Method and system for generating a round trip

Abstract

A method of generating a round trip in a network of transport lines on geospatial data is disclosed, wherein the geospatial data is partitioned using a grid of a plurality of adjacent geospatial cells, wherein each of the geospatial cells is associated with information at least about points of interest and transport lines enclosed therein. The method comprises obtaining an origin for the round trip, wherein the origin is a geospatial location, and setting a starting point and an end point for the round trip as a common point on the network of transport lines based on the obtained origin, and obtaining route parameters defining at least one of a requirement and a constraint for the round trip. One of the cells in which the origin is located is determined as an origin cell, and cells surrounding the origin cell are evaluated based on their respectively associated information in view of the obtained route parameters. Some of the evaluated cells are selected based on evaluation results, wherein the selected cells together with the origin cell form a closed ring of adjacent cells. One or more intermediate points are determined based on the assigned information about the points of interest, wherein in each cell one point of interest is selected as a respective one of the intermediate points. Route segments between the starting point, the one or more intermediate points and the end point are determined and connected to obtain a round trip from the starting point, via the one or more intermediate points to the end point.

Description

[0001] METHOD AND SYSTEM FOR GENERATING A ROUND TRIP

[0002] The present invention relates to the field of route planning, and more particularly to the field of generating a round trip. Specifically, the invention relates to a method and system for generating a round trip along geographical transport lines, e.g., roads, including one or more intermediate points, such as points of interest along the round trip.

[0003] Modern navigation systems, online map applications or other services often include a route planning feature for generating a route typically from a start to an end, or between two or more points. Routes may be generated that consider certain constraints or preferences of a user and may, thus, apply a multitude of optimization parameters, e.g., the shortest distance, shortest time, most fuel efficient, traffic jam avoidance or the like. The points of a route may be part of a network of transport lines on geospatial data. Transport lines may be all kind of roads, streets, highways or the like but may also include unpaved paths or other geographical lines suitable for any transport. The type of transport line used for a planned route may depend on the type of transportation means, which may include, e.g., a car or other motorized vehicle. It is also known, however, to plan routes for walking tours.

[0004] Route planning may particularly be based on a transport line grid included in geospatial data. Geospatial data is information that has a geographic aspect to it. This means that the records in a dataset have coordinates, addresses, or other locational identifiers that tie them to a place on the Earth's surface. The term “geospatial” combines “geo” (a prefix meaning “earth” or “land”) and “spatial” (pertaining to space or location). Geospatial data may particularly include latitude and longitude, address, zip code, or even more complex references like census tracts or areas within political boundaries. Geospatial data can be either vector or raster data. Vector data represent features as points, lines, and polygons, while raster data represent features as a grid of pixels, often used for different kinds of imaging data, such as satellite imagery or scanned maps. Geospatial data may also include a temporal dimension, e.g., weather data that changes over time and influences route planning accordingly.

[0005] Points of interest (“POIs”) may also be described by geospatial data. A point of interest is a specific location that someone may find useful or interesting. POIs include various places such as landmarks, tourist attractions, businesses, and government buildings. In navigation and mapping systems, POIs are used to help users find destinations and services like hotels, restaurants, gas stations, and museums. Therefore, POIs may often play a role in the context of route planning. While in many cases route planning includes a start and an end, and possibly one or more intermediate points, in certain situations it may be required that the route shall be a round trip visiting certain points or regions of the transport line grid, while the start and end of the route are the same point, e.g., for touristic travel, or delivery logistics. The automatic generation of such a round trip, especially with addition of customization parameters, is no trivial task.

[0006] Generating a round trip is comparable to the traveling salesman problem (TSP), which is a classic problem in theoretical computer science and combinatorial optimization. The task is to find the shortest possible round trip that visits a series of cities (each exactly once) and returns to the starting point. Formally speaking, this is a graph problem: given a graph in which the nodes represent the cities, and the edges are valued with costs or distances. The goal is to find a permutation of the cities so that the entire itinerary is as short as possible and each city is visited only once, returning to the starting city. However, while the TSP requires intermediate points (the cities) that are given as input, it may be desirable to plan a round trip without giving or even knowing any intermediate points but only the origin where the route shall start and end.

