: Since the measured velocity error in the Y direction exceeds the threshold, the system enters the radar detection control process. The system triggers the two radars to detect the target ahead simultaneously.

[0145] The system decides to add a radar, and at this time, there are three radars for detection; based on the data provided by the three radars, the system constructs a system of equations with multiple unknowns; the system uses the least squares method to solve the system of equations and obtains more accurate information about the speed of the target ahead; the system calculates that the optimized measurement error in the Y direction of speed is ±0.25 m / s (lower than the preset threshold) according to the new measurement data; since the optimized measurement error is lower than the preset threshold, the system terminates the process of increasing the number of radars and solving the system of equations; in this example, by increasing the number of radars and applying the least squares method to solve the system of equations, the system successfully reduces the measurement error of the speed in the Y direction and makes it lower than the preset threshold; this demonstrates the role of step S143 in the autonomous driving system, that is, by dynamically adjusting the number of radars and using mathematical optimization methods to improve the accuracy and reliability of measurement.

[0146] When the number of radars is 3, the system of equations is as follows:

[0147]

[0148] According to the least squares method, the optimized measurement error can be obtained:

[0149]

[0150] At this time, v 1x and v 1y are the speed in the X direction and the speed in the Y direction of the target ahead respectively; the optimized measurement error The angle between the line connecting the second radar and the target ahead and the X-axis is α2; the angle between the line connecting the first radar and the target ahead and the X-axis is α1; the angle between the line connecting the third radar and the target ahead and the X-axis is α3; the radial speed of the target ahead measured by the second radar is v 12L ; the radial speed of the target ahead measured by the second radar is v 11L .

[0151] Reference Figure 6 , in step S15, if the target ahead is not detected by multiple radars simultaneously, then the multiple radars perform independent speed measurement with respect to the target ahead, and determine the speed in the X direction and the speed in the Y direction of the target ahead based on the angles corresponding to the linear speeds detected by the multiple radars for the target ahead;

[0152] In the specific implementation process of the present invention, the specific steps are as follows:

[0153] S151: Collect the position coordinates of the target ahead. If the position coordinates of the target ahead are not in the overlapping area between multiple radars, at the same time, the distance between the first radar and the target ahead is less than the radar distance resolution of the first radar, the distance between the second radar and the target ahead is less than the radar distance resolution of the second radar, and the target ahead is not detected by multiple radars simultaneously;

[0154] S152: Multiple radars perform independent speed measurement on a forward target. The forward target outputs corresponding target speeds under the independent speed measurement of multiple radars, and combined data is formed based on the fusion of multiple target speeds. At this time, the combined data includes the angles corresponding to the linear speeds detected by multiple radars for the forward target.

[0155] S153: The angles corresponding to the linear speeds detected by multiple radars for the forward target are operated on under an approximate formula, and the speed of the forward target in the X direction and the speed in the Y direction are determined.

[0156] Approximate formula:

[0157] v x = v 1iL cos(α1), v y = v 1iL sin(α1);

[0158] At this time, the angle between the line connecting the first radar and the forward target and the X-axis is α1; v 1iL 、v 1iL are the speed parameters of the combined data, v x is the approximate speed of the forward target in the X direction; v y is the approximate speed of the forward target in the Y direction.

[0159] In the embodiments of the present application, the position coordinates of the forward target are collected. If the position coordinates of the forward target are not in the overlapping area between multiple radars, and at the same time, the distance between the first radar and the forward target is less than the radar distance resolution of the first radar, and the distance between the second radar and the forward target is less than the radar distance resolution of the second radar, the forward target is not detected by multiple radars simultaneously.

[0160] At this time, the autonomous driving system scans and detects the forward target through the radar system equipped thereon; the radar system can emit electromagnetic waves and receive the signals reflected from the target, so as to determine the position coordinates of the target; these position coordinates are usually represented in the form of (x, y, z) in a three-dimensional space, where x and y represent the position of the target on the horizontal plane, and z represents the height of the target.

[0161] The autonomous driving system is usually equipped with multiple radars, and the detection ranges of these radars will partially overlap; in this step, the system needs to determine whether the position coordinates of the forward target are located within these overlapping areas; if the target is located in the overlapping area, then multiple radars detect the target simultaneously, so as to provide richer data for subsequent processing.

