Radar signal processing method and device

Abstract

The invention provides a radar signal processing method and device. The method comprises the following steps: step 1, installing a preset number of ultrasonic radar probes according to a preset condition to form an ultrasonic radar system; 2, setting a measurement mode of the ultrasonic radar system as a composite measurement mode; 3, establishing a vehicle body coordinate system, and calculating the position of an obstacle; 4, generating a point cloud according to the position of the obstacle, and carrying out fusion clustering; step 5, determining a point cloud cluster corresponding to the target shielding object according to the position of the target shielding object; and step 6, processing the point cloud cluster corresponding to the target shelter according to a preset rule. According to the invention, the influence of the target shielding object on the radar system is avoided, and the product reliability is improved.

Description

Technical Field

[0001] The present invention relates to the field of radar technology, and in particular to a radar signal processing method and device. Background Art

[0002] Ultrasonic radar (USS) is a common automotive sensor, typically operating at 40kHz, 48kHz, and 58kHz. Higher operating frequencies increase sensitivity, but also reduce the horizontal and vertical detection angles. Therefore, 40kHz is the most widely used ultrasonic radar. Ultrasonic radar is waterproof and dustproof, even small amounts of mud and sand obstruct detection. Its detection range is between 0.1 and 3 meters, and its accuracy is high, making it ideal for parking assistance.

[0003] Medium and large trucks with trailer hooks also have the need to reverse. For such trucks, the main difficulty in installing an ultrasonic radar system is the blocking and reflection of radar waves by the trailer hook and the limitation of the installation position, which makes the radar unable to detect obstacles within a certain range behind the trailer hook, resulting in the risk of false alarms and increased blind spots. At the same time, since the trailer hook or rear bumper lever is usually made of metal, it will strongly reflect the ultrasonic wave and easily generate secondary echoes, which will cause radar misjudgment or false alarms, affecting the accuracy of distance and speed measurement. Summary of the Invention

[0004] The present invention provides a radar signal processing method and device, aiming to solve the defects in the prior art, avoid the influence of target obstructions on the radar system, and improve product reliability.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is:

[0006] One aspect of the present invention provides a radar signal processing method, comprising:

[0007] Step 1: Install a preset number of ultrasonic radar probes according to preset conditions to form an ultrasonic radar system;

[0008] Step 2: Setting the measurement mode of the ultrasonic radar system to a composite measurement mode;

[0009] Step 3: Establish the vehicle body coordinate system and calculate the position of the obstacle;

[0010] Step 4: Generate a point cloud based on the location of the obstacle and perform fusion clustering;

[0011] Step 5: Determine the point cloud cluster corresponding to the target occluder according to the position of the target occluder;

[0012] Step 6: Process the point cloud cluster corresponding to the target occluder according to preset rules.

[0013] Specifically, the preset number is at least 2.

[0014] Specifically, the preset condition arranges the ultrasonic radar probes evenly on both sides of the target obstruction and close to the outside of the vehicle.

[0015] Specifically, the step 3 includes:

[0016] Step 301: Establish a vehicle body coordinate system with the center of the vehicle's rear axle as the origin, the X-axis positive direction perpendicular to the vehicle's rear axle and pointing toward the rear of the vehicle, and the Y-axis positive direction parallel to the vehicle's rear axle and pointing toward the right side of the vehicle.

[0017] Step 302: Obtain the coordinates of the first ultrasonic radar in the vehicle body coordinate system and a first distance between the first ultrasonic radar and an obstacle, and obtain the coordinates of the second ultrasonic radar in the vehicle body coordinate system and a second distance between the first ultrasonic radar and the obstacle.

[0018] Step 303: Calculate the coordinate position of the obstacle according to a first preset relationship.

[0019] Specifically, the first preset relationship is:

[0020]

[0021] and:

[0022]

[0023] Where (x, y) is the coordinate of the obstacle, is the coordinate of the first ultrasonic radar in the vehicle body coordinate system, is the coordinate of the second ultrasonic radar in the vehicle body coordinate system, L k is the first distance, L k+1 is the second distance, and θ is the detection angle of the ultrasonic radar.

[0024] Specifically, step 4 includes:

[0025] Step 401: input all point clouds of the ultrasonic radar system and construct a KD tree object;

[0026] Step 402: Determine the cluster radius and cluster threshold;

[0027] Step 403: randomly select an unprocessed point from the point cloud as an initial point, and calculate the distance between the initial point and other points using the standardized Euclidean distance;

[0028] Step 404: Use KD tree search to find a point whose distance from the initial point is less than or equal to the cluster radius as a candidate point;

[0029] Step 405: determine whether the number of candidate points is greater than the clustering threshold; if so, mark the initial point as a core point, and find all objects densely connected to the initial point to form a point cloud cluster;

[0030] Step 406 : Determine whether all points in the point cloud have been processed. If so, complete the fusion clustering; otherwise, return to step 403 .

