WO2024022740A1

WO2024022740A1 - Creating a logical route network in parking areas

Info

Publication number: WO2024022740A1
Application number: PCT/EP2023/068100
Authority: WO
Inventors: Sebastian BUSSE; Daniel LAUBRICH; Jonas Jung; Stefan Wappler
Original assignee: Volkswagen Aktiengesellschaft
Priority date: 2022-07-26
Filing date: 2023-06-30
Publication date: 2024-02-01
Also published as: DE102022207651A1

Abstract

The invention relates to a method and to a device for determining a route network for a motor vehicle, comprising: providing at least two trajectories, each of which is indicative of a route traveled by a motor vehicle within its surroundings; and comprising waypoints, each waypoint having one dedicated point of the motor vehicle along the path of travel in the surroundings at a corresponding point in time, furthermore: determining corresponding trajectory sections from the trajectories; determining centroids of corresponding waypoints of corresponding trajectory sections, each centroid having a position and orientation within the surroundings; determining centroid clusters based on the determined centroids; determining nodes from the centroid clusters; determining further centroids based on centroid clusters oriented substantially parallel; determining node links based on the further centroids and the nodes such that one node is linearly connected to each of its next neighboring nodes; checking whether nodes exist where adjacent nodes can be reached via more than one node link; then determining node centroids of these nodes in question; replacing all node links between each node in question and each of the connected nodes which are not in question; deleting the nodes in question; and determining a route network in the surroundings based on the remaining nodes, the node centroids and the node links.

Description

Creating a logical network of paths in parking spaces

The present invention relates to the creation of digital maps for parking spaces from sensor data from motor vehicles.

US 10 012 991 B1 discloses a method and a device for controlling the operation of an autonomous vehicle. In one aspect, the autonomous vehicle can track the trajectories of other vehicles on a road. Based on the trajectories of the other vehicles, the autonomous vehicle can generate a pool of combined trajectories. The autonomous vehicle can then select one of the combined trajectories as a representative trajectory. The representative trajectory can be used to change the speed or direction of the autonomous vehicle.

US 2020 0217458 A1 discloses a device and a method that provide trajectory data for a geographical area. Multiple probe data sets are received or identified by one or more mobile devices. The probe data sets include time values in a first sequence that are associated with location values in the first sequence. At least one of the probing data sets is modified to reverse the location values so that the modified probing data set includes time values in a first sequence that are associated with location values in a second sequence. A location clustering algorithm is performed based on the plurality of test data sets and the modified test data set based on the location values.

It is therefore desirable to provide an improved card system.

This is done using a method and a device for a motor vehicle for determining a route network for a motor vehicle.

The method for determining a route network for a motor vehicle, preferably an autonomously driving motor vehicle, has the provision of at least two trajectories. Each trajectory is indicative of a path traveled by a motor vehicle within the surroundings of the motor vehicle. The trajectory has waypoints. Each waypoint of the trajectory has a dedicated point of the motor vehicle along the route in the area at a corresponding point in time. And the surroundings point preferably a park environment. The method further comprises: determining corresponding trajectory sections from the trajectories. Determining waypoint centroids of corresponding waypoints of corresponding trajectory sections. Each waypoint centroid has a position and orientation within the environment. Determine centroid clusters based on the identified waypoint centroids. Determining nodes from the focal clusters. Determine additional waypoint centroids based on essentially parallel oriented centroid clusters. Determining node connections based on the other waypoint centroids and the nodes, such that a node is linearly connected to its nearest node neighbors. Check whether nodes exist where it is possible to reach neighboring nodes via more than one node connection. If such nodes exist: Determine the node centroids of those nodes that can reach each other via more than one node connection. Replace all node connections between each affected node and the respective connected unaffected nodes with node connections between the corresponding node centroid and the unaffected connected nodes. Delete the affected nodes. Furthermore, the method comprises determining a route network in the environment based on the remaining nodes, the node centroids and the node connections.

A trajectory according to the present invention is a path, a path or a route that a motor vehicle travels. Thus, a trajectory can be a movement path of an object, represented by the temporal sequence of coordinates during runtime.

This has the advantage that a network of paths can be created and kept up to date in an efficient and automated manner, especially in parking spaces such as parking areas, multi-storey car parks and underground car parks.

