Sensing range calibration method and device, electronic equipment and storage medium

Abstract

The invention provides a sensing range calibration method and device, electronic equipment and a storage medium. The calibration method comprises the following steps: acquiring positions, speeds and movement directions of a plurality of target objects on a road; obtaining the actual position, speed and movement direction of each target object; acquiring three preset areas on a road; a probability density function corresponding to each preset area is generated, a weighted probability density function corresponding to an intersection area of the three preset areas is generated, the target area meeting the condition that the weighted probability density function is larger than or equal to tau is in the intersection area, and the sensing range of the roadside sensing system is the intersection of the preset sensing area and the target area. The calibration method can calibrate the sensing range of the roadside sensing system.

Description

Technical Field

[0001] This invention relates to the field of vehicle-road-cloud integration, and in particular to a method, apparatus, electronic device, and storage medium for calibrating sensing range. Background Technology

[0002] Vehicle-road-cloud integration refers to the integration of elements such as people, vehicles, roads, and the cloud through next-generation information and communication technologies to achieve collaborative perception, decision-making, and control of the transportation system. This concept originated in the field of intelligent connected vehicles and is considered an important path to achieving high-level autonomous driving.

[0003] In the vehicle-road-cloud integrated system, roadside perception systems are usually installed on the side of the road. These systems are typically equipped with sensors (such as cameras, lidar, and millimeter-wave radar) to acquire information such as traffic conditions.

[0004] In practical applications, it is necessary to calibrate the sensing range of these sensors, and the higher the accuracy of this calibration, the better. In existing technologies, the following calibration methods are typically used:

[0005] 1. Sensor Model-Based Geometric Projection Method: This method typically calculates the sensing range directly through 3D modeling based on geometric parameters such as the sensor's field of view and installation location. However, this method relies on an idealized mathematical model, considering only the sensor's theoretical parameters while completely ignoring the impact of sensor measurement noise on sensing accuracy. In practical applications, radar suffers from distance measurement errors, cameras exhibit pixel distortion, and lidar experiences point cloud data deviations due to environmental factors. These noise sources significantly affect the actual sensing performance of the sensor.

[0006] 2. Static Threshold Method Based on Signal Strength: Traditional sensor sensing range calibration methods typically employ a static threshold division strategy, defining the effective sensing area by setting a fixed threshold for the sensor's physical signal. For example, with LiDAR, this method determines target presence based on a factory-set echo intensity threshold; for cameras, it uses a signal-to-noise ratio threshold to define the clear imaging area. However, this calibration method is highly dependent on the sensor's factory parameters and has significant technical drawbacks: the distance measured by this method generally refers to the sensor's detection distance. The "effective" distance at which different sensors can truly identify targets is often much smaller than their detection distance. This effective distance is also strongly correlated with the target's reflectivity and the sensing algorithm, easily leading to misunderstandings by design units, who may misuse the detection distance as the sensing range parameter, resulting in inadequate design.

[0007] 3. Post-hoc calibration method based on target detection: This method defines the perception range by using the farthest reliable recognition distance of a known target (such as a standard-sized obstacle). In practice, test targets need to be manually and repeatedly placed in the test area, and the sensor's response data to the targets is obtained through point-by-point testing. This method has significant limitations: First, manually placing test targets is not only time-consuming and labor-intensive, but also easily introduces human error, making it difficult to guarantee the consistency and reliability of the calibration results. Second, its detection mode can only acquire discrete point perception boundary data, and cannot generate a continuous spatial perception range surface based on these discrete data, resulting in a fragmented perception range that is difficult to intuitively reflect the sensor's true perception capability in three-dimensional space. Third, in dynamic traffic scenarios, the relative positions and motion states of vehicles and obstacles are constantly changing. This static testing-based calibration method cannot meet the real-time perception range calibration requirements, severely restricting the accurate application of sensors in roadside perception scenarios.

