Method for supporting camera-based surroundings recognition of a mobile vehicle using road wetness information from a first ultrasonic sensor

Abstract

The present invention relates to a method and a device for supporting camera-based surroundings recognition of a moving tool (80) by means of road wetness information from a first ultrasonic sensor (30). The method comprises the following steps: detecting a first signal representing the surroundings (60) of the moving tool (80) by means of the first ultrasonic sensor (30) of the moving tool (80); detecting a second signal representing the surroundings (60) of the moving tool (80) by means of a camera (40) of the moving tool (80); obtaining road wetness information based on the first signal; selecting a predefined parameter group from a plurality of predefined parameter groups based on the road wetness information; and performing surroundings recognition based on the second signal in combination with the predefined parameter group.

Description

Technical Field

[0001] This invention relates to a method for camera-based environmental recognition of a vehicle using road moisture information from a first ultrasonic sensor. Background Technology

[0002] Existing technology knows of vehicles that perform camera-based ambient awareness to obtain information about objects in the vehicle's surroundings. This information is received and used by, for example, driver assistance systems and / or systems used for autonomous control of the vehicle. This ambient awareness is based on algorithms known in the art for image analysis and object classification, which typically use one or more classifiers for identified objects.

[0003] Furthermore, rain sensors for motor vehicles are known from existing technology, used to identify the presence of precipitation. These rain sensors are typically positioned in the upper area of ​​the vehicle's windshield and are designed to detect precipitation on the windshield. The moisture information obtained from such rain sensors can be used to select a suitable environmental recognition classifier.

[0004] Ultrasonic sensors are also known from the prior art, and their integration with vehicles is often used in parking assistance systems or similar driver assistance systems. For this purpose, such ultrasonic sensors are mostly arranged on the vehicle such that their radiation direction and detection direction are substantially horizontal relative to the vehicle, so that the distance from objects in the surrounding environment to the vehicle can be determined based on the signal propagation time of the ultrasonic signal. Summary of the Invention

[0005] According to a first aspect of the invention, a method is proposed to support camera-based ambient environment recognition of a vehicle using road moisture information from a first ultrasonic sensor. The vehicle may be, for example, a road vehicle (e.g., a motorcycle, car, transport vehicle, truck), a rail vehicle, an aircraft / aircraft, or a ship. Furthermore, the following method steps can be implemented, in whole or in part, by a device of the vehicle according to the invention. This device may include an analysis and processing unit, preferably having a data input terminal. The analysis and processing unit may be configured, for example, as an ASIC (Application Specific Integrated Circuit), FPGA (Field Programmable Gate Array), processor, digital signal processor, microcontroller, etc., and may be connected to internal and / or external storage units via information technology. Furthermore, the analysis and processing unit may be configured to execute the method according to the invention in conjunction with a computer program implemented by the analysis and processing unit.

[0006] In the first step of the method according to the invention, a first signal representing the environment surrounding the vehicle is detected by means of a first ultrasonic sensor of the vehicle. This ultrasonic sensor can be an ultrasonic sensor of the vehicle or for other purposes. Alternatively or additionally, a dedicated ultrasonic sensor can be used in the method according to the invention. The ultrasonic sensor of the vehicle can be, for example, an ultrasonic sensor of a parking assist system, or an ultrasonic sensor of other driver assistance systems of the vehicle. The ultrasonic sensor can be arranged, for example, in the front fender or in the rear region of the vehicle, or in other locations of the vehicle, such that not only tire noise can be detected, but also the lane ahead or behind and its surrounding environment. Furthermore, the ultrasonic sensor can be connected directly or indirectly (i.e., for example, through other control units of the vehicle) to the data input terminal of the analysis and processing unit according to the invention via information technology. This connection can be established, for example, via the bus system of the vehicle's onboard electrical network (e.g., CAN, LIN, MOST, Ethernet, etc.). For subsequent processing by the analysis and processing unit, the first signal received by the analysis and processing unit from the first ultrasonic sensor can be first stored in a storage unit connected to the analysis and processing unit.