According to a further aspect of the invention, a device for a motor vehicle is provided. The device for determining a route network for a motor vehicle, preferably an autonomously driving motor vehicle, has an interface that is set up to communicate control information with at least one specific vehicle component of a motor vehicle, and a control module. The control module is set up to provide at least two trajectories. Each trajectory is indicative of a path traveled by a motor vehicle within the surroundings of the motor vehicle. The trajectory has waypoints. Each waypoint of the trajectory has a dedicated point of the motor vehicle along the route in the area at a corresponding point in time. And the environment points preferably a park environment. Furthermore, the control module is set up to: determine corresponding trajectory sections from the trajectories; to determine waypoint centroids of corresponding waypoints of corresponding trajectory sections, each waypoint centroid having a position and orientation within the environment; Determine centroid clusters based on the identified waypoint centroids; to identify nodes from the focus clusters; identify additional waypoint centroids based on essentially parallel oriented centroid clusters; to determine node connections based on the other waypoint centroids and the nodes, such that a node is linearly connected to its nearest node neighbors; to check whether nodes exist where it is possible to reach neighboring nodes via more than one node connection; and if such nodes exist: to determine the node centroids of those nodes that can reach each other via more than one node connection; Replace all nodal connections between each affected node and the respective connected unaffected nodes with nodal connections between the corresponding nodal centroid and the unaffected connected nodes; and delete the affected nodes. Furthermore, the control module is set up to determine a network of paths in the area based on the remaining nodes, the node centroids and the node connections.

According to yet another aspect of the invention, a computer program product is provided with program code for carrying out a method according to the invention when the computer program product is executed on a computing unit according to the invention, a device according to the invention or a system according to the invention. This has the advantage that a possibility can be provided with which the map update or map creation and communication between the motor vehicle and a device can be carried out automatically.

According to a first embodiment, the determination of further waypoint centroids is additionally based on at least one node point of at least one of the essentially parallel oriented centroid clusters.

This has the advantage that the path network can be created even more efficiently. According to a further embodiment, determining node connections based on the further waypoint centroids and the node points involves replacing the node points that were used to determine the further waypoint centroids with the corresponding further waypoint centroids.

This has the advantage that the path network can be optimized or created more precisely.

According to a further embodiment, the focus clusters are indicative of sections of a path network in the area.

This has the advantage that the path network can be created even more efficiently.

According to a further embodiment, the nodes are indicative of starting points, end points and crossing points of the network of paths in the area.

According to a further embodiment, a focal cluster only has two nodes. This has the advantage that the path network can be optimized or created more precisely.

According to a further embodiment, the method further comprises dividing the determined road network in the area into a first level, which is indicative of a road network in the area, and into a second level, which is indicative of lanes in the area.

This has the advantage that the created path network is easier to process.

According to a further embodiment, the method further comprises providing the road network to a motor vehicle.

This can be the motor vehicle that has provided the at least two trajectories. However, the motor vehicle to which the road network is provided can also be another motor vehicle.

This has the advantage that the route network provided can be distributed to many motor vehicles, and therefore broadly.

Various aspects of the present disclosure relate to an apparatus comprising one or more processors and one or more storage devices configured to perform the previously presented method.

Various aspects of the present disclosure relate to a program having program code for performing the previously presented method when the program code a computer, a processor, a control module or a programmable hardware component.

The features of the method, the corresponding device, a corresponding computer program and the vehicle in relation to the method are described below. Features that are described in connection with the method can also be incorporated into the corresponding device, the corresponding computer program and the corresponding vehicle.

Various aspects of the present disclosure relate to a computer-implemented method for controlling one or more settings of a vehicle, as well as a corresponding device and a corresponding computer program. The method, the device and the computer program can be used to adjust (that is, change) the one or more settings based on the following criteria. Not all vehicle settings are equally suitable for this concept. In particular, the present concept is suitable for one or more settings that can be changed by the one or more occupants of the vehicle via a user interface (of the vehicle), that is, settings that can also be changed by the users themselves without this Changes to the vehicle's hardware or software would be necessary. A possible limitation to such settings also has the advantage that the actions of the occupants, i.e. the changes in the settings by the occupants, can be recorded and used to train the machine learning model. Accordingly, the method may include recording occupants changing settings, along with corresponding environmental indicators. For example, the one or more settings of the vehicle may relate to functions of an interior of the vehicle, such as comfort functions of the vehicle. These can be, for example, one or more settings of the vehicle, at least one of an air conditioning setting of the vehicle, a seat setting of the vehicle, a lighting setting of the vehicle, a windshield wiper setting of the vehicle. But the settings can also include a vehicle suspension setting, such as switching between a comfort and a sport suspension setting. Other settings are also conceivable.

Some examples will now be described in more detail with reference to the accompanying figures. However, other possible examples are not limited to the features of these embodiments described in detail. These may include modifications to the features as well as equivalents and alternatives to the features. Furthermore, the Terminology used herein to describe specific examples is not intended to limit other possible examples.

The same or similar reference numerals throughout the description of the figures refer to the same or similar elements or features, which can each be implemented identically or in a modified form while providing the same or a similar function. Furthermore, in the figures, the thicknesses of lines, layers and/or areas may be exaggerated for clarity.

