Method for detecting a vehicle washing system and method for operating a vehicle

Abstract

Method for detecting a vehicle washing system (FW), wherein - a vehicle's surroundings are detected using a vehicle's lidar sensor system, - for the detection and recognition of environmental objects, a location and an intensity of points (P1 to Pn) in point clouds (W1 to Wm) recorded by means of the lidar sensor system are determined, a direction of movement of the points (P1 to Pn) in the vehicle environment and relative to the vehicle is taken into account and a segmentation and classification of the point clouds (W1 to Wm) is carried out on the basis of this data, characterized in that - the vehicle wash system (FW) is recognized if at least three of the following requirements are met: - first requirement: an entrance (E) of the car wash (FW) is detected by a roller door (RT) and a roller door movement (RB) in the point clouds (W1 to Wm); - second requirement: static elements characteristic of the vehicle wash (FW) are detected in the point clouds (W1 to Wm); - third requirement: a tunnel-like structure with side walls, a floor (B) and a ceiling is detected in the point clouds (W1 to Wm); - fourth requirement: a lane with a transport device for the vehicle is detected in the point clouds (W1 to Wm); - fifth requirement: water droplets sprayed by surrounding objects are detected, whereby sprayed water droplets are distinguished from points (P1 to Pn) of static surrounding objects.

Description

[0001] The invention relates to a method for detecting a vehicle washing system.

[0002] The invention further relates to a method for operating a vehicle.

[0003] DE 10 2016 009 552 A1 discloses a method for preparing a vehicle for a car wash in a car wash. Upon entering the car wash, a car wash mode comprising preparatory and post-processing procedures is automatically activated in the vehicle. This mode is deactivated upon exiting the car wash. Entry and exit into and out of the car wash are detected by the vehicle's environment detection system, which is performed using a front camera in the vehicle.

[0004] Furthermore, DE 10 2020 102 113 A1 discloses a method for assisted alignment of a vehicle into a moving transport device of a car wash. The method comprises detecting a position and orientation of the vehicle relative to the transport device using lidar sensors and generating assistance information for aligning the vehicle into the moving transport device based on the detected position and orientation of the vehicle.

[0005] DE 10 2020 007 654 A1 describes a camera-based car wash detection system, where camera images are evaluated using an artificial neural network. Since a car wash is a tunnel-shaped structure, artificial intelligence detects an entry into a tunnel. Therefore, it is only concluded that a car wash has been entered if a tunnel is detected and a car wash-typical parameter is also met.The car wash-typical parameter is the connection state of a power transmission between a drive motor and the vehicle's wheels, the presence of wetness on a vehicle's windshield, the detection of a yaw rate profile characteristic of driving through a car wash, the detection of a speed, an acceleration, and / or a jolt characteristic of driving through a car wash, and / or a location corresponding to a car wash or a car wash portal. Additionally or alternatively, the detection of wetness on the vehicle's windshield with a rain sensor can be used to infer that a car wash has been entered.

[0006] DE 10 2020 000 764 A1 discloses a method for operating a vehicle with environmental sensors, comprising a front camera and at least one lidar sensor aimed at an area in front of the vehicle. Based on signals emitted by the lidar sensor and detected reflected signals, a three-dimensional image of the environment ahead of the vehicle is created. Within tunnels and buildings, objects protruding into the path of travel are detected, thus reducing any existing collision risk.

[0007] US 2021 / 0 188 321 A1 discloses a system for autonomously guiding a vehicle into a transport car wash. A LiDAR sensor mounted on the vehicle detects objects in the vehicle's surroundings and determines the distances, angles, and speeds of detected objects. In particular, the LiDAR sensor detects a conveyor structure of an automatic conveyor belt used to control and move vehicles in the car wash.

[0008] The invention is based on the object of providing a novel method for detecting a vehicle washing system and a novel method for operating a vehicle.

[0009] The object is achieved according to the invention by a method for detecting a vehicle washing system, which has the features specified in claim 1, and by a method for operating a vehicle, which has the features specified in claim 7.

[0010] Advantageous embodiments of the invention are the subject of the subclaims.

[0011] In the method for detecting a car wash system, the vehicle's surroundings are detected using a vehicle's lidar sensor system. To detect and recognize surrounding objects, the location and intensity of points in point clouds captured by the lidar sensor system are determined, and the direction of movement of the points in the vehicle's surroundings and relative to the vehicle is taken into account. Segmentation and classification of the point clouds are performed based on this data.

[0012] According to the invention, a vehicle washing system is recognized if at least three of the following requirements are met: - first requirement: an entrance to the car wash is detected by a roller shutter and a roller shutter movement in the point clouds; - second requirement: static elements and elements characteristic of a car wash system are detected in the point clouds; - third requirement: a tunnel-like structure with side walls, a floor and a ceiling is detected in the point clouds; - fourth requirement: a lane with a transport device for a vehicle is detected in the point clouds; - fifth requirement: water droplets sprayed by surrounding objects are detected, whereby sprayed water droplets are distinguished from points of static surrounding objects.