[0007] The problem of generating a round trip may become even more complex because road networks can be very diverse in density and shape. In some regions, the density of road networks might not be sufficient to provide round trips in all directions or not at all. Therefore, it is not possible to reliably create round trips that meet given criteria without having extensive knowledge about the network. Yet, collecting comprehensive information with every round trip request is costly and tends to exceed the self-imposed latency of a few seconds processing time.

[0008] It is an object of the present invention to provide an improved approach for generating a round trip. Specifically, it is desirable to improve one or more of efficiency, effectiveness, accuracy, and error rate of generating a round trip.

[0009] A solution to this problem is provided by the teaching of the independent claims. Various preferred embodiments of the present invention are provided by the teachings of the dependent claims.

[0010] A first aspect of the invention is directed to a, particularly computer-implemented, method of generating a round trip in a network of transport lines on geospatial data. In the method an origin for the round trip is obtained, wherein the origin is a geospatial location, and setting a starting point and an end point for the round trip as a common point on the network of transport lines based on the obtained origin. Route parameters are obtained defining at least one of a requirement and a constraint for the round trip. One or more intermediate points are determined on the network of transport lines in a geospatial region extending from the origin. In order to obtain a round trip from the starting point, via the one or more intermediate points to the end point, route segments between the starting point, the one or more intermediate points and the end point are determined and connected to each other. A validation value is determined for the obtained round trip using one or more validation parameters and, if the validation value is below the predetermined threshold, the route parameters are adjusted and the determining steps are repeated, or if the validation value is equal to or above a predetermined threshold, the obtained round trip is outputted.

[0011] Accordingly, the method may be considered a sophisticated method for calculating or constructing a round trip for a given location on a network of transport lines on geospatial data. A round trip can be generated that starts and ends at an origin, which may, e.g., be set by a user. The method may then automatically determine intermediate points which shall be travelled along the round trip. The round trip is then validated to ensure that the round trip fulfils certain requirements, such as quality requirements, to provide a round trip as a result, which can be pleasing and meaningful and may look like it has been created by an expert, a local or tour guide. In particular, as will also be explained in more detail below, the method can generate a round trip that can consider a user’s request, preferences, or desires. The round trip can be used, e.g., for touristic purposes or to plan other tours, e.g., for delivery, maintenance, or security services.

[0012] By validating the round trip, low quality tours can be avoided, such as tours that have many doubled visited places, streets that must be driven twice or more, confusing intersections or loops. If the validation value is below a predetermined value, the round trip is recalculated with adjusted route parameters, i.e. , the intermediate points and route segments are newly determined. The resulting newly obtained round trip is again validated until a valid round trip is found. This iterative process, thus, can deliver a valid round trip as result. No manual planning or even knowledge about the region to be travelled is necessary as the intermediate points and the connecting route segments are automatically determined. As a basis, a network of transport lines on geospatial data, such as electronic maps is used to automate the process of finding routes. Geospatial data may further help to improve the resulting round trip. Overall, the method provides an improved way of generating round trips that addresses the above-described needs. The term “round trip”, as used herein, refers particularly to a route or tour that starts and ends at the same point, which may be considered as the “origin”. The origin may be a geospatial location, i.e., a location on the surface of the earth. The round trip may be described on a network of transport lines, which is why the starting point and an end point as well as intermediate points shall be part of the network of transport lines. The origin as well as intermediate points of the round trip may be other geospatial locations, such as points of interest, which can be mapped on the network of transport lines.

[0013] The term “transport line”, as used herein, refers particularly to a line, i.e., a longer straight or curved continuous, predominantly one-dimensional geometric structure. A transport line shall be on “geospatial data”, i.e., on the surface of the earth, and may be a line forming or being part of a transport infrastructure, e.g., for transporting people or goods. Examples of transport lines are street, roads, highways or any other paved or unpaved way or path that can be travelled along, either using any kind of vehicle, such as a car, or by cycling or walking. A transport line may correspond to a section of a more complex network comprising multiple transport lines, such as for example a road network comprising multiple interconnected roads.

