DE102021207609A1 - Procedure for marking a movement path - Google Patents

Abstract

The present invention relates i.a. Method for identifying a movement path of a vehicle (1), in which sensor data including detection points are recorded using a sensor, with targets along the movement path being determined using the sensor data, location points for the respective position of the targets and the sensor being determined using the location points several polygons are determined, in particular continuously, so that a polygon mesh is created, the method comprising the following method steps:- identifying polygons which intersect with the ego path;- determining the edges which intersect with the ego path, for each polygon intersecting the ego path, and- defining the boundaries of the path of motion by taking all edges of the polygons intersecting the ego path that also intersect the ego path as the surface of the path of motion and all edges of the polygons intersecting with the Ego path intersecting polygons that do not intersect with the ego path en, are marked as borders.

Description

The present invention relates to a method for identifying a movement path of a means of transportation, in particular a vehicle, and a system for estimating the geometry of a movement path, which is based on the implementation of a method according to the invention.

Technological background

Modern means of transportation such as B. motor vehicles or motorcycles are increasingly equipped with driver assistance systems, which use sensor systems to detect the environment, recognize the traffic situation and support the driver, z. B. by a braking or steering intervention or by the output of a visual or acoustic warning. Radar sensors, lidar sensors, camera sensors or the like are regularly used as sensor systems for detecting the surroundings. From the sensor data determined by the sensors, conclusions can then be drawn about the environment, e.g. B. an object and / or environment classification can be made. Furthermore, the detection of the surroundings is almost indispensable in the field of (partially) autonomous driving. By processing the sensor and environment data, assistance functions can then be carried out, e.g. B. Accidents with other road users can be avoided or complicated driving maneuvers can be made easier by supporting or even completely taking over the driving task or vehicle guidance (partly / fully automated).

Environment detection using radar sensors is often used in modern means of transportation. This is based on the emission of bundled electromagnetic waves (radar waves or radar signals) and their reflection, e.g. B. by other road users, obstacles on the road or the edge of the road. The differentiation of obstacles or objects such. B. other road users, traffic signs, road markings, peripheral buildings, crash barriers and the like is of great importance to determine when z. B. a braking or steering intervention is to be initiated.

In the context of environment detection, information about the environment is usually obtained, which is mainly divided into two parts: static and dynamic environment. To construct the static environment only detections from non-moving objects are considered, while for the dynamic environment moving objects are considered. However, the interpretation of both environments is important so that driving maneuvers can be planned safely. The design of the static environment includes obtaining information about the driving path in front of the vehicle, i. i.e. where to determine the travel corridor of the ego vehicle for the coming seconds (e.g. 3, 5, 6, 10 seconds or the like) and for the given corridor, and whether there are any obstacles on the predicted travel corridor.

The present invention focuses primarily on the first task of static environmental interpretation, i. H. obtaining information about the route in front of the vehicle. So far, this task is mainly solved by hand-made algorithms, where developers write methods to determine which static detections from tactile sensors (or pixels from camera image) can be used for driving corridor estimation. Normally, such algorithms are scenario-dependent, with a special solution having to be developed for each scenario. On the other hand, a new trend is observed that relies on artificial intelligence to interpret the static scenarios. The sensory information is interpreted as an image, and the algorithms used primarily for image processing are used for scene interpretation of static environments. First, such algorithms are trained with images that have already been marked, which indicate, for example, where the estimate of the course of the road ahead of the car is. After training, these algorithms are used on another dataset to identify static environmental features they were trained for, e.g. B. crash barrier sections on the highway, grass edges, lane markings or curbs. A major disadvantage of these algorithms is that they work best on tagged data. This means that it must already be marked which features belong to which parts of the sensor image.

Often the annotation or labeling is done manually and requires human input to identify the appropriate features in the image. Human input is a limited and expensive resource, and often error-prone. However, the performance of the Al-based algorithms also depends directly on the amount and quality of the labeled data. Both of these factors increase the need for an automatic static environment labeling process. There is therefore a particular need to improve such identification methods. This is where the present invention comes in.

