Method, environmental sensor and vehicle for object classification

Abstract

The invention relates to a method for object classification of an environmental sensor, in particular a radar sensor (2, 9a-9d), for environmental detection, which emits a signal, which is reflected on an object, which is a target vehicle, and is received again by the environmental sensor, based on which the main speed (V veh ) of the target vehicle is ascertained, and the speed is ascertained for the signal reflected on at least one rim of the target vehicle, from which the dispersion (σ s ) is determined, which is derived from the deviation of the ascertained speed from the main vehicle speed (V veh ), and based on the dispersion (σ s ) the ratio (s) of the rim diameter (d Felge ) to the wheel diameter (d Rad ) is ascertained, which is used for the object classification of the target vehicle.

Description

Technical Field

[0001] The present invention relates to a method for object classification using environmental sensors, particularly radar sensors, for vehicles, and a vehicle having an environmental sensor according to the present invention. Background Technology

[0002] Modern vehicles, such as motor vehicles or motorcycles, are increasingly equipped with driver assistance systems. These systems use sensor systems to detect the environment, identify traffic conditions, and assist the driver, for example, through braking or steering intervention, or by issuing visual or audible warnings. Radar sensors, lidar sensors, camera sensors, and ultrasonic sensors are commonly used as sensor systems for environmental detection. Conclusions about the environment can then be drawn from the sensor data obtained. Environmental detection using radar sensors is based on the emission and reflection of beams of electromagnetic waves, such as those reflected by other road users, obstacles in the lane, or edge structures of the lane. Pedestrian detection is typically performed using camera sensors; however, radar sensors are increasingly being used for this purpose as well.

[0003] The radar sensors used in the aforementioned systems are often combined with sensors from other technologies, such as cameras or lidar sensors. Furthermore, radar sensors offer the advantage of reliable operation even in adverse weather conditions and can directly measure the radial relative velocity of objects via the Doppler effect, in addition to distance. Commonly used transmission frequencies are 24 GHz, 77 GHz, and 79 GHz; however, other permitted frequencies will be used in the future. As the functional range of such systems continues to expand, the requirements for maximum effective detection range and environmental classification performance are constantly increasing. Besides environmental detection of motor vehicles used in these systems, monitoring the interior space of motor vehicles is also a focus, for example, identifying which seats are occupied; frequencies in the 60 GHz range are used, for example.

[0004] For modern radar sensors, the classification of detected objects is particularly important. For example, vehicle classification is primarily based on the measured size and reflectivity of the observed or detected vehicle. Large vehicles (e.g., commercial vehicles (LKW) or transport vehicles) generate multiple, or relatively many, reflections in the sensor image. However, the actual size of the vehicle cannot be accurately determined. For instance, it cannot be definitively determined whether the detected vehicle is a truck with a trailer or two passenger cars (PKW) traveling close together at the same speed. This can lead to misclassification and negatively impact various driving functions (control strategies, human-machine interface displays, etc.), thereby limiting performance and reducing vehicle user acceptance. Therefore, for radar sensors or radar detection methods, it is particularly interesting to correctly and robustly classify different vehicle types (e.g., passenger cars (PKW) and commercial vehicles (LKW)), where better grouping of different reflections of individual vehicles is necessary to obtain the correct size. In addition, it should be able to correctly display road users on the sensor vehicle's display (e.g., human-machine interface; HMI) and implement appropriate adjustment strategies for various driving functions (e.g., adaptive cruise control (ACC) or electronic steering control (EBA)).

[0005] A radar device is known from US 2016 284 213 A1 that, during operation, determines whether a detected object is a moving vehicle based on performance values ​​of directional correlation and normalized directional correlation values. Here, the vehicle type is determined and roughly classified as a truck or passenger car by examining the properties of the echo signal components of the vehicle's rotating wheels, taking into account both the frequency shift and signal strength of the received signal. This classification is achieved as follows: reflections at the front of the vehicle (at the height of the axis of rotation) have only a small frequency shift due to the small velocity component of the rotating wheels in the longitudinal direction of the vehicle, but have high signal strength due to reflections on an almost vertical surface. Conversely, reflections at the upper side of the wheels have a large frequency shift due to the large velocity component of the rotating wheels in the longitudinal direction of the vehicle, but have low signal strength due to reflections on an almost horizontal surface, because the reflections are mainly caused by the small vertical surface of the tire tread. Summary of the Invention

[0006] Based on existing technology, the present invention aims to provide a general environmental sensor, particularly a radar sensor, for object detection, which enables improved object classification in a simple and economical manner.