When two elements A and B are combined using an “or”, it is to be understood that all possible combinations are disclosed, that is, only A, only B, and A and B, unless expressly defined otherwise in each individual case. As an alternative formulation for the same combinations, “at least one of A and B” or “A and/or B” can be used. This applies equivalently to combinations of more than two elements.

If a singular form, e.g. For example, if "a, an" and "the, the, the" are used and the use of only a single element is neither explicitly nor implicitly defined as mandatory, further examples can also use multiple elements to implement the same function. Where a function is described below as being implemented using multiple elements, further examples may implement the same function using a single element or processing entity. It is further understood that the terms “comprises,” “comprising,” “having,” and/or “having,” when used, mean the presence of the specified features, integers, steps, operations, processes, elements, components, and/or group describe the same, but do not exclude the presence or addition of one or more other features, integers, steps, operations, processes, elements, components and/or a group thereof.

Some examples of devices and/or methods are explained in more detail below with reference to the accompanying figures, merely by way of example. Show it:

1 shows a flowchart of an example of a method according to the invention for determining a route network for a motor vehicle;

2 shows a flowchart of a further example according to the invention of a method for determining a route network for a motor vehicle; 3 shows a flowchart of a further example of a method according to the invention for determining a route network for a motor vehicle;

4 shows a block diagram of a further example according to the invention of a device for determining a route network for a motor vehicle;

5 shows a graph of an example according to the invention for determining a route network for a motor vehicle;

6 to 6c show further graphs of an example according to the invention for determining a route network for a motor vehicle;

7 to 7a show further graphs of an example according to the invention for determining a route network for a motor vehicle;

8 to 8a show further graphs of an example according to the invention for determining a route network for a motor vehicle; and

9 to 9a show further graphs of an example according to the invention for determining a route network for a motor vehicle.

Fig. 1 shows a flowchart of an example of a method according to the invention for determining a route network for a motor vehicle. The method for determining a route network for a motor vehicle 100, preferably an autonomously driving motor vehicle, has the provision 10 of at least two trajectories 110. Each trajectory 110 is indicative of a path traveled by a motor vehicle 100 within an environment 200 of the motor vehicle 100. The trajectory 110 has waypoints 111 - also called waypoints.

Each waypoint 111 of the trajectory 110 has a dedicated point of the motor vehicle 100 along the route in the environment 200 at a corresponding point in time. And the environment 200 preferably has a parking environment. The method further comprises: determining 20 corresponding trajectory sections 112 from the trajectories 110; Determining 30 waypoint centroids 113 - also called WaypointCentroids - corresponding waypoints 111 of corresponding trajectory sections 112, each waypoint centroid 113 having a position and orientation within the environment 200; Determine 40 centroid clusters 114 - also called Centroid Cluster - based on the determined waypoint centroids 113; Determine 50 nodes 115 - also called intersection clusters - from the focus clusters 114; Determine 60 further waypoint centroids 116, based on essentially parallel oriented centroid clusters 114; Determining 70 node connections 117, based on the further waypoint centroids 116 and the nodes 115, such that a node 115 is linearly connected to its nearest node neighbors; Check 80 whether nodes 115 exist where it is possible to reach neighboring nodes via more than one node connection 117; and if such nodes exist: determining 82 node centers 118 - also called T-junctions - of those nodes 115 that can reach each other via more than one node connection 117; replacing 84 all node connections 117 between each affected node 115 and the respective connected unaffected nodes 115, with node connections 117 between the corresponding node centroid 118 and the unaffected connected nodes 115; and deleting 86 the affected nodes 115.

Furthermore, the method comprises determining 90 a path network 210 of the environment 200, based on the remaining nodes 115, the node centroids 118 and the node connections 117.

Fig. 2 shows a flowchart of a further example according to the invention of a method for determining a route network for a motor vehicle. 1 furthermore has that the determination 70 of node connections 117, based on the further waypoint centroids 116 and the nodes 115, involves replacing 72 of the nodes 115, which were used to determine 60 the further waypoint centroids 116, by the corresponding further waypoint focal points 116.

3 shows a flowchart of a further example according to the invention of a method for determining a route network for a motor vehicle. The method shown in FIG is indicative, for lanes around 200.