Printed state of the art

From the U.S. 6,751,547 B2 a method for accurately estimating the geometry of a vehicle's forward path is known, which is based on a two-clothoid road model. Here road data provided by a camera or a radar system is collected. A measurement transfer function of the two-clothoid road model is then calculated using the road data. The near and far range clothoid coefficients are estimated simultaneously and the forward path of the vehicle is estimated from the data provided by the two clothoid model.

the U.S. 6,718,259 B1 discloses a method for estimating the geometry of a vehicle's forward path using an adaptive Kalman filter bank and a two-clothoid road model. A number of Kalman filters are used here, with weighting factors being provided for each filter which indicate the probability that the upcoming road geometry will match the road model assumed in the filter. After each filter has been assigned a weighted value, the weighted value road models are fused using a fusion element so that a weighted road model can be output.

Furthermore, the DE 10 2016 107 705 A1 a method in which a path for an autonomously driving vehicle is planned using Delaunay triangulation, collecting all detections including static and moving objects. Then the triangulation method and looking for triangle edges (triangle segments) that lie on the road surface and meet a set of criteria, one of the criteria being that each edge is not a road boundary edge. After finding the possible edges or the possible paths, the vertices of these edges are then extracted, at which the path is planned. A function of the various path costs is then estimated, choosing the path with the lowest cost.

Other methods, which are used to estimate the geometry of a movement path of a means of locomotion, are from DE 10 2015 013 084 A1 and the DE 10 2013 201 796 A1 known.

Furthermore, the DE 10 2018 215 448 B3 a method for estimating the geometry of a movement path of a means of locomotion according to the preamble of claim 1.

Object of the present invention

Proceeding from the prior art, the object of the present invention is now to provide a method in which the identification of the movement path is improved in a simple and cost-effective manner.

solution of the task

The above object is achieved by the entire teaching of claim 1 and the independent claims. Expedient developments of the invention are claimed in the dependent claims.

The present invention discloses an automatic labeling method to label road boundaries left and right of a vehicle based on the inputs from automobile sensors such as cameras, radars, and radars. Information from ADAS sensors and/or ego movement of the vehicle is used. The output of the method or algorithm is a global map marker with the left and right road boundaries. No manual input is required, which means that the susceptibility to errors and the costs can be reduced. In addition, road boundary sections can be marked much faster and more precisely, i. H. the amount of marking can be increased additionally.

• - Identifizieren von Polygonen, die sich mit dem Ego-Pfad überschneiden;
• - Bestimmen der sich mit dem Ego-Pfad überschneidenden Kanten, für jedes sich mit dem Ego-Pfad überschneidende Polygon, sowie
• - Festlegen der Grenzen des Bewegungsweges indem alle Kanten der sich mit dem Ego-Pfad überschneidenden Polygone, die sich auch mit dem Ego-Pfad überschneiden, als Oberfläche des Bewegungsweges und alle Kanten der sich mit dem Ego-Pfad überschneidenden Polygone, die sich nicht mit dem Ego-Pfad überschneiden, als Grenzen beschriftet bzw. gekennzeichnet werden.

In the method according to the invention for detecting road boundaries or for marking or labeling the movement path of a means of transportation, in particular a vehicle, sensor data are collected by means of a sensor. This is done in that, in particular, static targets or objects along the movement path are determined using the sensor data and location points are defined for the respective position of the targets and the sensor. Furthermore, a polygon is determined based on the location points, z. B. by serving a definable number of location points as corner points of the polygon. In particular, the polygon can include location points on each side of the street. The polygon is then used to identify the movement path or to label the road boundaries. This marking can e.g. B. be implemented in the form of an (offline) algorithm, ie the inner road boundary of the road can be determined or calculated based on the sensor measurements using the invention. Because a number of polygons are also determined, in particular continuously, so that an entire polygon network is determined, the polygon network can be used to estimate the movement be used as a guide. According to the invention, the process comprises the process steps:

• - Identifying polygons that intersect with the ego path;
• - determining the edges intersecting the ego path, for each polygon intersecting the ego path, and
• - Determining the boundaries of the movement path by taking all edges of polygons intersecting the ego path that also intersect with the ego path as the surface of the movement path and all edges of polygons intersecting the ego path that do not intersect with intersect the ego path, be labeled as boundaries.

According to a preferred embodiment, a radar sensor is provided as the first sensor, which generates radar detections due to the emission of bundled electromagnetic waves (radar waves or radar signals) and their reflection, z. B. to other road users, obstacles on the road or the edge of the road. However, other sensors known from the prior art can also be provided as the first sensor, e.g. B. camera, ultrasonic or lidar sensors, whose detection points are fed to the algorithm. Furthermore, a camera sensor is preferably provided as the second sensor, by means of which road edges or road or lane markings can be detected as static objects. However, other known sensors can also be used here.