[0007] In the second step of the method according to the invention, a second signal representing the environment surrounding the vehicle is detected using a camera on the vehicle. Preferably, the second signal can be detected at substantially the same time as the first signal, ensuring that the two signals each include temporally corresponding information about the surrounding environment. Due to different sensor types, different signal processing links, and signal transmission links, a temporal deviation may occur between the two signals. A preferred, but not mandatory, range for the method according to the invention is, for example, between a few milliseconds and several hundred milliseconds, or it may be within a second. The camera may be, for example, a 2D camera, a 3D camera, or an infrared camera, having a standard image resolution, a high-definition image resolution, or an ultra-high-definition image resolution. Preferably, the camera can be arranged and oriented on the vehicle such that it detects the surrounding environment located in front of the vehicle. However, this arrangement and / or orientation of the camera is not limited to this example. Similar to the first ultrasonic sensor, the camera can be directly or indirectly connected to the analysis and processing unit information according to the invention via information technology using the vehicle's onboard electrical network. In a preferred embodiment of the invention, the camera can be connected via information technology to an image processing unit of the vehicle, the image processing unit being configured to receive and process image signals from the camera. This image processing unit can also be a component of a driver assistance system or a system for the autonomous operation of the vehicle. Furthermore, in this preferred embodiment, the image processing unit can be connected via information technology to an analysis processing unit according to the invention, such that the analysis processing unit can transmit road moisture information, described in detail below, to the image processing unit. As an addition or alternative to this embodiment, the image processing unit can be a component of the analysis processing unit itself (or vice versa), so that communication between the two components can be conducted directly rather than through the vehicle's onboard electrical system. This can be achieved, for example, by implementing the logic implemented by the analysis processing unit according to the invention in the form of a computer program using the image processing unit, the logic being used to implement the method steps according to the invention.

[0008] In the third step of the method according to the invention, road moisture information is obtained based on the first signal. For this purpose, the analysis and processing unit can compare the noise level of the first signal with a predefined AND threshold for the noise level, which can be stored in a storage unit connected to the analysis and processing unit. Preferably, the predefined AND threshold for the noise level is selected such that if the noise level exceeds the predefined AND threshold, the current road is considered moist. Conversely, if the noise level is below the predefined threshold, the road surface is considered dry. The result of obtaining the road moisture information can be stored again in the storage unit. As an alternative or addition to the pure distinction between moist and dry road surfaces, the analysis and processing unit can also estimate the degree of moisture if the noise level of the first signal exceeds the predefined AND threshold by considering the magnitude by which the noise level of the first signal exceeds the predefined AND threshold. Furthermore, it is also possible to consider using multiple predefined AND thresholds for the noise level, wherein the corresponding predefined AND thresholds can correspond to different speeds and / or speed ranges of the vehicle. In other words, the advantage is that by receiving information about the current speed of the vehicle from the onboard electrical network via the vehicle's electrical system through the analysis and processing unit, the unit can select from multiple predefined thresholds based on the current speed value, since higher speeds are typically accompanied by higher noise levels in the first signal. In this way, it is possible to prevent the analysis and processing unit from incorrectly identifying a wet road surface when the vehicle's speed is high, even if the road surface is actually dry.

[0009] In the fourth step of the method according to the invention, a predefined parameter set is selected from a plurality of predefined parameter sets based on road moisture information. The plurality of predefined parameter sets may, for example, represent different configurations of a classifier set up for analyzing the surrounding environment of the vehicle and identifying objects in the surrounding environment based on a second signal. Preferably, such a classifier may be part of a computer program implemented by an image processing unit and / or an analysis processing unit. In the case where the analysis processing unit and image processing unit according to the invention are implemented as separate components, the predefined parameter set can be selected by the image processing unit based on road moisture information and, if necessary, also based on other information (e.g., the speed of the vehicle), which may be provided by the analysis processing unit via the vehicle's onboard electrical network. One purpose of using different predefined parameter sets is to use the predefined parameter set matched to the current surrounding environment (i.e., wet or dry) for surrounding environment identification. This is because, in wet surrounding environments, such as due to water (spray) kicked up by vehicles ahead, a classifier trained for dry surrounding environments typically only ensures insufficient or unreliable identification results during the surrounding environment identification process. Conversely, classifiers trained for humid environments often fail to provide optimal recognition results in dry environments. When the humidity information provided by the analysis and processing unit includes not only the distinction between humid and dry conditions but also information about the degree of humidity, a predefined set of parameters matching the corresponding measurement level can be additionally selected based on the degree of humidity.