Fig. 4 shows a block diagram of a further example according to the invention of a device for determining a route network for a motor vehicle. The device 300 for determining a road network 210 for a motor vehicle 200, preferably an autonomously driving motor vehicle, has: an interface 310, which is set up to communicate control information with at least one specific vehicle component 110 of a motor vehicle 100; and a control module 320 configured to: provide at least two trajectories 110, each trajectory 110 being indicative of a path traveled by a motor vehicle 100 within an environment 200 of the motor vehicle 100; the trajectory 110 has waypoints 111; each waypoint 111 of the trajectory 110 has a dedicated point of the motor vehicle 100 along the route in the environment 200 at a corresponding time; and the environment 200 preferably includes a parking environment. Furthermore, the control module 320 is set up to determine corresponding trajectory sections 112 from the trajectories 110; to determine waypoint centroids 113 of corresponding waypoints of corresponding trajectory sections 112, each waypoint centroid 113 having a position and orientation within the environment 200; to determine centroid clusters 114 based on the determined waypoint centroids 113; to determine nodes 115 from the focus clusters 114; to determine further waypoint centroids 116 based on essentially parallel oriented centroid clusters 114;

to determine node connections 117, based on the further waypoint centroids 116 and the nodes 115, such that a node 115 is linearly connected to its nearest node neighbors; to check whether nodes 115 exist where it is possible to reach neighboring nodes 115 via more than one node connection 117; and if such nodes exist: to determine node centroids 118 of those nodes 115 that can reach each other via more than one node connection 117; to replace all node connections 117 between each affected node (115 and the respectively connected, unaffected nodes 115 by node connections 117 between the corresponding node centroid 118 and the unaffected connected nodes 115; and to delete the affected nodes 115. And furthermore, the control module 320 is there for this set up to determine a path network 210 of the environment 200, based on the remaining nodes 150, the node centers 118 and the node connections 117.

Fig. 5 shows a graph of an example according to the invention for determining a route network for a motor vehicle. The road network consists of several objects that are related to one another, and through these relationships a vehicle 100 can plan a route through the map. The road network consists of three layers, two logical graphs and the geometric information. The two graphs are the links/linkjnintersections and the lanes/lane_intersections, where the connections represent the higher logical level. The geometric information is represented by the shaping lines (shapepointjines). The relationships between these layers are as follows (according to MTC Json format):

- each connection is connected to 1-N lanes - each lane is connected to exactly one link and one shapepoint jine

- each shapepoint jine is linked to exactly one lane

- each connection/lane has exactly two intersections, one at the beginning and one at the end

- each intersection is connected to 1-N connections/tracks and can be the start or end of several connections/tracks.

These relationships must also be created when creating the road network or the path network 210.

6 to 6c show further graphs of an example according to the invention for determining a path network 210 for a motor vehicle 100. The graphs are created by merging and clustering poses of one or more tracks. The original trajectories 110 and the result of the clustering are shown in Figures 6 to 7a.

The result of the waypoint and centroid clustering shown in Fig. 5 is used to create the road network objects. Each centroid cluster 114 - also called CentroidCluster - consists of a path of waypoint centroids 113 - also called WaypointCentroids. In the following figure, each CentroidCluster is represented by a different line style. They show the geometric information of the merged trajectories 110 and function as shapepoint lines “shapepointjines” of the road networks. The lanes represent the shapepointjines in a logical graph and only contain the start and end points of each Centroid Cluster.

In Fig. 6: A node structure consisting of nodes 115 - also called intersection cluster - is formed. To obtain the link objects for the first level logical graph 212 - also called the top level - the node structure - also called the IntersectionCluster structure - can be added, which also contains WaypointCentroids 113. The initial intersection clusters 115 are the start and end points of the Centroid Cluster 114.

In Fig. 6a: These intersection clusters 115 are now expanded to contain further WaypointCentroids 113, which are close enough to the original center of gravity - also called centroid - and have a parallel orientation, thus in the same or the opposite direction. If different intersection clusters 115 overlap, they can be merged into a single cluster. All WaypointCentroids 113 can also know whether they are part of an intersection cluster 115.

In Fig. 6b: The centers of these intersection clusters 115 later function as connection intersections. Now a new structure called intersection connections 117 - also called IntersectionClusterConnection - can be added. These objects may contain references to the CentroidClusters 114 involved. They may also contain information about which WaypointCentroid 113 from which cluster is in the connection, references to two intersection clusters 115 and a flag indicating whether the connection is bidirectional.

To create these IntersectionClusterConnections 117, all CentroidClusters 114 can be iterated over. Each CentroidCluster 114 begins and ends with an Intersection Cluster 115. When iterating over the centroids of a cluster, one “walks” from one Intersection Cluster 115 to another and the Intersection Cluster Connection objects 117 are filled iteratively. Only the waypoint centroids 113 that are part of an intersection cluster 115 are processed. This is shown by “circles” in Fig. 6b.

This leads directly to the desired connections.

In Fig 6c: In the last step, the road network for the intersections can be simplified. 6 shows that the node centroids 118 - also called T-junctions - have three Intersection Clusters 115, all of which are connected to one another. This can be simplified so that there is only one intersection cluster 115 in the middle of the intersection that has connections to the different directions a vehicle can take. Since the T-junction 118 in FIG. 6c is a very simple one, another example is shown in FIG. 7 as an intersection.