[0008] Therefore, how to calibrate the sensing range of the roadside sensing system has become an urgent problem to be solved. Summary of the Invention

[0009] The purpose of this invention is to provide a method, apparatus, electronic device, and storage medium for calibrating sensing range.

[0010] To achieve one of the aforementioned objectives, an embodiment of the present invention provides a method for calibrating the sensing range of a roadside sensing system, comprising the following steps: controlling the roadside sensing system to acquire the position Pos1, speed Vel1, and movement direction Yaw1 of several target objects on the highway; and acquiring the actual position Pos2, actual speed Vel2, and actual movement direction Yaw2 of each target object; and acquiring a preset area T on the highway. Pos Preset area T Vel and preset area T yaw In the preset area T Pos In the context of the target object, the absolute value of the difference between its position Pos1 and position Pos2 is less than the first threshold; within the preset region T... Vel In the context of the target object, the absolute value of the difference between velocity Vel1 and velocity Vel2 is less than the second threshold; within the preset region T... yaw In the process, the angle between the movement directions Yaw1 and Yaw2 corresponding to each target object is less than a third threshold; wherein the first, second, and third thresholds are all greater than zero; the preset sensing area A of the roadside perception system on the highway is obtained. design Generate a preset region T Pos The corresponding probability density function is f pos (x,y), preset region T vel The corresponding probability density function is f vel(x,y), and the preset region T yaw The corresponding probability density function is f yaw (x,y)); Generate the preset region T Pos T vel and T yaw The weighted probability density function f corresponding to the intersection region total (x,y), the intersection region satisfies f total The region A with (x,y)≥τ effective The sensing range of the roadside sensing system is A. design ∩A effective , where τ>0.

[0011] As a further improvement to one embodiment of the present invention, the preset region T Pos Corresponding probability density function Among them, the preset area T Pos The number of target objects in the middle is n pos , will n pos The target objects are arranged in a queue, and the X-coordinate of the i-th target object is x. i The Y-axis coordinate is y i K is the Gaussian kernel function, h is the bandwidth parameter; preset region T vel Corresponding probability density function Among them, the preset area T vel The number of target objects in the middle is n vel , will n vel The target objects are arranged in a queue, and the X-coordinate of the i-th target object is x. i The Y-axis coordinate is y i Preset area T yaw Corresponding probability density function Among them, the preset area T yaw The number of target objects in the middle is n yaw , will n yaw The target objects are arranged in a queue, and the X-coordinate of the i-th target object is x. i The Y-axis coordinate is y i .

[0012] As a further improvement to one embodiment of the present invention, f total (x,y)=ω1f pos (x,y)+ω2f vel (x,y)+

[0013] ω3f yaw (x,y), where ω1+ω2+ω3=1, and ω1, ω2 and ω3 are all greater than zero.

[0014] As a further improvement to one embodiment of the present invention, the preset region T Pos T vel and T yaw All are trapezoidal.

[0015] This invention also provides a calibration device for the sensing range of a roadside sensing system, comprising the following modules: an information acquisition module, used to control the roadside sensing system to acquire the position Pos1, speed Vel1, and movement direction Yaw1 of several target objects on the highway; and to acquire the actual position Pos2, actual speed Vel2, and actual movement direction Yaw2 of each target object; and to acquire a preset area T on the highway. Pos Preset area T Vel and preset area T yaw In the preset area T Pos In the context of the target object, the absolute value of the difference between its position Pos1 and position Pos2 is less than the first threshold; within the preset region T... Vel In the context of the target object, the absolute value of the difference between velocity Vel1 and velocity Vel2 is less than the second threshold; within the preset region T... yaw In the process, the angle between the movement directions Yaw1 and Yaw2 corresponding to each target object is less than a third threshold; wherein the first, second, and third thresholds are all greater than zero; the processing module is used to acquire the preset sensing area A of the roadside perception system on the highway. design Generate a preset region T Pos The corresponding probability density function is f pos (x,y), preset region T vel The corresponding probability density function is f vel (x,y), and the preset region T yaw The corresponding probability density function is f yaw (x,y)); Generate the preset region T Pos T vel and T yaw The weighted probability density function f corresponding to the intersection region total (x,y), the intersection region satisfies f total The region A with (x,y)≥τ effective The sensing range of the roadside sensing system is A. design ∩A effective , where τ>0.