[0007] The aforementioned objective is achieved through the overall teaching of independent claim 1 and the independent claims conjoining therewith. Advantageous embodiments of the invention are claimed in the dependent claims.

[0008] In the method for object classification using environmental sensors (especially radar sensors, lidar sensors, camera sensors, or ultrasonic sensors for vehicles) according to the present invention, the environmental sensor first emits a signal, which is reflected off an object and received again by the environmental sensor. If the object is a target vehicle, the target vehicle should be classified. For this purpose, the main speed of the target vehicle is determined based on the signal, and the speed is also determined based on the signal reflected from at least one rim (i.e., multiple rims) of the target vehicle. The speed dispersion is determined from the known speed, which is derived from the deviation between the known speed and the main vehicle speed. Furthermore, the ratio of the rim diameter to the wheel diameter is determined based on the dispersion, where this ratio is used for object classification of the target vehicle.

[0009] Advantageously, this method is based on velocity measurement, or radial velocity measurement, which can be determined using the Doppler effect. This involves an independent method that can improve the robustness and performance of modern radar sensors. Furthermore, it can significantly improve object classification.

[0010] The ratio is preferably determined by calculating the quotient of the wheel diameter divided by the rim diameter.

[0011] According to a preferred design of the present invention, a limit value can be specified for this ratio.

[0012] If the speed dispersion of at least one wheel of the target vehicle is less than a threshold, the target vehicle can advantageously be classified as a passenger vehicle (PKW); and / or, if the speed dispersion of at least one wheel of the target vehicle is greater than a threshold, the target vehicle can be classified as a commercial vehicle (LKW).

[0013] Furthermore, this method can also be applied to vehicle categories other than commercial vehicles / passenger vehicles, such as motorcycles, light motorcycles, four-wheeled motorcycles, electric scooters, bicycles, SUVs, vans, transport vehicles, and agricultural vehicles. Specifically, it allows for the classification of all vehicle categories by specifying (e.g., pre-determined limits or thresholds) the ratio of rim diameter to wheel diameter and the associated dispersion.

[0014] Furthermore, the reflectivity of the reflected signal can be analyzed in parallel or additionally (e.g., using radar cross section RCS) and added to the object classification of the target vehicle. This enables the implementation of an additional parameter that validates previously known results, or object classification, i.e., a verification, thereby further improving the object classification.

[0015] Advantageously, the measured dimensions of the target vehicle can also be analyzed in parallel or additionally, thereby enabling the addition of these dimensions to the object classification of the target vehicle.

[0016] Furthermore, the present invention also relates to an environmental sensor, particularly a radar sensor (or also a lidar sensor, camera sensor, or ultrasonic sensor), for object recognition or environmental detection of a vehicle, classifying objects based on sensor data of the vehicle, wherein the object classification is performed using the method according to the present invention.

[0017] Furthermore, the present invention also claims protection for a vehicle having an environmental sensor, particularly a radar sensor, according to the present invention. Attached Figure Description

[0018] The invention is further described below based on advantageous embodiments. The accompanying drawings illustrate:

[0019] Figure 1 A simplified schematic diagram showing the dispersion of wheel reflection in wheels with rim diameters larger than wheel diameters (left figure) and wheels with rim diameters smaller than wheel diameters (right figure);

[0020] Figure 2 Simplified diagrams showing a passenger car wheel with a rim diameter larger than its wheel diameter (top image) and a commercial vehicle wheel with a rim diameter smaller than its wheel diameter (bottom image);

[0021] Figure 3 Show Figure 2 A simplified schematic diagram of a wheel, where the dispersion or velocity distribution of the wheel velocity is independent of the wheel size or diameter.

[0022] Figure 4 A simplified design of the vehicle according to the present invention is shown. Detailed Implementation

[0023] The proposed method for object classification is primarily based on the velocity analysis of reflections, which deviate from the main vehicle speed V. veh This allows for classification by utilizing the intensity of the radar signal via the Doppler effect. In motor vehicles, these speed deviations primarily originate from the rotating metal rims of the wheels. For a given vehicle speed V... veh The dispersion σ of the deviation speed s Directly depends on the diameter d of the metal rim Felge With wheel diameter d Rad The ratio s, where the applicable ones are:

[0024] .

[0025] Calculate the corresponding dispersion σ sOr, in other words, the main vehicle speed V detected by the passenger car. veh The dispersion σ s1 And the detected main vehicle speed V of commercial vehicles veh The dispersion σ s2 At times, such as Figure 1 As shown, the relationships that are important for object classification can be applied:

[0026] .