7 to 7a show further graphs of an example according to the invention for determining a route network for a motor vehicle.

An intersection with four intersection clusters 115 is shown, all of which are connected to one another. As in the example of FIG. 6c, there should only be one intersection cluster 115 with four connections to the different directions. In terms of graph theory, the nodes and edges of the intersection form a maximal cluster structure that should be shrunk to a node. Unfortunately, finding maximal cluster structures in a graph is an NP-complete problem, so it is not possible to shrink the entire cluster structure at once in polynomial time. However, it is possible to shrink relevant IntersectionClusterConnections 117 one by one.

This makes it possible to find out which connections need to be shrunk to form an intersection cluster 115 by iterating over all connections in the graph. If two of them have a common neighbor, they are part of a triangle and any connection that is part of such a triangle can then be shrunk. Care should be taken when changing IntersectionClusterConnections 117 to Intersectionclusters 115 be shrunk or combined. The two intersection clusters 115 belonging to the connection should be merged. In particular, the WaypointCentroids 113 that point to the Intersectionclusters 115 should be adjusted to point to the new one (which should also have references back to the Centroids 114). Since the IntersectionClusterConnection 117 to be collapsed is part of one or more triangles, there are also connections that should be merged. To avoid information loss, the newly created Intersectioncluster 115 representing the shrunken IntersectionClusterConnection 117 should receive references to the CentroidClusters 114 previously stored by the connection. Even if this intersection cluster 115 is later merged with another when another connection is shrunk, care should be taken to ensure that this information is still available. When shrinking each connection that belongs to a triangle, care should be taken to adjust each reference in the entire ClusterGraph accordingly to obtain the desired result.

The last thing that should now be done is to convert the Intersectionclusters 115 and IntersectionClusterConnections 117 with the included WaypointCentroids 113 and CentroidClusters 114 into links, lanes and shapepoint lines, thus into a path network. The first and easiest part is creating the connections and their intersections. The intersection clusters 115 can be translated directly into link intersections. A link can be created between these link intersections if there is an IntersectionClusterConnection 117 between two of these intersection clusters 115. For the lanes and the shapepoint lines, the CentroidClusters 114 stored within the IntersectionClusterConnections 117 and in the Intersectionclusters 115 should be taken into account (since some of the connections have been reduced to Intersectionclusters 115 and the corresponding CentroidClusters 114 have been referenced therein). It should be noted that an IntersectionCluster(Connection) 114 or 117 can only store part of the WaypointCentroids 113 of a CentroidCluster 114.

This means that a CentroidCluster 114 can be divided into different Shapepoint Lines/Lanes. Using the WaypointCentroids 113 of a CentroidCluster 114, to which an intersection cluster 115 or a connection refers, a lane intersection can be created from its first and last WaypointCentroids 113 and the respective lane or connection in between. A shapepoint line can also be created from these WaypointCentroids 113 and relationships can be added between these newly created lanes and shapepoint lines. Preferably, the relationships between the links and the lanes are created.

If a created lane was constructed from the WaypointCentroids 113 referenced by an IntersectionClusterConnection 117, a relationship can easily be added to the connection created from the same object. If the lane was created from the WaypointCentroids 113 referenced by an intersection cluster 115, a lane is created within an intersection that describes the transition from one link to another.

There is a convention that this lane should connect to the link going into the intersection. For this reason, the first WaypointCentroid, 113, which belongs to the CentroidCluster 114 and is referenced by the Intersectioncluster 115, can be examined. If this first WaypointCentroid 113 also points to an IntersectionClusterConnection 117, a relationship can be added between the created lane and the connection created by this IntersectionClusterConnection 117. If the first WaypointCentroid 113 does not reference an IntersectionClusterConnection 117, it can be checked to see if it has a predecessor that can be checked for a reference to an IntersectionClusterConnections 117. This continues until a (perhaps previous) predecessor is found that points to an IntersectionClusterConnection 117 to establish the relationship as described previously.

8 to 8a show further graphs of an example according to the invention for determining a route network for a motor vehicle. The previous steps already lead to a completely functional graph. However, some improvements can still be made so that the cluster graph (and the links, lanes and shapepoint lines generated from it) reflects reality even better.

The following two steps can be applied:

Removing short dead end connections. Merging (partially) parallel end connections

Furthermore, adjustments can be made to merging trajectories when merging waypoints and clustering merged waypoints.

Adjustments: ignore trace IDs

Remove neighborhood diamonds

A “neighborhood diamond” means that two centroids have both a common predecessor and a common successor. In Fig. 8:

The diamond can be removed by merging the two centers of gravity - also called centroids - into one.