[0016] As a further improvement to one embodiment of the present invention, the preset region T Pos Corresponding probability density function Among them, the preset area T Pos The number of target objects in the middle is n pos , will n posThe target objects are arranged in a queue, and the X-coordinate of the i-th target object is x. i The Y-axis coordinate is y i K is the Gaussian kernel function, h is the bandwidth parameter; preset region T vel Corresponding probability density function Among them, the preset area T vel The number of target objects in the middle is n vel , will n vel The target objects are arranged in a queue, and the X-coordinate of the i-th target object is x. i The Y-axis coordinate is y i Preset area T yaw Corresponding probability density function Among them, the preset area T yaw The number of target objects in the middle is n yaw , will n yaw The target objects are arranged in a queue, and the X-coordinate of the i-th target object is x. i The Y-axis coordinate is y i .

[0017] As a further improvement to one embodiment of the present invention, f total (x,y)=ω1f pos (x,y)+ω2f vel (x,y)+

[0018] ω3f yaw (x,y), where ω1+ω2+ω3=1, and ω1, ω2 and ω3 are all greater than zero.

[0019] As a further improvement to one embodiment of the present invention, the preset region T Pos T vel and T yaw All are trapezoidal.

[0020] This invention also provides an electronic device, including a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the calibration method described above.

[0021] This invention also provides a storage medium storing a program or instructions, which, when executed by a processor, implement the steps of the calibration method described above.

[0022] Compared to existing technologies, the technical advantages of this invention are as follows: This invention provides a method, apparatus, electronic device, and storage medium for calibrating the sensing range. The calibration method includes the following steps: acquiring the position, speed, and direction of motion of several target objects on a highway; acquiring the actual position, speed, and direction of motion of each target object; acquiring three preset regions on the highway; generating a probability density function corresponding to each preset region; and generating a weighted probability density function corresponding to the intersection region of the three preset regions. The intersection region contains target regions satisfying a weighted probability density function ≥ τ. The sensing range of the roadside sensing system is the intersection of the preset sensing regions and the target regions. This calibration method can calibrate the sensing range of the roadside sensing system. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the calibration method in an embodiment of the present invention;

[0024] Figure 2 , Figure 3 , Figure 4 and Figure 5 This is a scene diagram of the experiment in an embodiment of the present invention;

[0025] Figure 6 This is an interface diagram showing the parameter settings for the calibration method experiment in this embodiment of the invention;

[0026] Figure 7 This is a flowchart illustrating the calibration method in an embodiment of the present invention. Detailed Implementation

[0027] The present invention will now be described in detail with reference to the embodiments shown in the accompanying drawings. However, these embodiments do not limit the present invention, and any structural, methodological, or functional modifications made by those skilled in the art based on these embodiments are included within the scope of protection of the present invention.

[0028] The terms used herein, such as “above,” “over,” “below,” and “under,” indicating spatial relative position, are for illustrative purposes to describe the relationship of one unit or feature relative to another unit or feature as shown in the accompanying drawings. These terms may be intended to include different orientations of the device in use or operation other than those shown in the figures. For example, if the device in the figures is flipped, a unit described as being “below” or “under” another unit or feature would be “above” that unit or feature. Therefore, the exemplary term “below” can encompass both above and below orientations. The device may be oriented in other ways (rotated 90 degrees or otherwise) and the spatially related descriptive terms used herein will be interpreted accordingly.

