Search

CN-113744327-B - Method, control device, computer program and storage medium for classifying measurement points of a point cloud

CN113744327BCN 113744327 BCN113744327 BCN 113744327BCN-113744327-B

Abstract

A method for classifying measurement points of a point cloud (in particular, a point cloud acquired by a lidar sensor, a radar sensor and/or a camera sensor) acquired by at least one sensor by a control device is disclosed, wherein local surface vectors to adjacent measurement points are acquired for any measurement point of the point cloud, the angle between the local surface vectors relative to a gravity vector is calculated for any local surface vector, a maximum surface vector and a normalized surface vector having a maximum angle relative to the gravity vector are acquired for any measurement point of the point cloud based on the calculated angle, and any measurement point of the point cloud having a normalized surface vector and/or a maximum surface vector is classified as a non-ground point, wherein the angle of the normalized surface vector and/or the maximum surface vector relative to the gravity vector is greater than a limit value. A control device, a computer program and a machine readable storage medium are also disclosed.

Inventors

  • FU CHENGXUAN

Assignees

  • 罗伯特·博世有限公司

Dates

Publication Date
20260505
Application Date
20210531
Priority Date
20200529

Claims (7)

  1. 1. A method for classifying measurement points (4) of a point cloud (P) determined by at least one sensor (2) by a control device (6), For any measuring point (4) of the point cloud (P), a local surface vector (10) is determined to an adjacent measuring point (12), -Calculating the angle (WD) between any local surface vector (10) with respect to the gravity vector (g) for said local surface vector (10) respectively, For any measuring point (4) of the point cloud (P), a maximum surface vector (20) and a normalized surface vector (16, 18) with a maximum angle (WD) relative to the gravity vector (g) are determined on the basis of the calculated angle (WD), Classifying any measured point of the point cloud having a normalized surface vector (16, 18) and/or a maximum surface vector (20) as a non-ground point, the angle (WD) of the normalized surface vector and/or the maximum surface vector with respect to the gravity vector (g) being greater than a threshold value, Wherein for each unclassified measurement point of the point cloud (P), a height value (z) is compared with the height of the sensor (2) above the ground, wherein if the height value (z) of an unclassified measurement point of the point cloud (P) corresponds to the height of the sensor (2) above the ground, the unclassified measurement point is classified as a ground point, wherein a measurement point (4) of the point cloud (P) classified as a ground point having at least one unclassified neighboring measurement point (12) is determined, and an area growing method (24) is applied.
  2. 2. The method of claim 1, wherein the point cloud is determined by a lidar sensor, a radar sensor, and/or a camera sensor.
  3. 3. Method according to claim 1 or 2, wherein unclassified neighboring measurement points (12) of measurement points classified as non-ground points are found, wherein unclassified neighboring measurement points (12) of measurement points classified as non-ground points are classified as non-ground points, said unclassified neighboring measurement points having the same azimuth angle (WA) and a higher or the same elevation angle (WE).
  4. 4. Method according to any one of claims 1 or 2, wherein the measuring points (4) of the point cloud (P) are stored at least temporarily in a structured form with a plurality of rows and columns in a storage unit configured as a machine-readable storage medium (8).
  5. 5. Control device (6), the control device (6) being configured to classify measurement points (4) of a point cloud (P) determined by means of at least one sensor (2), wherein the control device (6) is configured to implement the method according to any one of claims 1 to 4.
  6. 6. A computer program product comprising instructions which, when the computer program product is implemented by a computer or a control device (6), cause the computer or the implementation control device to implement the method according to any of claims 1 to 4.
  7. 7. A machine-readable storage medium (8) on which a computer program product according to claim 6 is stored.

Description

Method, control device, computer program and storage medium for classifying measurement points of a point cloud Technical Field The invention relates to a method for classifying measuring points of a point cloud determined by at least one sensor, in particular of a lidar sensor, a radar sensor and/or a camera sensor. The invention also relates to a control device, a computer program and a machine-readable storage medium. Background In the field of automated driving assistance functions and automated driving, a lidar sensor, a radar sensor, or a camera sensor is generally used as an environment sensor to perform environment sensing. The environment can be scanned by means of an environment sensor in order to determine a plurality of measuring points in the form of a three-dimensional point cloud, which have distance information to the objects in the scanning area. In this case, for example, a travel Time measurement or so-called Time-of-Flight measurement is performed, and the distance travelled by the emitted beam is calculated from the measured travel Time. In order to detect objects from measurement points of a point cloud, it is generally necessary to classify the measurement points into ground points and non-ground points that are assigned to the ground. Subsequent object recognition is performed based on the measurement points classified as non-ground points. Methods for classifying measurement points of a point cloud are known. However, the known methods are complex and therefore require high computational power. Because of the complexity of these methods, real-time processing of the measurement data can only be achieved with high technical outlay. Furthermore, the known methods are inadequate in terms of false positive rate (Falsch-Positiv-Raten) and false negative rate (Falsch-Negativ-Raten). Disclosure of Invention The task on which the invention is based may be seen as proposing a method for classifying measurement data, which method has a reduced computational power requirement and is real-time. The object is achieved by a method, a control device, a computer program and a machine-readable storage medium for classifying measuring points of a point cloud determined by at least one sensor, in particular of a lidar sensor, a radar sensor and/or a camera sensor. Advantageous configurations of the invention are also given below. According to one aspect of the invention, a method for classifying measurement points of a point cloud determined by at least one sensor is provided. The at least one sensor can determine measurement data in the form of measurement points and can be configured, for example, as a lidar sensor, a radar sensor and/or a camera sensor. The method may be implemented by a control device. The control device may be configured as a field programmable gate array (Field Programmable GATE ARRAY, FPGA), an Application Specific Integrated Circuit (ASIC), a microprocessor, a computer, or as a hardware accelerator. The algorithm may also be implemented in an FPGA, ASIC, or other type of hardware accelerator to reduce CPU load. In one step of the method, local surface vectors to adjacent measurement points are found for any measurement point of the point cloud. The local surface vector can be determined as a normal vector and oriented toward the adjacent measuring point. Here, for any measurement point, a surface vector to its neighboring point or neighboring measurement points is first calculated. Alternatively, each unknown surface vector may also be calculated by a cross product (Kreuzprodukt) of two known surface vectors. The angles between the corresponding local surface vectors with respect to the gravity vector are calculated for any local surface vector, respectively. Then, for any measured point of the point cloud, a maximum surface vector and a normalized surface vector having a maximum angle with respect to the gravity vector are found based on the calculated angle. The gravity vector is oriented perpendicular to the ground (Untergrund or Boden) corresponding to the gravitational force. The angle between the local surface vector with respect to the gravity vector may preferably lie in the range between 0 ° and 90 °, including 0 ° and 90 °. In the case of an angle greater than 90 °, the angle can be determined in the form of subtraction from 180 °. The maximum surface vector may be determined for any measurement point. Any measurement point may be assigned at least one surface vector. In the case where one measurement point is assigned a plurality of surface vectors directed from the measurement point to an adjacent measurement point, the surface vector having the largest angle with respect to the gravity vector may be defined as the largest surface vector. Thus, the maximum surface vector may be oriented substantially parallel with respect to the X-Y plane, whereby the angle with respect to the gravity vector is greatest. The normalized surface vector can be calculated by constructin