[0162] The radar range resolution refers to the minimum distance that the radar can distinguish between two adjacent targets; in this step, the system needs to calculate the distance between the first radar and the target ahead and compare it with the radar range resolution of the first radar; if the distance is less than the resolution, then the first radar cannot accurately distinguish the target ahead from the surrounding background or noise, resulting in detection failure or increased error; at the same time, this step is similar to step three, but for the second radar (and other radars); the system needs to calculate the distance between each radar and the target ahead respectively and compare it with the corresponding radar range resolution; if the distance between a certain radar and the target is less than its resolution, then that radar cannot accurately detect the target either.

[0163] The system synthesizes the judgment results of steps two to four; if the position coordinates of the target ahead do not fall within the overlapping area between multiple radars and the distance between at least one radar and the target is less than its radar range resolution, then the system determines that the target ahead is not detected simultaneously by multiple radars; this is due to reasons such as the target being too close to a certain radar, the target being located in the radar detection blind area, or the performance limitations of the radar system itself.

[0164] Specifically, assume the following scenario: An autonomous vehicle is equipped with two radars: Radar A and Radar B; the radar range resolution of Radar A is 1 meter, and the radar range resolution of Radar B is 0.8 meter; the true position coordinates of the target ahead are (3 meters, 4 meters, 2 meters), that is, 3 meters ahead of the vehicle, 4 meters to the right, and the height is 2 meters; now, let's perform specific judgments and calculations: The system detects the position coordinates of the target ahead as (3 meters, 4 meters, 2 meters) through the radar system.

[0165] Assume that the detection ranges of Radar A and Radar B are circular areas centered on the vehicle with radii of 5 meters and 4 meters respectively; then, the position coordinates (3 meters, 4 meters) of the target ahead are within the detection range of Radar A but not within the detection range of Radar B (because the horizontal distance from the vehicle center is 5 meters, exceeding the 4-meter detection range radius of Radar B), and even less within the overlapping area between the two; the distance between Radar A and the target ahead is 3 meters, which is greater than the 1-meter range resolution of Radar A; therefore, Radar A can detect the target ahead.

[0166] If the target is within the detection range of Radar B, it is necessary to calculate its distance from Radar B; but in this example, the target is not within the detection range of Radar B, so this step is not applicable; but for the purpose of illustrating the method, assume that the target is within the detection range of Radar B and the distance is 0.6 meter, then this will be less than the 0.8-meter range resolution of Radar B; however, in this specific scenario, this step of judgment is not required.

[0167] Since the target ahead is not within the overlapping detection area of Radar A and Radar B, and (if the target were within the detection range of Radar B) at least one radar (Radar B) would have a distance to the target less than its radar range resolution (but in this example, the target is not within the detection range of Radar B at all), but in this scenario, only Radar A needs to be concerned; because Radar A can detect the target (the distance is greater than its resolution), and no other radar detects it simultaneously (because the target is not within the overlapping area and not within the detection range of Radar B); but the statement here is a bit confusing. Actually, it should be said that "the target ahead is not detected simultaneously by multiple radars, but at least one radar (Radar A) detects the target".

[0168] Furthermore, multiple radars perform independent speed measurement on the target ahead. The target ahead outputs corresponding target speeds under the independent speed measurement of multiple radars, and combined data is formed based on the fusion of multiple target speeds. At this time, the combined data includes the angles corresponding to the linear speeds detected by multiple radars for the target ahead.

[0169] At this time, the autonomous driving system uses multiple radars equipped to measure the speed of the target ahead; each radar independently emits electromagnetic waves and receives the signals reflected from the target, and calculates the rate of change of the distance between the target and the radar by measuring the round-trip time of the signal, that is, the radial speed (also known as the linear speed) of the target; this speed represents the linear motion speed of the target relative to the radar.

[0170] After each radar completes the speed measurement, it will output a corresponding target speed value; this speed value is calculated by the radar based on the received signal, and it reflects the motion speed of the target in the direction of the radar's line of sight; since the autonomous driving system is usually equipped with multiple radars, multiple target speed values will be obtained; these speed values are different due to the positions, detection angles of the radars and the actual motion state of the target; in order to obtain more accurate target speed information, the system needs to fuse these speed values from different radars; the fusion process involves algorithms such as weighted average and Kalman filtering to improve the accuracy and robustness of speed measurement.