[0029] Embodiment 1 of the present invention provides a method for calibrating the sensing range of a roadside sensing system, such as... Figure 7 As shown, it includes the following steps:

[0030] Step 701: Control the roadside sensing system to acquire the position Pos1, velocity Vel1, and direction of motion Yaw1 of several target objects on the highway; and acquire the actual position Pos2, actual velocity Vel2, and actual direction of motion Yaw2 of each target object; acquire the preset area T on the highway. Pos Preset area T Vel and preset area T yaw In the preset area T Pos In the context of the target object, the absolute value of the difference between its position Pos1 and position Pos2 is less than the first threshold; within the preset region T... Vel In the context of the target object, the absolute value of the difference between velocity Vel1 and velocity Vel2 is less than the second threshold; within the preset region T... yaw In the above, the angle between the motion direction Yaw1 and the motion direction Yaw2 corresponding to each target object is less than the third threshold; where the first, second and third thresholds are all greater than zero.

[0031] This roadside sensing system can be equipped with several sensors. In practice, several target objects (such as pedestrians, non-motorized vehicles, and motorized vehicles, which can be moving or stationary) can be moved at preset positions (i.e., actual positions) at preset speeds (i.e., actual speeds) and preset directions of motion (i.e., actual directions of motion). Then, the sensors are controlled to acquire road information, and the road information is processed to identify the target objects, as well as the position, speed, and direction of motion of each target object.

[0032] In the inventor's experiments, the roadside sensing system was a temporary setup, using four test pan-tilt units, such as... Figure 5 As shown, it also comes with a dedicated sensor mounting clamp, which can support the fixed installation of sensors (such as cameras, radar, etc.). After the equipment is installed, the installation angle can be controlled by the motorized pan-tilt head, and the installation height can be adjusted by the lifting rod, with a height range between 2.0 meters and 6.5 meters. Figure 2 The car used by the inventor in the experiment is shown. Figure 3 The image shows pedestrians during the inventor's experiment. Figure 4 The non-motorized vehicle used by the inventor in the experiment is shown.

[0033] In practice, the inventors conducted numerous experiments, the following table shows the sensor's installation height and angle for each experiment. In each experiment, both cars and non-motorized vehicles traveled at a constant speed of 20 km / h on the lane.

[0034] Installation height Installation angle Installation height Installation angle 4.5m 0° 6.5m 0° 4.5m 5° 6.5m 5° 4.5m 10° 6.5m 10° 4.5m 15° 6.5m 15°

[0035] Preset area T Pos Preset area T vel and preset area T yaw It can be generated automatically by the system or manually by the user. Figure 6 A user-defined preset area T is shown. Pos Preset area T vel and preset area T yaw The graphical interface allows for the selection of target objects using different initial thresholds, each displayed in a different color. Three preset perceptual trapezoidal boxes, each corresponding to a preset region T, are also provided. Pos Preset area T vel and preset area T yaw The user drags a sensing trapezoid based on the positioning accuracy sample distribution. It supports scaling and dragging the short and long sides of the trapezoid based on their endpoints. After dragging, it automatically acquires and visualizes the minimum sensing range width, blind zone length, and maximum sensing distance. After user confirmation, it outputs the sensor system's performance at a specific installation angle.

[0036] Set the preset area T Pos T Vel and T Yaw Determining whether the error between the perceived data and the actual data within these areas is less than the corresponding threshold helps to filter out areas where the perception system performs well. For example, in certain road sections, roadside perception systems may be less affected by environmental interference, and their position, speed, and direction perception errors may be within acceptable ranges. This method can help identify areas where the system performs well.