It has proven to be particularly advantageous if the polygon is a triangle, since this can be calculated extremely easily with little computing effort. In the same advantageous manner, an entire polygon network can be created using triangular formations. This can be done in a simple way using a Delaunay triangulation or Delaunay triangle formation, as already described in DE 10 2018 215 448 B3 described. In particular, the term Delaunay triangulation also includes other known derivatives of a Delaunay triangulation, such as. B. Constrained Delaunay triangulation.

Crash barriers, infrastructure, parked road users, road markings, roadside buildings and the like can expediently be provided as static objects.

Furthermore, the type of boundary detected can also be identified based on a classification, e.g. B. Roadside buildings made of concrete, metal crash barriers, grass edges, road markings or the like.

Preferably, the detected boundaries are defined as left and right boundaries, where the edges of the polygons intersecting the ego path that do not intersect with the ego path and are to the left of the ego path are the left boundary and the edges of the polygons intersecting the ego path that do not intersect the ego path and are to the right of the ego path are mapped as the right boundary.

According to an advantageous embodiment of the method according to the invention, all associated edges can then be classified or labeled as belonging to the same boundary or to the same movement path if an edge is assigned to a side.

Each of the connected edges can then advantageously be classified or labeled/identified as a boundary.

Conveniently, a DGPS sensor can be provided, which can be extended in such a way that the output of the proposed algorithm in combination with DGPS coordinates can be used directly to create high-resolution maps (HD maps) that can be used in various ADAS functions .

In addition or subordinately, the present invention also claims a method for identifying or labeling a movement path of a vehicle, in which sensor data comprising detection points are recorded by means of a sensor and the geometry of the movement path is estimated on the basis of the sensor data, with sensor signals from the sensor Grid or map of cells is created to capture static detections. Based on the sensor signal (e.g. based on the intensity or the signal strength or the transit time of the sensor signal), a distinction is then made between strong and weak detections and a histogram is created, which extends from a vehicle axis of the vehicle. The occurrence of detections in the cells of the grid is displayed.

Preferably, the boundaries can be (additionally) extracted from the histogram, the boundaries being based on the occurrence of detections in the cells. Here z. B. the peaks in the histogram, which indicate the increased occurrence of detections,
A radar grid, an RCS grid, a camera image, a lidar image or the like can expediently be provided as the grid. In addition, is another type of sensor signal representation associated with the prior art is also conceivable.

Furthermore, the present invention includes a system for estimating the geometry of the movement path of a means of transportation or a vehicle. For this purpose, the system has a sensor for generating sensor data, in particular a radar sensor for generating radar detection points, and a control device. The geometry of the movement path is then estimated on the basis of the sensor data or the detection points and the identifiers generated according to the invention, in that the control device is set up to further process the sensor data of the sensor and the identifiers accordingly. The control device can expediently be set up in such a way that it can access actuators of the vehicle in order to carry out corresponding driving and assistance functions in a supportive and/or (partly) autonomous manner.

The methods according to the invention can expediently be purely computer-implemented methods, with the term “computer-implemented method” in the sense of the invention describing a flow plan or procedure that is realized or carried out using a computer. The calculator, such as a computer, a computer network, a control unit (e.g. ECU or Electronic Control Unit or ADCU or Assisted & Automated Driving Control Unit) or a computing device therein or another programmable device known from the prior art process the corresponding data using programmable calculation rules. With regard to the method according to the invention and the system according to the invention, the essential properties z. B. be caused by a new program, new programs, an algorithm or the like, which is executed on the control unit.

character list

• 1 eine vereinfachte schematische Darstellung eines Fahrzeuges, welches auf Kennzeichnungen zurückgreift, die anhand des erfindungsgemäßen Verfahrens erzeugt wurden;
• 2 eine vereinfachte schematische Darstellung von Radardetektionen in einem Radargrid (Links) mit einem Ausschnitt (Rechts) daraus;
• 3 eine vereinfachte schematische Darstellung einer statischen Umgebung (Links) sowie der Kombination aus Lateralem Offset und Ego-Pfad (Rechts), sowie
• 4 eine vereinfachte schematische Darstellung von Radardetektionen in einem Koordinatensystem, wobei der Bewegungsweg anhand einer Triangulierung durchgeführt wurde.