[0014] The term “geospatial data”, as used herein, refers to data, that can be defined or expressed in terms of a geographic coordinate system, such as longitude and latitude or UTM coordinates. In particular and without limitation, a transport line, such as a road, as represented by an electronic map, such as an electronic map based on map data for car navigation systems or map on navigation applications on a computer or handheld device, e.g., smart phone or tablet computer, may be described by geospatial data or respective map data.

[0015] The term “obtaining” (or “acquiring”) data, e.g., route parameters, refers particularly to (i) generating that data by the “obtaining” entity itself, e.g., by means of one or more sensors or by deriving it from input data, such as user input, or (ii) to receiving that data from an external data source.

[0016] The terms “first”, “second”, “third” and the like in the description and in the claims, are used for distinguishing between similar elements and not necessarily for describing a sequential or chronological order. It is to be understood that the terms so used are interchangeable under appropriate circumstances and that the embodiments of the invention described herein are capable of operation in other sequences than described or illustrated herein. Where the term "comprising" or “including” is used in the present description and claims, it does not exclude other elements or steps. Where an indefinite or definite article is used when referring to a singular noun e.g., "a" or "an", "the", this includes a plural of that noun unless something else is specifically stated.

[0017] Further, unless expressly stated to the contrary, "or" refers to an inclusive or and not to an exclusive or. For example, a condition A or B is satisfied by any one of the following: A is true (or present) and B is false (or not present), A is false (or not present) and B is true (or present), and both A and B are true (or present).

[0018] As already briefly mentioned above, the obtained (and validated) round trip resulting from the method of the first aspect may be used in many different applications. A round trip can be applied to a touristic context to mimic the behavior of a tour guide. Applied to a security context, it can provide an unbiased patrol route for police, military, or other security entities that covers a maximum of an area selected by objective properties, e.g., incidents reported in the last week. Another application may be in a maintenance context, where the round trip can provide an optimal plan for on-site visual inspection of the condition of roads and their surroundings. Based on general road grade, condition at last inspection, time since last inspection, and estimated strain by traffic, it could be possible to derive a deterioration model that simulates when a road section should be included into a round trip.

[0019] In the following, preferred embodiments of the method are described, which can be arbitrarily combined with each other or with other aspects of the present invention, unless such combination is explicitly excluded or technically impossible.

[0020] In some embodiments, the validation value is determined to comprise one or more of the following: a shape factor and a deviation factor. The shape factor represents a ratio of a perimeter and an area defined by the obtained round trip on the geospatial data. In other words, a valid round trip should have a minimal roundness. While this may depend on other constraints like landscape geography or simply user preferences, a round trip may be considered valid or at least preferred if it actually travels “around” rather than, e.g., only back and forth. The deviation factor represents a deviation of actual properties of the obtained round trip from respective ones of the obtained route parameters. This may help to ensure that the deviation of certain parameters of the resulting round-trip, e.g., distance or time, to the route parameters is below a threshold. In this way, a round trip can be provided that may be optimized, e.g., with regards to a user’s needs, such as a round trip that is not much longer or shorter than what the user expects or desires. In some embodiments, the validation value is determined by calculating a weighted and normalized linear combination of at least the shape factor and the deviation factor. In this way, the validation can be customized and adapted to certain needs. For instance, the above-mentioned factors (and possibly further factors contributing to the validation value) can be weighted with individual weights. Normalization can be performed on a unified domain, which could be a user specific setting or could be learned from past trips.

[0021] In some embodiments, the method further comprises postprocessing the obtained round trip to identify and remove invalid sections of the obtained round trip. This can be done to clean an obtained round trip. The invalid sections may comprise loops, double or multiple travels, clusters, or the like or a combination thereof, which could make the tour unpleasant. Such sections of the round trip can be charged with higher costs to decrease the likelihood of being selected again in a re-calculation of the round trip.

[0022] Loops can be removed by deleting transport lines that form the loop, such that each point on the network of transport lines is travelled through only once. By removing loops, also self-intersections of the round trips can be removed. However, it will be appreciated that only small loops may be removed, while larger loops could be allowed if reasonable in the context of a certain routing request.