[0010] In the fifth step of the method according to the invention, ambient environment recognition is performed based on the second signal combined with a predefined set of parameters. Because the presence of moisture information allows for the application of separately matched predefined sets of parameters to ambient environment recognition, the recognition performance of ambient environment recognition can be optimized accordingly. The resulting improved reliability of ambient environment recognition can further lead to greater safety when using vehicles.

[0011] Preferred extensions of the invention are shown in the dependent claims.

[0012] In an advantageous configuration of the invention, road wetness information can be additionally obtained based on the vehicle's speed and / or acceleration and / or motor speed. As described above, it is advantageous to consider the current speed when obtaining current road wetness information. Similarly, values ​​of the vehicle's current acceleration and / or current motor speed received via the vehicle's onboard electrical system can be advantageously considered in a similar manner.

[0013] In another advantageous configuration of the invention, a predefined set of parameters can represent the configuration of a trained self-learning system. That is, the aforementioned classifier can be implemented based on a self-learning system, such as a neural network (e.g., with a deep learning architecture). Furthermore, other types of self-learning systems can also be used. In this way, with the aid of such a self-learning system, training of the vehicle can be performed under different wet conditions, and the configuration of the corresponding trained self-learning system can be stored in the form of different predefined set of parameters.

[0014] In another advantageous configuration of the invention, alternatively or additionally, a predefined set of parameters is selected based on changes in noise levels and / or the current temperature and / or the amount of water present in the environment surrounding the vehicle. As mentioned above, changes in noise levels may be caused by varying amounts of water on the road surface. However, changes in noise levels may also be caused by changes in the distance to a vehicle traveling ahead. In both cases, it would be meaningful to use a correspondingly matched set of parameters for environmental identification due to the consequently altered visual conditions. The distinction between changes in noise levels caused by altered water levels and changes in distance to a vehicle traveling ahead can be made through additional analysis of the second signal, for example, by determining the size change of the vehicle traveling directly ahead. Alternatively or additionally, signals from other environmental sensors of the vehicle can also be considered for assessing the current situation. Particularly advantageous here is considering distance information from the vehicle's lidar system and / or radar system to the vehicle traveling ahead. The predefined parameter set generated and used for the above situation enables improved and / or faster identification of vehicles partially obscured by water mist, even when the camera can only detect a blurry outline of the vehicle in front.

[0015] Furthermore, considering the current external temperature when selecting predefined parameter sets is meaningful, as external temperatures below 4°C, and especially below 0°C, allow for the inference of potential snow accumulation at lane edges and / or on the lane itself. If the external temperature is 0°C or lower and road moisture is detected, the probability of snow accumulation, at least at the lane edges, is particularly high. Based on this information, other suitable predefined parameter sets can be selected and used in the surrounding environment recognition process, enabling reliable identification of lane boundaries, for example, even in the presence of snow.

[0016] In another advantageous configuration of the invention, the interference-free nature of the first signal can be considered. The road moisture information is obtained by means of various interfering factors that make reliable identification of road moisture difficult or even impossible, such as buildings at the road edge and / or other vehicles adjacent to the vehicle. Such interfering factors can be obtained, for example, based on a second signal or signals from other surrounding environmental sensors (e.g., lidar and / or radar sensors). In cases where such interfering factors exist, the analysis and processing unit can transmit road moisture information to the image processing unit representing the road moisture state before the occurrence of the interfering factors. Preferably, this value can continue to be used as road moisture information by the entire system for such a long period until one or more interfering factors disappear from the environment surrounding the vehicle. In this way, time-limited interfering factors can be advantageously avoided, as their temporary appearance and disappearance do not cause undesirable frequent changes in the road moisture information. This, in turn, avoids undesirable frequent switching of predefined parameter sets.

[0017] As described above, the first ultrasonic sensor can be arranged on the vehicle such that its detection range is in or opposite to the direction of travel of the vehicle. Additionally, the environment around the vehicle can be detected based on a second ultrasonic sensor, particularly by means of a second ultrasonic sensor, which is also arranged on the vehicle such that its detection range is in or opposite to the direction of travel of the vehicle. In a preferred variant, for example, the first ultrasonic sensor can be oriented in the direction of travel of the vehicle, and the second ultrasonic sensor can be oriented opposite to the direction of travel of the vehicle. In this way, road moisture information can be obtained based on two ultrasonic sensors, thereby enabling additional reliability checks on the road moisture information obtained from the corresponding first signals. Alternatively, road moisture information can be obtained alternately based on the first or second ultrasonic sensor by analyzing the first signals of the ultrasonic sensors in terms of road moisture, wherein the ultrasonic sensor has the lowest share of interference at the current moment. It should be noted that, in addition to the first and second ultrasonic sensors, other ultrasonic sensors can also be used in the method according to the invention. That is, for example, third, fourth, and more ultrasonic sensors can be used, and they can be combined and used in a configuration similar to that described above. The arrangement of the first, second, third, fourth, and other ultrasonic sensors, if necessary, is explicitly not limited to the front and / or rear regions of the vehicle.