Remove neighborhood triangles:

In Fig. 8a: A centroid z has a common successor m with one of its successors n. An edge should be removed. l.d.R. has n other predecessors or successors besides z and m. If there are other predecessors, remove the edge between z and n. If there are other successors, remove the edge between n and m.

Remove bidirectional connections:

A centroid has another centroid as both a predecessor and a successor. Use alignment values stored in centroids and calculate their average. Remove the edge direction that faces the opposite direction to the alignment.

9 to 9a show further graphs of an example according to the invention for determining a route network for a motor vehicle. They show further configurations that can further optimize the process.

In Fig. 9: If there are IntersectionClusterConnections 117 that are very short (e.g. < 10m) and lead to (or come from) a "dead end", it is very likely that you have driven into a parking lot. These connections should not be present in the ClusterGraph and the Roadnet objects obtained from it. To do this, the length of these connections can be calculated, and if this length is below a certain threshold, this connection is deleted. This can leave redundant intersection clusters 115 that do not provide any additional information for the graph, as shown in the middle representations of FIG. 9. Therefore, they can be removed and their neighboring connections merged. This merging of the neighboring connections also includes merging the CentroidClusters 114.

Redundancy of Intersectioncluster 115:

An intersection cluster 115 can be considered redundant if the following conditions are met:

Number of neighboring connections == 2 the travel directions of the neighboring connections agree both connections are bidirectional OR both are NOT bidirectional and have the same direction connection_1_end == connection_2_start OR connection_2_end == connection_1 start In Fig. 9a: Merging (partially) parallel end connections

In various situations, it is possible for two IntersectionClusterConnections 117 connected to the same intersection cluster 115 to have (at least partially) parallel CentroidClusters 114. These parallel parts can be merged into a single new IntersectionClusterConnections 117 with a new Intersectioncluster 115 at the point where the CentroidClusters 114 are no longer parallel. At this point, it is checked whether the distance between this new intersection cluster 114 and the remaining end intersection clusters 114 is greater than a certain threshold (as in the removal of short dead end connections). If this is not the case, the intersection clusters 115 are not connected and the short end connections are removed, otherwise they are added to the ClusterGraph.

The following paragraphs explain the mentioned objects for creating the path network 210 in more detail.

links

A connection describes a section of road between two connecting intersections. Every connection has a starting point and an ending point. The start and end points are not presented as independent features. They always belong to the connection intersection at the beginning or end of the connection. The direction from the start point to the end point is the forward direction. A connection can be unidirectional or bidirectional and describes the permitted traffic flow on the road section.

Connection intersections

A connector intersection represents a street intersection where two or more streets meet or intersect. An intersection is also used to model the end point of a dead end. Two streets that intersect but are not topologically connected do not form an intersection. An intersection is represented by exactly one point in the road network.

Lanes

A lane describes a section of lane between two lane intersections. Each lane has a start and an end point. The start and end points are not presented as independent features. They always belong to the lane intersection at the beginning or end of the lane. Lanes are always unidirectional and describe a logical path from the start to the end point. Lane intersections

A lane intersection represents a lane intersection where two or more lanes meet or separate. A lane intersection is also used to model the end point of a dead end. Two lanes that intersect but are not topologically connected do not form an intersection. An intersection is represented by exactly one point in the lane network.

shape_points_lines

A shape dot line is the geometric representation of a lane that describes the commonly used center line of the lane. It is represented by a polyline. A waypoint 111 represents a single position of a single trajectory 110 and is the smallest increment of information. The trajectory 110, to which the waypoint 111 belongs, is determined via the track ID. This information can be used optionally. In addition to the pose information (position and orientation), each waypoint 111 contains the information as to which other waypoint 111 is its predecessor and which is its successor in the trajectory 110. After the WaypointCentroids 113 have been calculated, each waypoint 111 also knows which centroid, i.e. center of gravity, it is assigned to.

WaypointCentroid

The WaypointCentroid 113 is the result of grouping waypoints 111 of different trajectories 110 and has a pose (position and orientation) that depends on the waypoints 111 assigned to it. It contains references to these waypoints 111 and the IDs of their trajectories 110. A centroid can be part of several Centroid Clusters 114 and thus have several predecessor or successor centroids. Each centroid may also contain information about which Intersection Cluster 115 or Intersection Cluster Connections 117 it belongs to.

Centroid Cluster

A CentroidCluster 114 is a polyline of WaypointCentroids 113.

Intersection cluster

The intersection clusters 115 are the start, intersection and end points of the road network (link intersections). Each intersection cluster 115 consists of at least one waypoint centroid 113, but can also have several centroids, especially in the case of intersections. Their position, and thus the position of their waypoints 113, determines the position of the intersection cluster 115. However, since the intersection cluster 115 is part of a logical graph, its position is only important for visualization. An intersection cluster 115 has at least one neighboring IntersectionClusterConnection 117 to connect it to another intersection cluster 115.