[0171] When fusing the speed data of multiple radars, in addition to the speed values themselves, the angle between the linear speed detected by each radar and the target ahead also needs to be considered; this angle is the angle between the radar's line of sight and the reference coordinate system of the autonomous driving system (such as the vehicle coordinate system); by considering this angle, the system converts the speed data of different radars to the same coordinate system, so as to perform more accurate fusion processing; the combined data will finally include the linear speed detected by each radar and its corresponding angle information.

[0172] Specifically, an autonomous vehicle is equipped with two radars: Radar A and Radar B. Radar A and Radar B are respectively located on the left and right sides of the vehicle, forming a certain angle with the vehicle's longitudinal axis (for example, Radar A forms a 30-degree angle with the longitudinal axis, and Radar B forms a -30-degree angle with the longitudinal axis). The target ahead is an object moving in a straight line at a constant speed. Now, let's conduct a specific step-by-step analysis: Radar A and Radar B respectively emit electromagnetic waves towards the target ahead and receive the reflected signals. By measuring the round-trip time of the signals, Radar A and Radar B respectively calculate the radial velocity (linear velocity) of the target ahead.

[0173] Radar A outputs a velocity value representing the motion velocity of the target ahead in the line-of-sight direction of Radar A. Radar B also outputs a velocity value representing the motion velocity of the target ahead in the line-of-sight direction of Radar B. The system receives the velocity values from Radar A and Radar B. Since the detection angles of Radar A and Radar B are different, the velocity values they detect are also different. The system uses algorithms such as weighted average or Kalman filtering to fuse the velocity values from different radars to obtain more accurate target velocity information.

[0174] During the fusion process, the system also needs to consider the angles between the linear velocities detected by Radar A and Radar B and the target ahead (i.e., the angles between the radar line-of-sight directions and the vehicle's longitudinal axis). These angle information are used to convert the velocity data from different radars to the same coordinate system (such as the vehicle coordinate system). Finally, the combined data will include the linear velocities detected by Radar A and Radar B and their corresponding angle information, as well as the fused target velocity value.

[0175] Specifically, assume that the linear velocity detected by Radar A is 5 m / s, and the angle with the target ahead is 30 degrees; the linear velocity detected by Radar B is 4.5 m / s, and the angle with the target ahead is -30 degrees. The system calculates the true velocity (including magnitude and direction) of the target ahead in the vehicle coordinate system by fusing these two velocity values and their angle information. This true velocity will more accurately reflect the actual motion state of the target ahead and provide key information for the subsequent decision-making of the autonomous driving system.

[0176] Therefore, the angles corresponding to the linear velocities detected by multiple radars for the target ahead are operated under an approximate formula, and the velocities of the target ahead in the X direction and Y direction are determined. Further, the velocities of the target ahead in the X direction and Y direction are controlled, taking into account the situation where the target ahead is not detected by multiple radars simultaneously, and the accurate control of the velocities of the target ahead in the X direction and Y direction is achieved.

[0177] At this time, the autonomous driving system collects detection data from multiple radars, including the linear velocity of the front target detected by each radar and the angle between this linear velocity and the vehicle reference coordinate system (usually the longitudinal axis of the vehicle); this data is the output result of previous steps (such as S152); the system uses approximate formulas to calculate the X-direction and Y-direction velocities of the front target in the vehicle reference coordinate system based on the linear velocity and angle data of the radars; these approximate formulas are usually based on the principles of trigonometry or the concept of velocity decomposition.

[0178] For each radar, the following formulas are used to calculate the velocity components of the target in the X-direction and Y-direction: X-direction velocity component = linear velocity * cos(angle); Y-direction velocity component = linear velocity * sin(angle); note that the angle here is the angle between the radar line of sight and the X-axis of the vehicle reference coordinate system, and it needs to be converted to an angle consistent with the positive direction of the X-axis (if the angle is negative, use its absolute value and consider the directionality).

[0179] After obtaining the X-direction and Y-direction velocity components of each radar, the system needs to further process this data to determine the final velocity of the front target; this usually involves weighted averaging the velocity components from different radars or applying more complex fusion algorithms to consider the reliability of the radars, the differences in detection angles, and the complexity of the target movement; finally, the system outputs the velocities of the front target in the X-direction and Y-direction, which are defined in the vehicle reference coordinate system and can be used for subsequent decision-making and control of the autonomous driving system.