[0018] In another advantageous configuration of the invention, the reliability of the road moisture information obtained from the first signal can be verified by the road moisture information obtained from the second signal. This can be achieved, for example, based on the analysis of light source reflections in a camera image, by checking, for example, whether the light source is above or appears to be below the roadway plane. Alternatively, the reliability of the road moisture information can also be verified by other sensors and / or control devices of the vehicle. Here, for example, a rain sensor mounted on the windshield of the vehicle, or other sensors of the vehicle, could also be considered.

[0019] Subsequently, the information about objects in the surrounding environment obtained by the surrounding environment identification can be transmitted to driver assistance systems and / or systems for autonomous control of the vehicle, and the information can be used there.

[0020] According to a second aspect of the invention, a device is provided to support camera-based ambient environment recognition of a vehicle using road moisture information from a first ultrasonic sensor. The device includes an analysis and processing unit and a data input terminal. The analysis and processing unit may be configured as, for example, an ASIC, FPGA, processor, digital signal processor, microcontroller, etc., and can be connected to internal and / or external storage units via information technology. Furthermore, the analysis and processing unit may be configured to execute the method according to the invention in conjunction with a computer program implemented by the analysis and processing unit. Additionally, the analysis and processing unit is configured to detect a first signal and a second signal in conjunction with the data input terminal, the first signal being obtained by means of the first ultrasonic sensor of the vehicle and representing the ambient environment of the vehicle, and the second signal being obtained by means of a camera of the vehicle and representing the ambient environment of the vehicle. Preferably, the ultrasonic sensor may be an existing ultrasonic sensor of the vehicle. Furthermore, the ultrasonic sensor may be arranged, for example, in the front fender, or in the rear region of the vehicle, or in other locations on the vehicle, such that either the roadway ahead or the roadway behind and its surrounding environment can be detected. The camera may be, for example, a 2D or 3D camera with standard image resolution, high-definition image resolution, or ultra-high-definition image resolution, or it may be an infrared camera. Preferably, the camera may be arranged and oriented on the vehicle such that it can detect the surrounding environment in front of the vehicle. The analysis and processing unit may be directly and / or indirectly connected to the ultrasonic sensors and camera via information technology through the vehicle's onboard electrical network. Furthermore, the analysis and processing unit is configured to obtain road moisture information based on a first signal, select a predefined parameter group from multiple predefined parameter groups based on the road moisture information, and perform surrounding environment recognition based on a second signal combined with the predefined parameter group. Attached Figure Description

[0021] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings. The drawings show:

[0022] Figure 1 A flowchart illustrating the steps of an embodiment of the method according to the present invention is shown;

[0023] Figure 2 A block diagram of the device according to the invention is shown in conjunction with the means of transportation;

[0024] Figure 3 A graph showing the velocity-dependent noise level of the first ultrasonic sensor is displayed. Detailed Implementation