[0031] Specifically, step 5 includes:

[0032] Step 501: Obtain the installation position of the target obstruction in the vehicle body coordinate system;

[0033] Step 502: Determine the obstacle point cloud whose absolute value of the distance difference between the clustered obstacle point cloud and the installation position of the target occluder is less than a preset threshold as the point cloud corresponding to the target occluder.

[0034] Another aspect of the present invention provides a radar signal processing device, comprising: an ultrasonic radar probe, an obstacle location module, a point cloud clustering module, a point cloud determination module, and a point cloud processing module connected in sequence, wherein the ultrasonic radar probe is further connected to a working mode setting module, and the point cloud determination module is further connected to an obstruction location module;

[0035] The working mode setting module is used to set the working mode of the ultrasonic radar probe;

[0036] The ultrasonic radar probe is used to control the transmitting and receiving waves of the radar probe according to the set working mode;

[0037] The obstacle location module is used to calculate the location of obstacles;

[0038] The obstruction position module is used to set the position of the target obstruction in the vehicle body coordinate system;

[0039] The point cloud clustering module is used to generate point clouds according to the locations of obstacles and perform fusion clustering;

[0040] The point cloud determination module is used to determine the point cloud cluster corresponding to the target occluder according to the position of the target occluder;

[0041] The point cloud processing module is used to process the point cloud cluster corresponding to the target occluder according to preset rules.

[0042] Specifically, there are at least two ultrasonic radar probes.

[0043] Specifically, the ultrasonic radar probe operates in a composite measurement mode.

[0044] The beneficial effects of the present invention are as follows: the present invention installs and constructs an ultrasonic radar system according to preset conditions, obtains the point cloud of the radar system, and performs fusion clustering, and then determines the point cloud cluster corresponding to the target obstruction according to the position of the target obstruction, and then processes the point cloud cluster corresponding to the target obstruction according to preset rules, thereby avoiding the influence of the target obstruction on the radar system and improving product reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 1 is a flow chart of a radar signal processing method of the present invention;

[0046] Figure 2 This is a diagram of the installation position of the ultrasonic radar of the present invention;

[0047] Figure 3 It is a schematic structural diagram of the radar signal processing device of the present invention. DETAILED DESCRIPTION

[0048] The following describes the embodiments of the present invention in detail with reference to the accompanying drawings, which are for reference and illustration only and do not limit the scope of patent protection of the present invention.

[0049] In the processes described in the specification, claims, or drawings of the present invention, if the steps are numbered (e.g., steps 10, 20, etc.), the numbers are used solely to distinguish the steps and do not represent any order of execution. It should be noted that the terms "first," "second," etc., used herein are used solely to distinguish the objects being described and do not indicate a sequential order or indicate different types of steps.

[0050] Example 1

[0051] like Figure 1 As shown, this embodiment provides a radar signal processing method, including:

[0052] Step 1: Install a preset number of ultrasonic radar probes according to preset conditions to form an ultrasonic radar system.

[0053] In this embodiment, the preset number is at least 2, preferably 4, and the preset condition is to evenly arrange the ultrasonic radar probes on both sides of the target obstruction (such as a trailer hook) and close to the outside of the vehicle, such as Figure 2 shown.

[0054] Step 2: Set the measurement mode of the ultrasonic radar system to the composite measurement mode.

[0055] The measurement mode of ultrasonic radar is divided into direct echo measurement mode and indirect echo measurement mode according to the different echo methods. The composite measurement mode is a measurement mode that has both.

[0056] Direct echo measurement refers to using the same ultrasonic radar to send and receive signals, while indirect echo measurement (often called cross-echo measurement) refers to using different ultrasonic radars to send and receive signals.

[0057] Step 3: Establish the vehicle body coordinate system and calculate the position of the obstacle.

[0058] In this embodiment, the target obstruction, the trailer hook, can be regarded as a point obstacle.

[0059] In this embodiment, step 3 includes:

[0060] Step 301: Establish a vehicle body coordinate system with the center of the vehicle's rear axle as the origin, the X-axis perpendicular to the vehicle's rear axle and pointing toward the rear of the vehicle as the positive direction, and the Y-axis parallel to the vehicle's rear axle and pointing toward the right side of the vehicle as the positive direction.