Additionally, an intersection cluster 115 may have information about CentroidClusters 114 and their centroids if this intersection cluster 115 was created during the reduction of an IntersectionClusterConnection 117. The connection information for its Centroid Clusters 114 and centroids are written to the new Intersection Cluster 115.

IntersectionClusterConnection

An IntersectonClusterConnection 117 is a connection between two IntersectionClusters. 115. It also represents a connection between intersections in the road network (link). Such a connection can be unidirectional or bidirectional, which is determined by the bidirectional member flag. If the connection is unidirectional, it points from the first to the second intersection cluster 115. The clusters member contains references to all CentroidClusters 114 that are at least partially represented by the connection, and the centra ids_of_cl usters member contains references to the specific WaypointCentroids 113 of these clusters, that are represented by the connection.

Note: An IntersectionClusterConnection 117 often only references a portion of the WaypointCentroids 113 of a CentroidCluster 114.

ClusterGraph

The ClusterGraph is the main object that contains all relevant objects for creating road networks.

The idea of the invention is to use the movement data from several trips to identify points at which “certain poses accumulate”. First, it ensures that all clusters are unidirectional. Furthermore, intersections and parallel trajectory sections are selected in the resulting trajectory clusters. On this basis, a graph structure with its nodes and edges is generated, which represents the logical path network. This has two levels: the street and its lanes.

These levels are logically linked to each other, so routing can access the information from both levels.

The aspects and features described in connection with a particular one of the previous examples may also be combined with one or more of the further examples to replace an identical or similar feature of that further example or to additionally introduce the feature into the further example . Examples may further include or relate to a (computer) program with program code for executing one or more of the above methods when the program is executed on a computer, a processor or other programmable hardware component. Steps, operations or processes of various of the methods described above can also be carried out by programmed computers, processors or other programmable hardware components. Examples may also include program storage devices, e.g. B. digital data storage media, which are machine-, processor- or computer-readable and encode or contain machine-executable, processor-executable or computer-executable programs and instructions. The program storage devices can e.g. B. include or be digital storage, magnetic storage media such as magnetic disks and magnetic tapes, hard disk drives or optically readable digital data storage media. Further examples may also include computers, processors, control units, (field-programmable logic arrays ((F)PLAs = (Field) Programmable Logic Arrays), (field-programmable gate arrays ((F)PGA = (Field) Programmable Gate Arrays), graphics processors ( GPU = Graphics Processor Unit), application-specific integrated circuits (ASIC = application-specific integrated circuit), integrated circuits (IC = Integrated Circuit) or system-on-a-chip (SoC) that are used to execute the steps of the procedures described above are programmed.

It is further understood that the disclosure of several steps, processes, operations or functions disclosed in the description or the claims should not be interpreted as necessarily being in the order described, unless this is explicitly stated in the individual case or is absolutely necessary for technical reasons . Therefore, the foregoing description does not limit the performance of multiple steps or functions to a particular order. Further, in further examples, a single step, function, process, or operation may include and/or be broken down into multiple substeps, functions, processes, or operations.

If some aspects have been described in the previous sections in connection with a device or a system, these aspects should also be understood as a description of the corresponding method. For example, a block, a device or a functional aspect of the device or system can correspond to a feature, such as a method step, of the corresponding method. Correspondingly, aspects that are described in connection with a method are also considered a description of a corresponding block Element, a property or a functional feature of a corresponding device or a corresponding system.

The following claims are hereby incorporated into the Detailed Description, with each claim standing on its own as a separate example. Furthermore, it should be noted that, although a dependent claim refers in the claims to a particular combination with one or more other claims, other examples may also include a combination of the dependent claim with the subject matter of any other dependent or independent claim. Such combinations are hereby explicitly suggested, unless it is stated in individual cases that a particular combination is not intended. Furthermore, features of a claim for any other independent claim are also intended to be included, even if that claim is not directly defined as dependent on that other independent claim.

Reference symbol list

Providing at least two trajectories

Determining corresponding trajectory sections

Determine waypoint centroids

Identifying focus clusters

Determining nodes

Determine additional waypoint centroids

Determining node connections

Replace those nodes that were used to determine the further waypoint centroids with the corresponding further waypoint centroids

Check whether nodes exist where it is possible to have neighboring ones

To reach nodes via more than one node connection

Determining node centroids

Replacing node connections

Delete affected nodes

Determine a network of paths in the area

Splitting the identified path network in the area

Providing the path network to a motor vehicle

motor vehicle

Trajectory

Trajectory sections corresponding to waypoints

Waypoint centroids

Focus cluster

Nodes

More waypoint focal points

Junction connections Node centroids Surrounding area Path network First level Second level