[0025] Figure 1A flowchart illustrating the steps of an embodiment of a method according to the invention is shown, which uses road moisture information from a first ultrasonic sensor 30 to support camera-based ambient environment recognition of a vehicle 80. In a first step 100, a first signal representing the ambient environment 60 of the vehicle 80 is detected by the first ultrasonic sensor 30 of the vehicle 80 using an analysis processing unit 10 (microcontroller) according to the invention. The first ultrasonic sensor 30 is disposed in the front fender of the vehicle 80 and oriented in the direction of travel of the vehicle 80. The analysis processing unit 10 receives the first signal via its data input terminal 12 and stores the ambient environment information represented by the first signal in the internal storage unit 20 of the microcontroller. In step 200, a second signal representing the ambient environment 60 of the vehicle 80 is detected by a camera 40 of the vehicle 80. The camera 40 is disposed in the upper region of the windshield of the vehicle 80 within the interior space of the vehicle 80, and is oriented such that it can detect the ambient environment 60 located in front of the vehicle 80. The image processing unit of the vehicle 80 receives a second signal from the camera 40, which is connected to the camera 40 via information technology. The first signal from the ultrasonic sensor 30 and the second signal from the camera 40 are detected substantially at the same time. In step 300, road moisture information is obtained based on the first signal using a computer program implemented by the analysis processing unit 10. For this purpose, the analysis processing unit 10 compares the noise level 70 of the first signal with a predefined threshold 75 for the noise level 70. If the noise level 70 exceeds the predefined threshold 75, it can be inferred that road moisture exists in the surrounding environment 60 of the vehicle 80. In this case, because road moisture is identified based on the predefined threshold 75, the analysis processing unit 10 sends a corresponding signal, including the current road moisture information, to the image processing unit via the vehicle bus of the vehicle's electrical network. In step 400 of the method according to the invention, the image processing unit selects a predefined parameter group from a plurality of predefined parameter groups based on the received road moisture information. In this context, the parameter set selected by the image processing unit represents the configuration of a neural network-based classifier, which was trained earlier (e.g., during the development phase of the vehicle 80) under similar wet road conditions. Subsequently, in step 500, the image processing unit performs ambient environment recognition based on a second signal combined with the predefined parameter set. The information obtained through ambient environment recognition (regarding objects in the surrounding environment 60 of the vehicle 80) is then transmitted via the vehicle's electrical network to a system for the autonomous control of the vehicle 80, and is used by this system during the autonomous control of the vehicle 80.

[0026] Figure 2A block diagram of the device according to the invention is shown in conjunction with a vehicle 80. The device includes an analysis and processing unit 10 (in this case, a microcontroller) and has a data input terminal 12. Through the data input terminal 12, the analysis and processing unit 10 is connected via information technology to a first ultrasonic sensor 30, and via information technology to a second ultrasonic sensor 35, the first ultrasonic sensor being oriented in the direction of travel of the vehicle 80, and the second ultrasonic sensor being oriented in the opposite direction of travel. Also via the data input terminal 12, the analysis and processing unit 10 is connected via information technology to a camera 40 oriented in the direction of travel of the vehicle 80. Furthermore, the analysis and processing unit 10 is connected via information technology to an external storage unit 20, which is configured to store information received by the analysis and processing unit 10 for subsequent processing. With the aid of the first ultrasonic sensor 30, the second ultrasonic sensor 35, and the camera 40, the analysis and processing unit 10 is able to detect the surrounding environment 60 of the vehicle 80 at substantially the same time. In this example, all the steps of the method according to the invention are implemented in the analysis and processing unit 10 itself. That is, the analysis and processing unit 10 is not only set to obtain road moisture information based on the first signals of the first ultrasonic sensor 30 and the second ultrasonic sensor 35, but also set to select a predefined parameter group corresponding to the road moisture information, and perform surrounding environment recognition based on the second signal of the camera 40 by means of the predefined parameter group.

[0027] Figure 3A graph showing the speed-dependent noise level 70 of the first ultrasonic sensor 30 is presented. In the first phase P1 of this graph, the vehicle 80 travels at a speed v corresponding to a predefined AND threshold 75 for the first phase P1, and the vehicle uses the first ultrasonic sensor 30 in the sense of the method according to the invention. In other words, since the vehicle 80 initially travels at a relatively low speed v in the first phase P1, a predefined AND threshold 75 from a plurality of predefined AND thresholds 75 is used for comparison with the noise level 70 of the first signal: this predefined AND threshold has been previously determined for this speed range. Because the noise level 70 is significantly higher than the predefined AND threshold 75 for the first phase P1 in the first phase P1, the presence of road wetness is determined by the analysis processing unit 10 according to the invention. It can be seen in the change of speed v that the speed v of the vehicle 80 continuously increases over time. Upon reaching a speed value v1, the analysis processing unit 10 selects a predefined AND threshold 75 for the second phase P2 based on the currently higher speed v: this predefined AND threshold is different from the predefined AND threshold 75 for the first phase P1. Therefore, the predefined threshold 75 of the second stage P2 is matched with the noise level 75 generated by the higher speed v. As in the first stage P1, the presence of road wetness is identified again here because the noise level 70 is higher than the predefined threshold 75 of the second stage P2 at the beginning of the second stage P2. At time t1 in the second stage P2, the curve of the noise level 70 drops below the predefined threshold 75 of the second stage P2. In response, the analysis processing unit 10 determines that there is a dry lane surface.