[0013] The lidar sensors used comprise one or more lidar sensors.

[0014] The method enables particularly reliable detection of car washes, also known as car wash tunnels. By using data acquired by lidar sensors, reliable depth information can be determined, which significantly increases the reliability of detection. In particular, the three-dimensional detection of the vehicle's surroundings reduces the susceptibility to errors compared to camera-based systems. The lidar sensors provide reliable depth information, so that a car wash can be plausibly verified more reliably than with camera-based systems. For example, the point clouds of the lidar sensors enable the detection of wash brushes and their carriers. Vehicles driving ahead in the car wash can also be detected in an object detection and classification process using data acquired by the lidar sensors.The reliability of detection is particularly enhanced by detecting roller shutter movement in three-dimensional space, detecting humidity and spray mist in the room, detecting water sprayed from nozzles, detecting water on the floor, and potentially redundancy and supplementation of data acquired by other sensors. The use of lidar sensors also offers the advantage of detecting a car wash from a greater distance, even when the vehicle is not directly in front of the car wash.

[0015] In the method for operating a vehicle according to the invention, a vehicle wash is automatically detected. Before the vehicle enters the vehicle wash, a vehicle wash mode that includes preparatory and follow-up processes on the vehicle is automatically activated. After the vehicle exits the vehicle wash, the vehicle wash mode is automatically deactivated.

[0016] Car washes can cause damage to vehicles, for example, to exterior mirrors, protruding cameras, or through open vehicle openings such as windows or sunroofs. The automatic and reliable detection of car washes using data acquired by lidar sensors enables automated activation of the car wash mode, so that countermeasures can be initiated in a timely manner before entering the car wash to prevent damage to the vehicle.

[0017] Embodiments of the invention are explained in more detail below with reference to a drawing.

[0018] It shows: Fig. 1 schematically shows an image of an entrance to a car wash, a point cloud distribution of a car wash recorded by a lidar sensor and an image of a car wash segmented based on the point cloud distribution.

[0019] In the only Fig. 1 shows an image BI of an entrance E of a vehicle wash system FW, a point cloud distribution V of the vehicle wash system FW recorded by a lidar sensor system with several point clouds W1 to Wm, and the image BI of the vehicle wash system FW, which is segmented based on the point cloud distribution V. The lidar sensor system comprises at least one lidar sensor and is a component of a vehicle.

[0020] To detect a car wash (FW), the vehicle's surroundings are recorded using the vehicle's lidar sensors. To detect and recognize objects in the vehicle's surroundings, the location and intensity of points P1 to Pn in point clouds W1 to Wm recorded by the lidar sensors are determined. Furthermore, the direction of movement of points P1 to Pn in the vehicle's surroundings and relative to the vehicle is determined and taken into account. Based on this data, the point clouds W1 to Wm are segmented and classified.

[0021] The lidar sensors can be used to detect various features in the vehicle environment, whereby a car wash FW is detected in the point clouds W1 to Wm if at least three of the five requirements listed below are met.

[0022] To fulfill a first requirement, it is checked whether an entrance E of the car wash system FW is detected. This detection is performed using a roller shutter RT and a roller shutter movement RB in the point clouds W1 to Wm. Corresponding point clouds W1 to Wm can be clearly identified as static via scans and / or tracking of static environmental objects from the point cloud distribution V. The roller shutter RT is extracted as a dynamic environmental object from the point cloud distribution V, and the roller shutter movement RB in the vertical direction is tracked via a vector v.

[0023] To fulfill a second requirement, it is checked whether static elements characteristic of a car wash FW are detected in the point clouds W1 to Wm. For example, static elements such as columns S1, S2, beams, wash brushes, and other elements are extracted from the point cloud distribution V, clustered, and classified as such based on their shape. For example, an element with a cylindrical shape and a "bristled" surface can be classified as a wash brush.

[0024] To fulfill a third requirement, it is checked whether a tunnel-like structure with side walls, a floor B, and a ceiling is detected in the point clouds W1 to Wm. This means that walls, ceilings, and floors B can be extracted from the point clouds W1 to Wm and classified as such. A car wash FW can then be detected and classified as a driveable space by means of boundaries formed by a detected ceiling, walls, and a floor B. Thus, a space detected in the point clouds W1 to Wm behind the entrance E of the car wash FW can be classified as a car wash space.

[0025] To fulfill a fourth requirement, it is checked whether a lane with a transport device for a vehicle is detected in the point clouds W1 to Wm.