[0037] Step 702: Obtain the preset sensing area A of the roadside sensing system on the highway. design Generate a preset region T Pos The corresponding probability density function is f pos (x,y), preset region T vel The corresponding probability density function is f vel (x,y), and the preset region T yaw The corresponding probability density function is f yaw (x,y)); Generate the preset region T Pos T vel and T yaw The weighted probability density function f corresponding to the intersection region total (x,y), the intersection region satisfies f total The region A with (x,y)≥τ effective The sensing range of the roadside sensing system is A. design ∩Aeffective Where τ>0. Here, A design ∩A effective Refers to A design and A effective In the intersecting areas, when installing a roadside sensing system, users can determine the preset sensing area A based on road conditions, weather conditions, and sensor performance. design To set up, that is, the envisioned preset perception area A design In this process, the absolute values ​​of the differences between positions Pos1 and Pos2, the absolute values ​​of the differences between velocities Vel1 and Vel2, and the motion directions Yaw1 and Yaw2 for each target object should all be less than preset values.

[0038] Probability density refers to the probability of an event occurring randomly. It equals the probability of an event occurring over a given interval (the range of values ​​the event can take) divided by the length of that interval. Its value is non-negative and can be very large or very small. Let's define probability density functions f for predetermined regions representing position, velocity, and direction of motion, respectively. pos (x,y),f vel (x,y),f yaw (x, y), using kernel density estimation (e.g., Gaussian kernel function K), can describe the probability density distribution of relevant attributes within a region based on the distribution of target objects within that region. For example, f pos (x,y), based on the preset region T Pos Calculating the location-related probability density by using the coordinates of the target object can reflect information such as the density of the target object's location distribution within the area, which is helpful for analyzing the performance of the roadside perception system in terms of location perception and for subsequent comprehensive evaluation.

[0039] In practical applications, different attributes may have different levels of importance to the roadside perception system. By setting weights, different levels of attention and integration of each attribute can be given according to specific needs, so as to obtain a more comprehensive indicator reflecting the performance of the perception system.

[0040] Obtain the preset sensing area A design And by filtering out regions that satisfy f total The region A with (x,y)≥τ (τ is the threshold) is... effective The final sensing range of the roadside sensing system was determined to be A. design ∩A effectiveThis approach reasonably combines the pre-designed sensing area of ​​the system with the effective area obtained based on performance evaluation (using probability density functions and threshold judgments). It can determine the final reliable sensing range based on actual performance while taking into account the system design expectations, which helps to more accurately evaluate the effective working area of ​​the roadside sensing system in real highway scenarios.

[0041] In this embodiment, the preset region T Pos Corresponding probability density function Among them, the preset area T Pos The number of target objects in the middle is n pos , will n pos The target objects are arranged in a queue, and the X-coordinate of the i-th target object is x. i The Y-axis coordinate is y i K is the Gaussian kernel function, h is the bandwidth parameter; preset region T vel Corresponding probability density function Among them, the preset area T vel The number of target objects in the middle is n vel , will n vel The target objects are arranged in a queue, and the X-coordinate of the i-th target object is x. i The Y-axis coordinate is y i Preset area T yaw Corresponding probability density function Among them, the preset area T yaw The number of target objects in the middle is n yaw , will n yaw The target objects are arranged in a queue, and the X-coordinate of the i-th target object is x. i The Y-axis coordinate is y i .

[0042] In roadside sensing systems, after acquiring information such as the position, velocity, and direction of motion of the target object, these probability density functions f pos (x,y),f vel (x,y) and f yaw (x, y) can be used for the preset region T respectively Pos T Vel and T Yaw The distribution of target objects within the area is described. For example, f pos (x,y) is based on a preset region T Pos The coordinates of the target object (x) i y iThe function f is used to calculate and reflects the density of target objects in the location dimension within a given area. In real-world highway scenarios, the location distribution of target objects may differ across different road sections. This function can quantify these distribution characteristics, providing a basis for subsequent analysis of the performance of roadside perception systems in location perception. Similarly, f vel (x,y) and f yaw (x,y) describes the distribution of target objects within the corresponding preset area from the dimensions of velocity and direction of motion, which helps to understand the system’s performance in these attribute perceptions.