[0180] Specifically, two radars are equipped on the autonomous driving vehicle: Radar A and Radar B; Radar A is located on the left side of the front of the vehicle and forms an angle of 30 degrees with the longitudinal axis of the vehicle; Radar B is located on the right side of the front of the vehicle and forms an angle of -30 degrees with the longitudinal axis of the vehicle (i.e., in the direction opposite to the positive direction of the longitudinal axis); the front target is an object moving in a certain direction at a constant speed (in the vehicle reference coordinate system); now, let's perform specific step analysis and calculations: the linear velocity detected by Radar A is 5 m / s and the angle is 30 degrees; the linear velocity detected by Radar B is 4.5 m / s and the angle is -30 degrees; however, for simplicity, directly consider the case where the angle of Radar B is -30 degrees but the velocity direction is negative.

[0181] Apply the approximate formula for calculation:

[0182] For Radar A:

[0183] X-direction velocity component = 5 m / s * cos(30 degrees) = 4.33 m / s (positive direction)

[0184] Y-direction velocity component = 5 m / s * sin(30 degrees) = 2.5 m / s (positive direction)

[0185] For Radar B (note that the velocity direction is negative):

[0186] The velocity component in the X direction = 4.5 m / s * cos(-30 degrees) = 3.87 m / s (but since it is in the negative direction, it is actually in the opposite direction to that detected by Radar A)

[0187] The velocity component in the Y direction = 4.5 m / s * sin(-30 degrees) = -2.25 m / s (negative direction)

[0188] However, it should be noted here that since the detection angles of Radar A and Radar B are relative, and the target is moving in front of the vehicle, their velocity components in the Y direction should be the same (except for the difference in direction); in this example, due to simplifying the problem and directly using the negative angle of Radar B for calculation, the velocity components in the Y direction are inconsistent; in actual applications, the relative positions of the radars and the actual movement direction of the target should be considered to correctly calculate the velocity components; for the sake of simplicity in explanation, assume that the target is moving obliquely in front of the vehicle, and the detections of Radar A and Radar B are accurate (although there are errors in reality); in this case, the true velocity of the target is estimated by weighted averaging the velocity components of the two radars (taking into account their angles and relative reliabilities).

[0189] Approximate formula:

[0190] v x = v 1iL cos(α1), v y = v 1iL sin(α1);

[0191] At this time, the angle between the line connecting the first radar and the target in front and the X-axis is α1; v 1iL 、v 1iL are the velocity parameters of the combined data, v x is the approximate velocity of the target in front in the X direction; v y is the approximate velocity of the target in front in the Y direction.

[0192] Please refer to Figure 7 , Figure 7 which is a schematic structural composition diagram of the vehicle's speed measurement system for a target based on multiple radars in an embodiment of the present invention; the vehicle's speed measurement system for a target based on multiple radars includes:

[0193] A radar data combination module 21, configured to collect the position coordinates and angles of the target in front of the vehicle by both radars to form two sets of radar data combinations;

[0194] A radial velocity module 22, configured to determine the corresponding radial velocity based on the two sets of radar data combinations;

[0195] The first velocity module 23 is configured to determine the X-direction velocity and Y-direction velocity of a front target according to the included angle corresponding to two radial velocities;

[0196] The error module 24 is configured to, if the measurement error of the X-direction velocity or the measurement error of the Y-direction velocity is greater than a preset measurement error threshold, gradually reduce the measurement error based on a further increase in the number of radars until the optimized measurement error is lower than the preset measurement error threshold, and then output the optimized measurement error;

[0197] The second velocity module 25 is configured to, if the front target is not detected by multiple radars simultaneously, perform independent speed measurement on the front target by multiple radars, and determine the X-direction velocity and Y-direction velocity of the front target according to the included angle corresponding to the linear velocities detected by the multiple radars for the front target.