Claims

Claims A method for determining a route network for a motor vehicle (100), preferably an autonomously driving motor vehicle, comprising the method: Providing (10) of at least two trajectories (110), each trajectory (110) being indicative, for one of a motor vehicle (100) path traveled within an environment (200) of the motor vehicle (100), the trajectory (110) has waypoints (111), each waypoint (111) of the trajectory (110) a dedicated point of the motor vehicle (100) along the route in the environment (200) at a corresponding time, and the environment (200) preferably has a parking environment, determining (20) corresponding trajectory sections (112) from the trajectories

(110),

Determining (30) waypoint centroids (113) of corresponding waypoints

(111) of corresponding trajectory sections (112), each waypoint center of gravity (113) having a position and orientation within the environment (200),

Determining (40) centroid clusters (114) based on the determined waypoint centroids (113),

Determining (50) nodes (115) from the center of gravity clusters (114), determining (60) further waypoint centers of gravity (116), based on essentially parallel oriented center of gravity clusters (114),

Determining (70) node connections (117), based on the further waypoint centers of gravity (116) and the node points (115), such that a node (115) is linearly connected to its nearest node neighbors,

Check (80) whether nodes (115) exist where it is possible to reach neighboring nodes via more than one node connection (117),

If such nodes exist:

Determining (82) node centers (118) of those nodes (115) that can reach each other via more than one node connection (117), replacing (84) all node connections (117) between each affected node (115) and the respectively connected unaffected nodes Nodal points (115), through nodal connections (117) between the corresponding node center of gravity (118) and the unaffected connected nodes (115), and

Delete (86) the affected nodes (115), and

Determining (90) a path network (210) in the environment (200), based on the remaining nodes (115), the node centers (118) and the node connections (117). The method according to claim 1, wherein the determination (60) of further waypoint centroids (116) is additionally based on at least one node (115) of at least one of the essentially parallel oriented centroid clusters (114). The method according to claim 2, wherein determining (70) node connections (117) based on the further waypoint centroids (116) and the nodes (115) comprises:

Replacing (72) the nodes (115) that were used to determine (60) the further waypoint centroids (116) with the corresponding further waypoint centroids (116). The method according to any one of the preceding claims, wherein the focus clusters (114) are indicative of sections of a trail network (210) in the surrounding area (200). The method according to any one of the preceding claims, wherein the nodes (115) are indicative of starting points, ending points and crossing points of the path network (210) of the environment (200). The method according to any one of the preceding claims, wherein a centroid cluster (114) has only two nodes (115). The method according to any one of the preceding claims, the method further comprising:

Dividing (92) the determined road network (210) of the surroundings (200) into a first level (212), which is indicative of a road network of the surroundings (200), and into a second level (214), which is indicative of Surrounding lanes (200). The method according to any one of the preceding claims, the method further comprising: Providing (94) the path network (210) to a motor vehicle (100). A device (300) for determining a road network (210) for a motor vehicle (200), preferably an autonomously driving motor vehicle, comprising the device (300):

An interface (310), set up to communicate control information with at least one specific vehicle component (110) of a motor vehicle (100), and a control module (320), set up to: provide at least two trajectories (110), wherein each trajectory (110) is indicative of a path traveled by a motor vehicle (100) within an environment (200) of the motor vehicle (100), the trajectory (110) has waypoints (111), each waypoint (111) of the trajectory ( 110) has a dedicated point of the motor vehicle (100) along the route in the environment (200) at a corresponding time, and the environment (200) preferably has a parking environment, to determine corresponding trajectory sections (112) from the trajectories (110), To determine waypoint centroids (113) of corresponding waypoints of corresponding trajectory sections (112), each waypoint centroid (113) having a position and orientation within the environment (200), to determine centroid clusters (114), based on the determined waypoint centroids (113), nodes (115) from the centroid clusters (114), to determine further waypoint centroids (116), based on essentially parallel oriented centroid clusters (114), to determine node connections (117), based on the further waypoint centroids (116) and the nodes ( 115), such that a node (115) is linearly connected to its nearest node neighbors, to check whether nodes (115) exist where it is possible to connect neighboring nodes (115) via more than one node connection (117). reach if such nodes exist:

to determine the focal points (118) of those nodes (115) that can reach each other via more than one node connection (117), To replace all node connections (117) between each affected node (115) and the respectively connected unaffected nodes (115) with node connections (117) between the corresponding node center of gravity (118) and the unaffected connected nodes (115), and the affected nodes (115) and to determine a path network (210) of the environment (200) based on the remaining nodes (150), the node centroids (118) and the node connections (117). A computer program product with program code for carrying out a method according to any one of claims 1 to 8, when the computer program product is executed on a computing unit or a device according to claim 9.