[0043] The function takes the form of kernel density estimation (e.g., Gaussian kernel function K), utilizing a finite number of target objects n within a preset region. pos n vel and n yaw Information is used to estimate the probability density distribution of the entire area. In practical applications, it is impossible to obtain precise information on all target objects on the highway. This sample-based estimation method is a feasible and commonly used approach. By reasonably selecting the bandwidth parameter h, the estimation bias and variance can be balanced to a certain extent, so that the obtained probability density function can better reflect the true distribution of the relevant attributes of target objects within the area.

[0044] For performance evaluation of roadside perception systems, position, velocity, and direction of motion are important attribute dimensions. These three probability density functions provide a quantitative evaluation metric for each attribute. For example, when analyzing the system's position perception performance, f... pos (x, y) can help determine whether the region T is within the preset area. Pos Which locations within the object are most likely to be perceived, and which locations might have perception problems? Similarly, f vel (x,y) and f yaw (x, y) are used to evaluate the performance of velocity and motion direction perception, providing effective tools for individual analysis of system performance. Therefore, these three functions form the basis of the comprehensive evaluation. Each function independently quantifies different attributes, allowing the contribution of each attribute to be considered separately during the comprehensive evaluation. Subsequently, these three probability density functions are combined in a weighted manner to form a weighted probability density function f. total (x,y) is used to comprehensively consider the impact of three attributes—position, speed, and direction of motion—on the performance of the roadside perception system.

[0045] In this embodiment, f total (x,y)=ω1f pos (x,y)+ω2f vel (x,y)+ω3f yaw(x, y), where ω1 + ω2 + ω3 = 1, and ω1, ω2, and ω3 are all greater than zero. Define the weighted probability density function f for the intersection region. total The probability density of position, velocity, and direction of motion is comprehensively considered based on different weights ω1, ω2, and ω3, where ω1 + ω2 + ω3 = 1. By setting different weights ω1, ω2, and ω3, the varying importance of each attribute in practical applications is reflected, thereby enabling a more comprehensive and accurate evaluation of the roadside perception system's performance and determination of its sensing range.

[0046] In this embodiment, the preset region T Pos T vel and T yaw All are trapezoidal. Here, for example... Figure 1 As shown, the closer the area is to the sensor, the smaller the error in the target object acquired by the sensor; the farther the area is from the sensor, the larger the error in the target object acquired by the sensor. Furthermore, the farther away from the sensor, the larger the field of view. Therefore, the preset area can be designed as a trapezoid.

[0047] The preset sensing area A is calculated based on the sensor's theoretical parameters (field of view, installation height, etc.). design The area is: Where θ min / θ msx For the horizontal field of view boundary, r min / r max To detect the distance boundary, the actual effective range R is obtained through the above probability model. effective Its area is: Design perception range is efficient Among them, A intersection =A design ∩A effective Through multiple measurements, the final value of A was obtained. effective The range is shown in the table below:

[0048]

[0049] Embodiment 2 of the present invention provides a calibration device for the sensing range of a roadside sensing system, comprising the following modules:

[0050] The information acquisition module is used to control the roadside sensing system to acquire the position Pos1, speed Vel1, and direction of movement Yaw1 of several target objects on the highway; and to acquire the actual position Pos2, actual speed Vel2, and actual direction of movement Yaw2 of each target object; and to acquire the preset area T on the highway. Pos Preset area T Vel and preset area T yaw In the preset area TPos In the context of the target object, the absolute value of the difference between its position Pos1 and position Pos2 is less than the first threshold; within the preset region T... Vel In the context of the target object, the absolute value of the difference between velocity Vel1 and velocity Vel2 is less than the second threshold; within the preset region T... yaw In the above, the angle between the motion direction Yaw1 and the motion direction Yaw2 corresponding to each target object is less than the third threshold; where the first, second and third thresholds are all greater than zero.