[0198] For any combination of the technical features of the above embodiments, for the sake of brevity of description, not all combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

Claims

1. A method for measuring the speed of a target based on multiple radars for a vehicle, characterized in that, Including: Both radars collect the position coordinates and included angles of the front targets of the vehicle to form two sets of radar data combinations; Based on the two sets of radar data combinations, determine the corresponding radial velocities; According to the included angles corresponding to the two radial velocities, determine the X-direction velocity and Y-direction velocity of the front target; If the measurement error of the X-direction velocity or the measurement error of the Y-direction velocity is greater than the preset measurement error threshold, then gradually reduce the measurement error based on the further increase in the number of radars until the optimized measurement error is lower than the preset measurement error threshold; If the front target is not detected by multiple radars simultaneously, then multiple radars perform independent speed measurement relative to the front target, and based on the included angles corresponding to the linear velocities detected by the multiple radars for the front target, determine the X-direction velocity and Y-direction velocity of the front target.

2. The vehicle speed measurement method for a target based on multiple radars according to claim 1, wherein Both of the two radars collect the position coordinates and included angles of the front targets of the vehicle to form two sets of radar data combinations, including: The two radars are arranged in the front of the vehicle and detect the front targets of the vehicle along the forward direction. At this time, there is an overlapping area in the detection ranges of the two radars; When the front target is in the overlapping area, the two radars detect the same front target and collect the position coordinates and included angles of the front targets of the vehicle. At this time, the two radars are the first radar and the second radar respectively; In the detection data of the first radar, use the position coordinates and included angles of the front targets of the vehicle as the first set of radar data combinations; in the detection data of the second radar, use the position coordinates and included angles of the front targets of the vehicle as the second set of radar data combinations.

3. The vehicle speed measurement method for a target based on multiple radars according to claim 1, characterized in that, The determining the corresponding radial velocities based on the two sets of radar data combinations includes: Based on the screening of each set of radar data combinations, determine the position coordinates of the front targets of the vehicle, and perform directional detection on the position coordinates of the front targets of the vehicle to collect the current velocity of the front targets of the vehicle; According to the operation of the current velocity of the front targets of the vehicle and the corresponding included angles, determine the corresponding radial velocities. At this time, there are two radial velocities, which correspond to the two radars.

4. The vehicle speed measurement method for a target based on multiple radars according to claim 1, characterized in that, The determining the X-direction velocity and Y-direction velocity of the front target according to the included angles corresponding to the two radial velocities includes: Collect the two radial velocities and the corresponding included angles, and construct a system of binary linear equations according to the two radial velocities and the corresponding included angles; According to the solution of the system of binary linear equations, determine the X-direction velocity and Y-direction velocity of the front target; Collect the current velocity of the vehicle, and determine the velocity difference between the vehicle and the front target according to the comparison of the current velocity of the vehicle, the X-direction velocity and Y-direction velocity of the front target; according to the velocity difference and time, determine the position change amount between the vehicle and the front target, and according to the position change amount between the vehicle and the front target and the relative distance between the vehicle and the front target, determine the real-time distance between the vehicle and the front target, and dynamically trigger the optimization of the dynamic detection mode of multiple radars for the same front target based on the regulation of the real-time distance between the vehicle and the front target.

5. The vehicle speed measurement method for a target based on multiple radars according to claim 4, wherein The determining the X-direction velocity and Y-direction velocity of the front target according to the included angles corresponding to the two radial velocities further includes: In the system of binary linear equations, the system of binary linear equations is: At this time, v 1x and v 1y are the X-direction speed and Y-direction speed of the front target respectively; the angle between the line connecting the second radar and the front target and the X-axis is α2; the angle between the line connecting the first radar and the front target and the X-axis is α1; the radial speed of the front target measured by the second radar is v 12L ; the radial speed of the front target measured by the second radar is v 11L .

6. The vehicle speed measurement method for a target based on multiple radars according to claim 1, characterized in that, If the measurement error of the X-direction speed or the measurement error of the Y-direction speed is greater than the preset measurement error threshold, the measurement error is gradually reduced based on the further increase of the number of radars until the optimized measurement error is lower than the preset measurement error threshold, including: Collect the X-direction speed and Y-direction speed of the front target, and determine the measurement error of the X-direction speed according to the X-direction speed of the front target, the real-time distance of the front target, and the measurement error mapping relationship; Determine the measurement error of the Y-direction speed according to the Y-direction speed of the front target, the real-time distance of the front target, and the measurement error mapping relationship; compare the measurement error of the X-direction speed and the measurement error of the Y-direction speed with the preset measurement error threshold respectively; If the measurement error of the X-direction speed or the measurement error of the Y-direction speed is greater than the preset measurement error threshold, trigger the detection and control of the front target by multiple radars; in the detection and control of the front target by multiple radars, increase the number of multiple radars one by one to further increase the number of radars, construct a corresponding equation set based on the further increase of the number of radars, and determine the optimized measurement error according to the equation set and the least squares method, and the optimized measurement error is lower than the preset measurement error threshold.