[0026] To fulfill a fifth requirement, it is checked whether water droplets sprayed by surrounding objects are detected in the point clouds W1 to Wm. Sprayed water droplets are differentiated from points P1 to Pn of static surrounding objects. This involves segmenting the point clouds W1 to Wm and classifying sprayed water droplets, also known as spray points. Sprayed water droplets can be identified in the lidar based on their detection properties. The segmentation and classification enable the plausibility of the presence of sprayed water droplets in a defined space or specific area. This allows a distinction to be made between spray mist and spraying surrounding objects.

[0027] Additionally, an exit from the car wash system FW can also be detected. This is especially the case if another roller shutter and its associated roller shutter movement are detected in the point clouds W1 to Wm.

[0028] Furthermore, an environment located outside the vehicle wash FW behind the exit can be detected in the point clouds W1 to Wm if the exit of the vehicle wash FW was previously detected and at least three requirements are no longer met.

[0029] In order to detect the vehicle washing system FW, a gear position or gear position of a vehicle transmission can also be taken into account.

[0030] To identify the vehicle washing system FW, image data captured by a camera sensor can also be evaluated with regard to the first to fifth requirements.

[0031] The automatic detection of the vehicle wash (FW) can be used to automatically activate a vehicle wash mode that includes preparatory and follow-up processes on the vehicle when a vehicle wash (FW) is detected before the vehicle enters it. After the vehicle exits the vehicle wash (FW), the vehicle wash mode can be automatically deactivated.

[0032] When the vehicle wash mode is activated, for example, automatically opened vehicle openings that can be closed using locking elements, such as window openings and / or sunroof openings, are closed. Vehicle components protruding from the exterior of the vehicle, such as extendable cameras, exterior mirrors, antennas, etc., can also be automatically moved to a protected position, in particular retracted and / or folded in and / or folded in.

[0033] In one possible embodiment, a position of the closure elements and / or protruding vehicle components existing before the execution of the vehicle wash mode is stored so that, after the deactivation of the vehicle wash mode, they can be automatically moved back to the position existing before the activation of the vehicle wash mode.

Claims

[1] Method for detecting a vehicle washing system (FW), wherein - a vehicle's surroundings are detected using a vehicle's lidar sensor system, - for the detection and recognition of environmental objects, a location and intensity of points (P1 to Pn) in point clouds (W1 to Wm) recorded by the lidar sensor system are determined, a direction of movement of the points (P1 to Pn) in the vehicle environment and relative to the vehicle is taken into account and a segmentation and classification of the point clouds (W1 to Wm) is carried out on the basis of this data, characterized by , that - the vehicle wash system (FW) is recognized if at least three of the following requirements are met: - first requirement: an entrance (E) of the car wash (FW) is detected by a roller door (RT) and a roller door movement (RB) in the point clouds (W1 to Wm); - second requirement: static elements characteristic of the vehicle wash (FW) are detected in the point clouds (W1 to Wm); - third requirement: a tunnel-like structure with side walls, a floor (B) and a ceiling is detected in the point clouds (W1 to Wm); - fourth requirement: a lane with a transport device for the vehicle is detected in the point clouds (W1 to Wm); - fifth requirement: water droplets sprayed by surrounding objects are detected, whereby sprayed water droplets are distinguished from points (P1 to Pn) of static surrounding objects. [2] Method according to claim 1, characterized by that a space detected in the point clouds (W1 to Wm) behind the entrance (E) of the vehicle wash (FW) is classified as a vehicle wash room. [3] Method according to claim 1 or 2, characterized bythat an exit of the vehicle wash system (FW) is detected when another roller shutter (RT) and a roller shutter movement (RB) are detected in the point clouds (W1 to Wm). [4] Method according to claim 3, characterized by that an environment located outside the vehicle wash (FW) behind the exit is detected in the point clouds (W1 to Wm) if the exit of the vehicle wash (FW) was previously detected and at least three requirements are no longer met. [5] Method according to one of the preceding claims, characterized by that in order to detect the vehicle washing system (FW), a gear position or gear position of a vehicle transmission is also taken into account. [6] Method according to one of the preceding claims, characterized by that in order to detect the vehicle washing system (FW), image data captured by a camera sensor system are additionally evaluated with regard to the first to fifth requirements. [7] Method for operating a vehicle, wherein - a vehicle washing system (FW) is automatically detected by means of a method according to one of the preceding claims, - before the vehicle enters the vehicle wash (FW), a vehicle wash mode comprising preparatory and follow-up processes on the vehicle is automatically activated and - after the vehicle has left the vehicle wash (FW), the vehicle wash mode is automatically deactivated. [8] Method according to claim 7, characterized by that when the vehicle wash mode is activated, vehicle openings that are automatically opened and can be closed using locking elements are closed and vehicle components that protrude from the outside of the vehicle are automatically moved into a protected position. [9] Method according to claim 8, characterized bythat the closure elements and / or the protruding vehicle components are automatically moved into a position they had before the vehicle wash mode was activated after the vehicle wash mode has been deactivated.

Images