In the following, the invention is explained in more detail by means of expedient exemplary embodiments. Show it:

• 1 a simplified schematic representation of a vehicle, which uses identifiers that were generated using the method according to the invention;
• 2 a simplified schematic representation of radar detections in a radar grid (left) with a section (right) thereof;
• 3 a simplified schematic representation of a static environment (left) and the combination of lateral offset and ego path (right), as well as
• 4 a simplified schematic representation of radar detections in a coordinate system, where the movement path was carried out using a triangulation.

Reference number 1 in 1 refers to a vehicle or ego vehicle with a control device 2 (ECU, Electronic Control Unit or ADCU, Assisted and Automated Driving Control Unit), through which sensor control, sensor data fusion, environment and/or object recognition, trajectory planning and/or vehicle control, in particular (partially) autonomously. To control the vehicle, the control device 2 can access various actuators (steering 3, motor 4, brake 5). Vehicle 1 also has various sensors for detecting the surroundings (lidar sensor 6, camera 7, radar sensor 8, ultrasonic sensors 9a-9d and a rearward-facing radar sensor 10). Advantageously, the sensor data can be used for environment and object detection, so that various assistance functions such. B. emergency brake assistant (EBA, Emergency Brake Assist), distance following control (ACC, Adaptive Cruise Control), lane keeping control or a lane departure warning system (LKA, Lane Keep Assist) or the like can be realized. Furthermore, the assistance functions can also be executed via the control device 2 or another control unit provided for this purpose. The trajectory is planned based on an estimate of the geometry of the movement path according to the method according to the invention.

The radar sensor 8 is designed in such a way that the system can distinguish between static and moving objects. This embodiment is well known and is not intended to form part of this invention. Information about the road geometry in front of the vehicle 1 can be derived from the static objects. However, such information can also be ambiguous: for example, there are scenarios where only corridors based on static information are good candidates for roadside estimation. In order to eliminate such ambiguities, the lane information from moving objects can also be used to estimate the roadside. The traces of the detected moving objects indicate the position history or movement history of the objects or road users in the traffic surrounding the vehicle 1 .

To implement the invention, a grid or radar grid is first created to capture static detections and accumulate them over time. Various can do this Types of accumulations are used, e.g. B. probabilistic accumulations such as Bayes grid or RCS-based accumulations. The latter has the advantage that the RCS value depends on the reflection parameters of the objects. Therefore, a distinction can be made between strong detections (usually from metal objects) or weak detections (from non-metal ones such as vegetation, concrete, etc.). After inserting the detections into the static grid, the lateral offset (edge) of the road must be taken. This can e.g. B. by a histogram, the z. B. extends up to 15 meters in the longitudinal direction and 50 meters laterally to the left and right or laterally in both directions from the sensor position. The histogram shows the occurrence of the occupied cells in the longitudinal direction. Then the first peaks of the histogram to the left and right of the sensor are detected. They designate the innermost limit position to the left and right of the vehicle axis, as in 2 shown.

In 2 shows side road edge extraction for a static grid used as static radar input detection. A histogram is created for the occupied cells up to X=15 m and the lateral edges on the left (road boundary left SGL) and right (road boundary right SGR) are extracted. The extraction is based on the first peaks of the histogram to the left and right of the vehicle axis FA. The extracted lateral edges on the left and right relate to the position of the vehicle. By combining the current lateral edges with the ego trajectory or the ego path EP of the vehicle, the road boundaries to the left and right of the vehicle can now be reconstructed (reconstructed road boundary on the left rSGL and reconstructed road boundary on the right rSGR), as in 3 shown.

For each time stamp (t ₀ , t ₁ , t ₂ . . . t _n ), the corresponding border can be identified on the left and right next to the ego position or the ego path EP in the global map. By connecting the points, the entire course of the movement path to the left and right can then be reconstructed. Depending on the type of road boundary to be reconstructed, this algorithm can be extended to recognize different types of boundaries. If only the road boundaries have to be recognized, which originate from the crash barriers, and are recorded or detected using radar or lidar sensors (which can detect such targets well), a dedicated static grid can be created that allows the Time to collect only strong reflections and reject or discard weaker ones. If grass edges or concrete buildings are to be detected, which are to be detected using radar or lidar sensors (but do not provide such strong detections), e.g. For example, a static grid can be created that only accumulates weak reflections over time and discards stronger ones.