[0061] Step 302: Obtain the coordinates of the first ultrasonic radar in the vehicle body coordinate system And the first distance L between the first ultrasonic radar and the obstacle k , obtain the coordinates of the second ultrasonic radar in the vehicle body coordinate system And the second distance L between the first ultrasonic radar and the obstacle k+1 .

[0062] Step 303: Calculate the coordinate position (x, y) of the obstacle according to a first preset relationship.

[0063] In this embodiment, the first preset relationship is:

[0064]

[0065] and:

[0066]

[0067] Where (x, y) is the coordinate of the obstacle, is the coordinate of the first ultrasonic radar in the vehicle body coordinate system, is the coordinate of the second ultrasonic radar in the vehicle body coordinate system, L k is the first distance, L k+1 is the second distance, and θ is the detection angle of the ultrasonic radar.

[0068] In this embodiment, the first ultrasonic radar and the second ultrasonic radar are not particularly limited to a certain ultrasonic radar, and only need to be different ultrasonic radars.

[0069] It is easy to understand that due to the volatility of radar signals, the position (x, y) of the obstacle calculated by the ultrasonic radar system each time is data that fluctuates within a certain range.

[0070] Step 4: Generate a point cloud based on the location of the obstacle and perform fusion clustering.

[0071] After the ultrasonic system runs for a period of time, a series of data points are obtained to form a point cloud. That is, the point cloud contains the obstacle positions (x, y) calculated multiple times. The point cloud is clustered and fused to determine the shape of the obstacle.

[0072] In this embodiment, the first method is to use the classic DBSCAN algorithm to fuse and aggregate the point clouds. This is an existing technology and will not be described in detail.

[0073] In this embodiment, the second method is to use the improved DBSCAN algorithm to fuse and aggregate the point clouds. Step 4 includes:

[0074] Step 401: Input all point clouds of the ultrasonic radar system and construct a KD tree object.

[0075] KD tree is a tree-type data structure that can store and quickly extract corresponding data in k-dimensional space, which can speed up the search of point clouds.

[0076] Step 402: Determine the cluster radius r and the cluster threshold m.

[0077] Step 403: Randomly select an unprocessed point from the point cloud as an initial point, and calculate the distance between the initial point and other points using the standardized Euclidean distance.

[0078] Using standardized Euclidean distance as the distance criterion can make the clustering results more accurate.

[0079] Step 404: Use KD tree search to find a point whose distance from the initial point is less than or equal to the cluster radius r as a candidate point.

[0080] Step 405: determine whether the number of the candidate points is greater than the clustering threshold m. If so, mark the initial point as a core point, and find all objects densely connected to the initial point to form a point cloud cluster.

[0081] Step 406 : Determine whether all points in the point cloud have been processed. If so, complete the fusion clustering; otherwise, return to step 403 .

[0082] Step 5: Determine the point cloud cluster corresponding to the target occluder according to the position of the target occluder.

[0083] In this embodiment, step 5 includes:

[0084] Step 501: Obtain the installation position of the target obstruction in the vehicle body coordinate system.

[0085] Step 502: Determine the obstacle point cloud whose absolute value of the distance difference between the clustered obstacle point cloud and the installation position of the target occluder is less than a preset threshold as the point cloud corresponding to the target occluder.

[0086] Step 6: Process the point cloud cluster corresponding to the target occluder according to preset rules.

[0087] In this embodiment, the preset rule includes: filtering or ignoring the point cloud cluster corresponding to the target occluder.

[0088] Example 2

[0089] like Figure 3 As shown, this embodiment provides a radar signal processing device, comprising: an ultrasonic radar probe, an obstacle location module, a point cloud clustering module, a point cloud determination module, and a point cloud processing module, which are connected in sequence. The ultrasonic radar probe is also connected to a working mode setting module, and the point cloud determination module is also connected to an obstruction location module.

[0090] The working mode setting module is used to set the working mode of the ultrasonic radar probe;

[0091] The ultrasonic radar probe is used to control the transmitting and receiving waves of the radar probe according to the set working mode;

[0092] The obstacle location module is used to calculate the location of obstacles;

[0093] The obstruction position module is used to set the position of the target obstruction in the vehicle body coordinate system;

[0094] The point cloud clustering module is used to generate point clouds according to the locations of obstacles and perform fusion clustering;

[0095] The point cloud determination module is used to determine the point cloud cluster corresponding to the target occluder according to the position of the target occluder;

[0096] The point cloud processing module is used to process the point cloud cluster corresponding to the target occluder according to preset rules.