[0051] The processing module is used to acquire the preset sensing area A of the roadside sensing system on the highway. design Generate a preset region T Pos The corresponding probability density function is f pos (x,y), preset region T vel The corresponding probability density function is f vel (x,y), and the preset region T yaw The corresponding probability density function is f yaw (x,y)); Generate the preset region T Pos T vel and T yaw The weighted probability density function f corresponding to the intersection region total (x,y), the intersection region satisfies f total The region A with (x,y)≥τ effective The sensing range of the roadside sensing system is A. design ∩A effective , where τ>0.

[0052] In this embodiment, the preset region T Pos Corresponding probability density function Among them, the preset area T Pos The number of target objects in the middle is n pos , will n pos The target objects are arranged in a queue, and the X-coordinate of the i-th target object is x. i The Y-axis coordinate is y i K is the Gaussian kernel function, h is the bandwidth parameter; preset region T vel Corresponding probability density function Among them, the preset area T vel The number of target objects in the middle is n vel , will n vel The target objects are arranged in a queue, and the X-coordinate of the i-th target object is x. i The Y-axis coordinate is y i Preset area T yaw Corresponding probability density function Among them, the preset area T yaw The number of target objects in the middle is n yaw , will n yaw The target objects are arranged in a queue, and the X-coordinate of the i-th target object is x. i The Y-axis coordinate is y i .

[0053] In this embodiment, f total (x,y)=ω1f pos (x,y)+ω2f vel (x,y)+ω3f yaw (x,y), where ω1+ω2+ω3=1, and ω1, ω2 and ω3 are all greater than zero.

[0054] In this embodiment, the preset region T Pos T vel and T yaw All are trapezoidal.

[0055] Embodiment 3 of the present invention provides an electronic device, including a processor, a memory, and a program or instructions stored in the memory and executable on the processor. When the program or instructions are executed by the processor, they implement the steps of the calibration method as described in Embodiment 1.

[0056] Embodiment 4 of the present invention provides a storage medium storing a program or instructions, which, when executed by a processor, implement the steps of the calibration method as described in Embodiment 1.

[0057] It should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This way of describing the specification is only for clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

[0058] The detailed descriptions listed above are merely specific descriptions of feasible embodiments of the present invention, and are not intended to limit the scope of protection of the present invention. All equivalent embodiments or modifications made without departing from the spirit of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for calibrating the sensing range of a roadside sensing system, characterized in that, Includes the following steps: The roadside sensing system is controlled to acquire the position Pos1, velocity Vel1, and direction of motion Yaw1 of several target objects on the highway; and to acquire the actual position Pos2, actual velocity Vel2, and actual direction of motion Yaw2 of each target object; and to acquire the preset area T on the highway. Pos Preset area T Vel and preset area T yaw In the preset area T Pos In the context of the target object, the absolute value of the difference between its position Pos1 and position Pos2 is less than the first threshold; within the preset region T... Vel In the context of the target object, the absolute value of the difference between velocity Vel1 and velocity Vel2 is less than the second threshold; within the preset region T... yaw In the above, the angle between the motion direction Yaw1 and the motion direction Yaw2 corresponding to each target object is less than the third threshold; where the first, second and third thresholds are all greater than zero; The roadside sensing system acquires a preset sensing area A on the highway. design Generate a preset region T Pos The corresponding probability density function is f pos (x,y), preset region T vel The corresponding probability density function is f vel (x,y) and the preset region T yaw The corresponding probability density function is f yaw (x,y)); Generate the preset region T Pos T vel and T yaw The weighted probability density function f corresponding to the intersection region total (x,y), the intersection region satisfies f total The region A with (x,y)≥τ effective The sensing range of the roadside sensing system is A. design ∩A effective , where τ>0.