7. The vehicle speed measurement method for a target based on multiple radars according to claim 6, characterized in that, If the measurement error of the X-direction speed or the measurement error of the Y-direction speed is greater than the preset measurement error threshold, the measurement error is gradually reduced based on the further increase of the number of radars until the optimized measurement error is lower than the preset measurement error threshold, and it also includes: When the number of radars is 3, the equation set is as follows: According to the least squares method, the optimized measurement error can be obtained: At this time, v 1x and v 1y are the X-direction speed and Y-direction speed of the front target respectively; the optimized measurement error The angle between the line connecting the second radar and the front target and the X-axis is α2; the angle between the line connecting the first radar and the front target and the X-axis is α1; the angle between the line connecting the third radar and the front target and the X-axis is α3; the radial velocity of the front target measured by the second radar is v 12L ; the radial velocity of the front target measured by the second radar is v 11L .

8. The vehicle speed measurement method for a target based on multiple radars according to claim 1, wherein If the front target is not detected by multiple radars at the same time, multiple radars perform independent speed measurement on the front target, and determine the X-direction speed and Y-direction speed of the front target based on the included angle corresponding to the linear speeds detected by multiple radars for the front target, including: Collect the position coordinates of the front target. If the position coordinates of the front target are not in the overlapping area between multiple radars, at the same time, the distance between the first radar and the front target is less than the radar distance resolution of the first radar, the distance between the second radar and the front target is less than the radar distance resolution of the second radar, and the front target is not detected by multiple radars at the same time; Multiple radars perform independent speed measurement on the front target, and the front target outputs corresponding target speeds under the independent speed measurement of multiple radars, and combined data is formed according to the fusion of multiple target speeds. At this time, the combined data includes the included angle corresponding to the linear speeds detected by multiple radars for the front target; The included angle corresponding to the linear speeds detected by multiple radars for the front target is operated under the approximate formula, and the X-direction speed and Y-direction speed of the front target are determined.

9. The vehicle speed measurement method for a target based on multiple radars according to claim 8, wherein If the front target is not detected by multiple radars at the same time, multiple radars perform independent speed measurement on the front target, and determine the X-direction speed and Y-direction speed of the front target based on the included angle corresponding to the linear speeds detected by multiple radars for the front target, and it also includes: Approximate formula: v x = v 1iL cos(α1), v y = v 1iL sin(α1); At this time, the angle between the line connecting the first radar and the forward target and the X-axis is α1; v 1iL , v 1iL are the velocity parameters of the combined data, and v x is the approximate velocity of the forward target in the X direction; v y is the approximate velocity of the forward target in the Y direction.

10. A vehicle speed measurement system for a target based on multiple radars, characterized in that, The vehicle speed measurement system for a target based on multiple radars is applied to the vehicle speed measurement method for a target based on multiple radars as described in any one of claims 1-9. The vehicle speed measurement system for a target based on multiple radars includes: A radar data combination module, configured to collect the position coordinates and angles of the front target of the vehicle by both radars to form two sets of radar data combinations; A radial velocity module, configured to determine the corresponding radial velocity based on the two sets of radar data combinations; A first velocity module, configured to determine the X-direction velocity and Y-direction velocity of the front target according to the angles corresponding to the two radial velocities; An error module, configured to, if the measurement error of the X-direction velocity or the measurement error of the Y-direction velocity is greater than a preset measurement error threshold, gradually reduce the measurement error based on the further increase in the number of radars until the optimized measurement error is lower than the preset measurement error threshold; A second velocity module, configured to, if the front target is not detected by multiple radars simultaneously, perform independent speed measurement of the multiple radars relative to the front target, and determine the X-direction velocity and Y-direction velocity of the front target based on the angles corresponding to the linear velocities detected by the multiple radars for the front target.

Images