In the event that a camera image is to be used as a sensor input, the image can be converted to a bird's eye view. After that, the lane markers can be extracted and then recombined with the ego-driven position or path. In terms of content, such a combination would then roughly correspond to a side boundary extraction using a radar grid or lattice. Instead of just one lateral edge, multiple lateral edges can be extracted and a multi-track map can be created. The extraction of lane markers is not the subject of this work and there are a lot of algorithms how to extract them. The invention here consists of the combination of the side marker position and the ego position as explained in the preceding paragraphs. We are approaching the goal of this invention, viz. H. the marking of the features of the road boundaries. This can easily be done in a second iteration, augmenting the recorded data with the estimated edges as discussed above. By selecting all detections up to a certain distance, e.g. B. 0.5 m far from the estimated road boundary, such detections can be flagged as road boundary detection. Furthermore, it can be identified to which side of the road boundary the respective detection belongs.

• - Identifizieren von Polygonen bzw. Dreiecken, die sich mit dem Ego-Pfad überschneiden;
• - Bestimmen der sich mit dem Ego-Pfad überschneidenden Kanten für jedes sich mit dem Ego-Pfad überschneidende Polygon bzw. Dreiecks, sowie
• - Festlegen der Grenzen des Bewegungsweges indem alle Kanten der sich mit dem Ego-Pfad überschneidenden Polygone, die sich auch mit dem Ego-Pfad überschneiden, als Oberfläche des Bewegungsweges und alle Kanten der sich mit dem Ego-Pfad überschneidenden Polygone, die sich nicht mit dem Ego-Pfad überschneiden, als Grenzen bzw. Straßengrenzen/Straßenränder klassifiziert werden.

Furthermore, there may be situations where the use of a grid is not advantageous. For example, the resolution of raster cells can be too coarse, which can result in blurred street boundaries. i.e. detection beyond the innermost road boundary may be misclassified as a road boundary. Also the principle of selecting all detections close to the road boundary up to a certain distance can contribute to this blurring (misclassification of detections). Therefore, labeling can also be performed that does not require a grid or histogram to identify the innermost boundaries. In this method according to the invention, the geometry of the movement path is estimated by determining or calculating the inner road boundary on the basis of sensor measurements. Exemplary embodiments are described below in which a radar sensor is used as the sensor. However, other sensors known from the prior art, such as e.g. B. lidar or camera sensors. In each scan cycle, the radar sensor supplies a list of objects or targets with the corresponding relative position of the respective ones Targets and their speed in relation to the vehicle that includes the radar sensor and corresponds to location 0 (x=0; y=0). Objects are then classified as static or mobile targets based on relative speed to the vehicle. In the present case, however, only the static targets to which a location point can be assigned are used for estimating the road boundary. The assessment then takes place by calculating or inserting a polygon network or a Delaunay triangulation (Delaunay triangle formation or Delaunay triangulation) on the basis of the position of the static targets or their location points and that of the vehicle. The creation of the polygon mesh or the Delaunay triangulation is carried out in particular by a method or a method configuration according to FIG DE 10 2018 215 448 B3 , to which direct reference is made. Here, the static detections from a tactile sensor can be used directly in order to build up the Delaunay triangulation described thereon. Then the ego path can be taken beyond this triangulation. The following procedural steps can also be taken:

• - Identifying polygons or triangles that intersect with the ego path;
• - determining the edges intersecting the ego path for each polygon or triangle intersecting the ego path, and
• - Determining the boundaries of the movement path by taking all edges of polygons intersecting the ego path that also intersect with the ego path as the surface of the movement path and all edges of polygons intersecting the ego path that do not intersect with intersect with the ego path are classified as boundaries or road boundaries/roadsides.

Here it can be recognized whether a (road) boundary is to the left or to the right by determining on which side of the ego path one of its segments or one of its edges lies. If a segment/edge is assigned as a side, all connected segments/edges can be classified on the same path/road, with each detection in this chain of segments/edges being classified as a (road) boundary.

In 4 is the intersection of the vehicle-driven path with the Delaunay triangulation Tri of the static detections from a tactile sensor, such as a B. Radar. Only the triangles that lead through the ego path EP are taken into account for the detection of the road boundary SG. In particular, the edges of such triangles that do not intersect the ego path EP form the innermost edge or the innermost border on the left and right.