Claims

1. A method for supporting camera-based surroundings recognition of a moving tool (80) by means of road wetness information from a first ultrasonic sensor (30), the method comprising the following steps: Detecting (100) a first signal representing the surroundings (60) of the traveling tool (80) by means of a first ultrasonic sensor (30) of the traveling tool (80); Detecting (200) a second signal representing the surroundings (60) of the traveling tool (80) by means of a camera (40) of the traveling tool (80); road wetness information is ascertained (300) based on the first signal, wherein the road wetness information is ascertained by comparing a speed-dependent noise level (70) of the first signal with a speed-dependent predefined threshold value (75) of the noise level (70), wherein an estimation of the degree of wetness is also performed if the noise level of the first signal exceeds the predefined threshold value, wherein the road wetness information ascertained from the first signal is subjected to a plausibility check using the road wetness information ascertained from the second signal; selecting (400) a predefined parameter group from a plurality of predefined parameter groups according to the road moisture information, wherein a predefined parameter group matching a corresponding moisture level is selected according to the moisture level, the plurality of predefined parameter groups representing different configurations of a classifier, the classifier being configured to analyze a surrounding environment of the traveling tool and identify an object in the surrounding environment based on the second signal; A surrounding environment recognition is performed (500) based on the second signal in combination with the predefined parameter set.

2. The method according to claim 1, wherein: The first signal and the second signal are detected at substantially the same time.

3. The method according to claim 1 or 2, wherein: The road wetness information is additionally ascertained (300) as a function of the speed and / or acceleration and / or motor speed of the travel device (80).

4. The method according to claim 1 or 2, wherein: The set of predefined parameters represents the configuration of a trained self-learning system.

5. The method according to claim 1 or 2, wherein: Alternatively or additionally, the set of predefined parameters is selected (400) as a function of a change in the noise level (70) and / or a current outside temperature and / or an amount of water present in the surroundings (60) of the traveling tool (80).

6. The method according to claim 1 or 2, wherein: The road wetness information is determined (300) based on the interference-free nature of the first signal.

7. The method according to claim 1 or 2, wherein: The first ultrasonic sensor (30) is arranged on the traveling tool (80) in such a way that a detection range of the first ultrasonic sensor (30) is in the traveling direction of the traveling tool (80) or opposite to the traveling direction.

8. The method according to claim 1 or 2, wherein: The surroundings (60) are additionally detected based on a second ultrasonic sensor (35), which is arranged on the traveling tool (80) in such a way that the detection range of the second ultrasonic sensor (35) is in the traveling direction of the traveling tool (80) or opposite to the traveling direction.

9. A device for supporting camera-based surroundings recognition of a moving tool (80) by means of road wetness information from a first ultrasonic sensor (30), the device comprising an evaluation unit (10) and a data input (12), wherein: The analysis and processing unit (10) is configured to: detecting, in conjunction with the data input (12), a first signal representing the surroundings (60) of the traveling tool (80) acquired by means of a first ultrasonic sensor (30) of the traveling tool (80); detecting, in conjunction with the data input, a second signal representing the surroundings (60) of the traveling tool (80) acquired by means of a camera (40) of the traveling tool (80); road wetness information is ascertained based on the first signal, wherein the road wetness information is ascertained by comparing a speed-dependent noise level (70) of the first signal with a speed-dependent predefined threshold value (75) of the noise level (70), wherein an estimation of the degree of wetness is also performed if the noise level of the first signal exceeds the predefined threshold value, wherein the road wetness information ascertained from the first signal is subjected to a plausibility check using the road wetness information ascertained from the second signal; selecting a predefined parameter group from a plurality of predefined parameter groups according to the road moisture information, wherein a predefined parameter group matching a corresponding moisture level is selected according to the moisture level, the plurality of predefined parameter groups representing different configurations of a classifier, the classifier being configured to analyze a surrounding environment of the traveling tool and identify an object in the surrounding environment based on the second signal; A surrounding environment recognition is performed based on the second signal in combination with the predefined parameter set.

Images