[0097] In this embodiment, there are at least two ultrasonic radar probes, and preferably four.

[0098] In this embodiment, the ultrasonic radar probe operates in a composite measurement mode.

[0099] The working process of the radar signal processing device in this embodiment is the same as that of the radar signal processing method in Example 1, and will not be repeated here.

[0100] The above disclosure is only a preferred embodiment of the present invention and cannot be used to limit the scope of protection of the present invention. Therefore, equivalent changes made according to the scope of the patent application of the present invention are still within the scope covered by the present invention.

Claims

1. A radar signal processing method, characterized in that: include: Step 1: Install a preset number of ultrasonic radar probes according to preset conditions to form an ultrasonic radar system; Step 2: Setting the measurement mode of the ultrasonic radar system to a composite measurement mode; Step 3: Establish the vehicle body coordinate system and calculate the position of the obstacle; Step 4: Generate a point cloud based on the location of the obstacle and perform fusion clustering; Step 5: Determine the point cloud cluster corresponding to the target occluder according to the position of the target occluder; Step 6: Process the point cloud cluster corresponding to the target occluder according to preset rules.

2. The radar signal processing method according to claim 1, wherein: The preset number is at least 2.

3. The radar signal processing method according to claim 2, wherein: The step 3 includes: Step 301: Establish a vehicle body coordinate system with the center of the vehicle's rear axle as the origin, the X-axis positive direction perpendicular to the vehicle's rear axle and pointing toward the rear of the vehicle, and the Y-axis positive direction parallel to the vehicle's rear axle and pointing toward the right side of the vehicle. Step 302: Obtain the coordinates of the first ultrasonic radar in the vehicle body coordinate system and a first distance between the first ultrasonic radar and an obstacle, and obtain the coordinates of the second ultrasonic radar in the vehicle body coordinate system and a second distance between the first ultrasonic radar and the obstacle. Step 303: Calculate the coordinate position of the obstacle according to a first preset relationship.

4. The radar signal processing method according to claim 3, characterized in that ,, the first preset relationship is: and: Among them, (x, y) is the coordinate of the obstacle, is the coordinate of the first ultrasonic radar in the vehicle body coordinate system, is the coordinate of the second ultrasonic radar in the vehicle body coordinate system, L k is the first distance, L k+1 is the second distance, and θ is the detection angle of the ultrasonic radar.

5. The radar signal processing method according to claim 1, wherein: The step 4 comprises: Step 401: input all point clouds of the ultrasonic radar system and construct a KD tree object; Step 402: Determine the cluster radius and cluster threshold; Step 403: randomly select an unprocessed point from the point cloud as an initial point, and calculate the distance between the initial point and other points using the standardized Euclidean distance; Step 404: Use KD tree search to find a point whose distance from the initial point is less than or equal to the cluster radius as a candidate point; Step 405: determine whether the number of candidate points is greater than the clustering threshold; if so, mark the initial point as a core point, and find all objects densely connected to the initial point to form a point cloud cluster; Step 406 : Determine whether all points in the point cloud have been processed. If so, complete the fusion clustering; otherwise, return to step 403 .

6. The radar signal processing method according to claim 1, wherein: The step 5 comprises: Step 501: Obtain the installation position of the target obstruction in the vehicle body coordinate system; Step 502: Determine the obstacle point cloud whose absolute value of the distance difference between the clustered obstacle point cloud and the installation position of the target occluder is less than a preset threshold as the point cloud corresponding to the target occluder.

7. A radar signal processing device, characterized in that: include: An ultrasonic radar probe, an obstacle location module, a point cloud clustering module, a point cloud determination module, and a point cloud processing module are connected in sequence, wherein the ultrasonic radar probe is further connected to a working mode setting module, and the point cloud determination module is further connected to an obstruction location module; The working mode setting module is used to set the working mode of the ultrasonic radar probe; The ultrasonic radar probe is used to control the transmitting and receiving waves of the radar probe according to the set working mode; The obstacle location module is used to calculate the location of obstacles; The obstruction position module is used to set the position of the target obstruction in the vehicle body coordinate system; The point cloud clustering module is used to generate point clouds according to the locations of obstacles and perform fusion clustering; The point cloud determination module is used to determine the point cloud cluster corresponding to the target occluder according to the position of the target occluder; The point cloud processing module is used to process the point cloud cluster corresponding to the target occluder according to preset rules.

8. The radar signal processing device according to claim 7, characterized in that: There are at least two ultrasonic radar probes.

9. The radar signal processing device according to claim 8, characterized in that: The ultrasonic radar probe operates in a composite measurement mode.

Images