[0027] However, here, the dispersion, or velocity distribution, should be independent of the wheel size or wheel diameter, and depend only on the relative dimensions of the rim (i.e., the reflecting portion) and the tire, such as... Figure 2 As shown. For example, heavy-duty trucks (LKWs) typically have rims smaller than the wheel diameter (the rim diameter relative to the total wheel diameter), and consequently have a larger σ than passenger cars. sPKW Smaller dispersion σ sLKW ,like Figure 3 As shown, the following relationship applies:

[0028] ,

[0029] thereby

[0030] .

[0031] Of course, this relationship can be applied to all vehicle categories that differ accordingly, or typically, or class-related, in terms of rim-wheel-proportion.

[0032] Advantageously, this logic or method can be further refined or improved by performing parallel or additional RCS (Radar Cross Section) analysis on the reflection of individual wheels. It is also advantageous to further refine or improve this logic or method by performing parallel or additional analysis on the geometry of the target vehicle.

[0033] Furthermore, this method can also be applied to other sensor technologies for detecting metal wheel rims, such as ultrasound.

[0034] Figure 4A vehicle 1 according to the invention is shown, which has a radar sensor 2 according to the invention for object detection or environmental detection. Reference numeral 2 here indicates a control device (ECU, electronic control unit, or ADCU, auxiliary and automated driving control unit), through which sensor control, sensor data fusion, environmental and / or object recognition, trajectory planning, and / or vehicle control are particularly capable of (semi-)automatic operation. For vehicle control, the control device 3 has access to various actuators (steering mechanism 4, engine 5, brakes 6). In addition to the radar sensor 2, the vehicle 1 also has other sensors for environmental detection (LiDAR sensor 7, camera 8, and radar sensors 9a-9d). Sensor data can be advantageously used for environmental and object recognition, thereby enabling various auxiliary functions such as Emergency Braking Assist (EBA), Adaptive Cruise Control (ACC), Lane Keeping Assist (LKA), etc. Furthermore, the execution of auxiliary functions can also be achieved via the control device 3 or other control units configured for this purpose.

[0035] List of reference numerals in the attached diagram:

[0036] 1 vehicle

[0037] 2. Radar Sensors

[0038] 3. Control equipment

[0039] 4. Steering mechanism

[0040] 5 engines

[0041] 6. Brakes

[0042] 7. LiDAR Sensor

[0043] 8 cameras

[0044] 9a-9d radar sensors.

Claims

1. A method for object classification using environmental sensors, particularly radar sensors (2, 9a-9d), for environmental monitoring, wherein, The environmental sensor emits a signal, which is reflected off the object and received again by the environmental sensor. The object is the target vehicle. Based on the signal, the main speed (V) of the target vehicle is obtained. veh ), Speed ​​is determined from signals reflected from at least one rim of the target vehicle, wherein, The dispersion (σ) is determined based on the speed. s The dispersion is determined by the known speed and the main vehicle speed (V). veh The deviation was obtained from the result. Based on dispersion (σ) s The rim diameter (d) is known. Felge ) and wheel diameter (d) Rad The proportion (s) of ) where, The ratio (s) is used for object classification of the target vehicle.

2. The method according to claim 1, characterized in that, The ratio (s) is calculated by measuring the wheel diameter (d). Rad Divide by the rim diameter (d) Felge (This information was obtained from the merchant.) 3. The method according to claim 1 or 2, characterized in that, Define a limit value for the ratio (s).

4. The method according to claim 3, characterized in that, If the velocity dispersion of at least one wheel of the target vehicle is (σ) s If the velocity dispersion (σ) of at least one wheel of the target vehicle is less than a threshold, the target vehicle is classified as a passenger vehicle; and / or, if the velocity dispersion (σ) of at least one wheel of the target vehicle is less than a threshold, the target vehicle is classified as a passenger vehicle. s If the value is greater than the threshold, the target vehicle will be classified as a commercial vehicle.

5. The method according to any one of the preceding claims, characterized in that, The reflectivity of the reflected signal is analyzed in parallel or additionally, and the reflectivity is added to the object classification of the target vehicle.

6. The method according to any one of the preceding claims, characterized in that, The measured dimensions of the target vehicle are analyzed in parallel or additionally, and the measured dimensions are added to the object classification of the target vehicle.

7. An environmental sensor, particularly a radar sensor (2, 9a-9d), for object recognition in a vehicle (1), classifying objects based on sensor data from the environmental sensor, wherein, The object classification is performed based on the method according to any one of the preceding claims.

8. A vehicle (1) having an environmental sensor, particularly a radar sensor (2, 9a-9d) according to claim 7.

Images