2. The calibration method according to claim 1, characterized in that, Preset area T Pos Corresponding probability density function Among them, the preset area T Pos The number of target objects in the middle is n pos , will n pos The target objects are arranged in a queue, and the X-coordinate of the i-th target object is x. i The Y-axis coordinate is y i K is the Gaussian kernel function, and h is the bandwidth parameter; Preset area T vel Corresponding probability density function Among them, the preset area T vel The number of target objects in the middle is n vel , will n vel The target objects are arranged in a queue, and the X-coordinate of the i-th target object is x. i The Y-axis coordinate is y i ; Preset area T yaw Corresponding probability density function Among them, the preset area T yaw The number of target objects in the middle is n yaw , will n yaw The target objects are arranged in a queue, and the X-coordinate of the i-th target object is x. i The Y-axis coordinate is y i .

3. The calibration method according to claim 2, characterized in that, f total (x,y)=ω1f pos (x,y)+ω2f vel (x,y)+ω3f yaw (x,y), where ω1+ω2+ω3=1, ω1, ω2 and ω3 is always greater than zero.

4. The calibration method according to claim 1, characterized in that, Preset area T Pos T vel and T yaw All are trapezoidal.

5. A calibration device for the sensing range of a roadside sensing system, characterized in that, Includes the following modules: The information acquisition module is used to control the roadside sensing system to acquire the position Pos1, speed Vel1, and direction of movement Yaw1 of several target objects on the highway; and to acquire the actual position Pos2, actual speed Vel2, and actual direction of movement Yaw2 of each target object; and to acquire the preset area T on the highway. Pos Preset area T Vel and preset area T yaw In the preset area T Pos In the context of the target object, the absolute value of the difference between its position Pos1 and position Pos2 is less than the first threshold; within the preset region T... Vel In the context of the target object, the absolute value of the difference between velocity Vel1 and velocity Vel2 is less than the second threshold; within the preset region T... yaw In the above, the angle between the motion direction Yaw1 and the motion direction Yaw2 corresponding to each target object is less than the third threshold; where the first, second and third thresholds are all greater than zero; The processing module is used to acquire the preset sensing area A of the roadside sensing system on the highway. design Generate a preset region T Pos The corresponding probability density function is f pos (x,y), preset region T vel The corresponding probability density function is f vel (x,y) and the preset region T yaw The corresponding probability density function is f yaw (x,y)); Generate the preset region T Pos T vel and T yaw The weighted probability density function f corresponding to the intersection region total (x,y), the intersection region satisfies f total The region A with (x,y)≥τ effective The sensing range of the roadside sensing system is A. design ∩A effective , where τ>0.

6. The calibration device according to claim 5, characterized in that, Preset area T Pos Corresponding probability density function Among them, the preset area T Pos The number of target objects in the middle is n pos , will n pos The target objects are arranged in a queue, and the X-coordinate of the i-th target object is x. i The Y-axis coordinate is y i K is the Gaussian kernel function, and h is the bandwidth parameter; Preset area T vel Corresponding probability density function Among them, the preset area T vel The number of target objects in the middle is n vel , will n vel The target objects are arranged in a queue, and the X-coordinate of the i-th target object is x. i The Y-axis coordinate is y i ; Preset area T yaw Corresponding probability density function Among them, the preset area T yaw The number of target objects in the middle is n yaw , will n yaw The target objects are arranged in a queue, and the X-coordinate of the i-th target object is x. i The Y-axis coordinate is y i .

7. The calibration device according to claim 6, characterized in that, f total (x,y)=ω1f pos (x,y)+ω2f vel (x,y)+ω3f yaw (x,y), where ω1+ω2+ω3=1, ω1, ω2 and ω3 is always greater than zero.

8. The calibration device according to claim 5, characterized in that, Preset area T Pos T vel and T yaw All are trapezoidal.

9. An electronic device, characterized in that, It includes a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the calibration method as described in any one of claims 1-4.

10. A storage medium, characterized in that, The storage medium stores a program or instructions that, when executed by a processor, implement the steps of the calibration method as described in any one of claims 1-4.

Images