The methods according to the invention are easy to implement or to retrofit in existing systems. Furthermore, they only require a reasonable amount of resources (RAM and processor) and most importantly, they can replace the human operator for road boundary marking. This method can thus be used to estimate road boundaries based on artificial intelligence, since they automatically output the marked road sections. Furthermore, the methods can also be used in the areas of trajectory planning in autonomous driving, robotics, geodesy or in the creation of HD maps, so that the present invention makes a very special contribution to the prior art.

Reference List

1

vehicle

2

control device

3

steering

4

engine

5

brake

6

lidar sensor

7

camera

8th

radar sensor

9a-9d

ultrasonic sensor

10

radar sensor

EP

ego path

SG

road border

SGL

street boundary on the left

SGR

street boundary on the right

rSGL

reconstructed road boundary on the left

rSGR

reconstructed road boundary on the right

FA

vehicle axle

tri

triangulation

QUOTES INCLUDED IN DESCRIPTION

This list of documents cited by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent Literature Cited

• US 6751547 B2 [0007]
• US 6718259 B1 [0008]
• DE 102016107705 A1 [0009]
• DE 102015013084 A1 [0010]
• DE 102013201796 A1 [0010]
• DE 102018215448 B3 [0011, 0017, 0035]

Claims (15)

Method for identifying a movement path of a vehicle (1), in which sensor data comprising detection points is recorded by means of a sensor, with targets along the movement path being determined using the sensor data, location points for the respective position of the targets and the sensor being determined, using the location points several Polygons are determined, in particular continuously, so that a polygon network is created, characterized in that the method comprises the following method steps: - identifying polygons which intersect with the ego path; - determining the ego-path-intersecting edges, for each ego-path-intersecting polygon, and - defining the boundaries of the movement path by including all edges of the ego-path-intersecting polygons that also intersect with the ego-path Intersect Path as the surface of the motion path and all edges of the polygons intersecting the ego path that do not intersect the ego path as boundaries. procedure after claim 1 , characterized in that at least one lidar sensor (6), a camera (7), a radar sensor (8) and/or an ultrasonic sensor (9a-9d) is provided as the sensor. procedure after claim 1 or 2 , characterized in that the polygon is a triangle or the polygons are triangles. procedure after claim 3 , characterized in that the polygon mesh is determined using a Delaunay triangulation. Method according to one of the preceding claims, characterized in that crash barriers, infrastructure, parked road users, road markings, roadside buildings and the like are provided as static objects. Method according to one of the preceding claims, characterized in that the recognized boundaries are marked on the basis of a classification, in particular the boundaries are marked as roadside development, crash barrier, grass edge, road marking or the like. Method according to one of the preceding claims, characterized in that the recognized boundaries are defined as left and right boundaries, the edges of the polygons intersecting with the ego path, which do not intersect with the ego path and are to the left of the ego path are assigned as the left boundary and the edges of the polygons intersecting the ego path that do not intersect with the ego path and are to the right of the ego path are assigned as the right boundary. procedure after claim 7 , characterized in that when an edge is assigned to a side, all associated edges can be classified on the same boundary or movement path. Method according to one of the preceding claims, characterized in that each of the connected edges is classified as a boundary. Method according to one of the preceding claims, characterized in that a GPS sensor, in particular a DGPS sensor, is provided. procedure after claim 10 , characterized in that a high-resolution map (HD maps) is created using the coordinates of the GPS sensor. Method for characterizing a movement path of a vehicle (1), in which sensor data comprising detection points is recorded by means of a sensor and the geometry of the movement path is estimated on the basis of the sensor data, in particular according to at least one of the preceding claims, wherein on the basis of sensor signals from the sensor a grid with cells is created in order to record static detections, a distinction being made between strong and weak detections based on a sensor signal and a histogram, which extends from a vehicle axis (FA) of the vehicle (1), is created, which shows the occurrence of detections in the cells of the grid. procedure after claim 12 , characterized in that the boundaries are extracted from the histogram, the boundaries being based on the occurrence of detections in the cells. procedure after claim 12 or 13 , characterized in that a radar grid, an RCS grid, a camera image, a lidar image or the like is provided as the grid. System for estimating the geometry of a movement path of a vehicle (1), with at least one sensor for generating detection points comprising sensor data, and a control device (2), the geometry of the movement path being estimated by the control device (2) using the sensor data of the sensor , in that the control device (2) carries out the estimation of the geometry of the movement path on the basis of markings which were generated by means of a method according to one of the preceding claims.

Images