[0023] Multiple travels may be removed by indicating transport lines that are travelled more than once and at least partially re-calculating the round trip avoiding the indicated transport lines such that each transport line is travelled only once. Multiple travels can be regarded as enhancing a “tentacle factor”, which can describe the portion of double or multiple usage of transport lines. The tentacle factor, thus, should be below a certain threshold. However, it will be appreciated that in certain cases double travels may be allowed, e.g., if removing them would result in a dead end.

[0024] There may also be clusters of transport lines around an intermediate point, wherein the clusters are removed by moving a respective intermediate point to a location nearby on the network of transport lines and at least partially recalculating the round trip. Clusters may lead to many maneuvers at an intermediate point and may be removed by (slightly) relocating an intermediate point, e.g., by putting it on the first (local) maneuver and omitting the set of complex maneuvers. Clusters may be an indication for a complex topology that may malform the round trip and should be reduced or avoided. In some embodiments, the invalid sections are identified by determining a lateral distance between sections of the round trip, wherein sections are identified as invalid sections if the determined lateral distance is below a predetermined threshold. The distance between two sections of the round trip can be calculated in various ways. For instance, a clearance band may be constructed along the route, which should not overlap. Another method may be to cut the round trip into small sections, which when virtually rotated e.g., by 90 degrees should not intersect with another section. It will be appreciated that the lateral distance may be assumed zero for sections of the round trip that are travelled twice.

[0025] In some embodiments of the present invention, which may be considered a first variant of constructing the round trip, the determining the one of more intermediate points comprises obtaining a travel direction as one of the route parameters, and determining a center for the geospatial region in which the intermediate points are determined, wherein the center is located in the travel direction relative to the origin. This provides a method in which the area where the round trip shall be located can be defined by giving a travel direction, the travel direction particularly originating from the origin. A center is then determined, which may be considered a point or region around which the round trip shall travel. A distance of the center from origin may depend on route parameters, such as travel distance or travel time as the distance between the origin and the center influences the overall length and duration of the round trip.

[0026] In some embodiments, the travel direction is obtained by receiving a user input that specifies a desired direction as the travel direction or by determining a direction as the travel direction based on environmental criteria of a geospatial region surrounding the origin. For instance, a direction, such as “south” or “north” (from the origin) may be indicated by a user. The travel direction may also be determined depending, e.g., on a description of the type of travel or by indicating certain points of interest relative to the origin.

[0027] In some embodiments, the method further comprises selecting a basic route geometry, wherein the basic route geometry is selected to extend around the center and to contain the origin along an outline thereof and defining one or more auxiliary intermediate points located on the outline of the basic route geometry. The intermediate points may then determined by projecting the auxiliary intermediate points to respective points on the network of transport lines in a respective vicinity of the auxiliary intermediate points. Specifically, the intermediate points are locations that are actually visited along the round trip, whereas the auxiliary intermediate points are points on the outline of a selected basic route geometry. This is an auxiliary measure to roughly define a way the round trip shall take around the center. By providing a basic route geometry as a “control shape”, the efficiency of generating the round trip can be increased. In particular, the iterative validation process may be accelerated compared to cases, in which intermediate points are arbitrarily chosen without respect to a basic route geometry. When projecting the auxiliary intermediate points to intermediate points, a plausibility check can be performed, e.g., by checking whether the intermediate points are actually accessible (and not, e.g., in a water area). The intermediate points shall be projected from the auxiliary intermediate points on the basic route geometry to points actually located on the network of transport lines.

[0028] In some embodiments, the basic route geometry is selected as a regular geometric shape, wherein the regular geometric shape is a circular, elliptic, or polygonal shape. This facilitates the procedure. For instance, a circle can be selected to describe the basic geometry of the round trip, wherein the center of the circle is the aforementioned center, and the origin lies on the outline of the circle.

[0029] In some embodiments, the auxiliary intermediate points are defined as a selected number of equidistant points along the outline of the basic route geometry. While in principle a single intermediate point may be sufficient, it is preferable to select at least two or more intermediate points that “span” the area of the round trip. It will be appreciated that equidistant points may be advantageous in many cases, but other choices, e.g., depending on landscape conditions or user preferences (e.g., manually input intermediate points) may be also appropriate.

[0030] In some embodiments, the determining of the route segments comprises determining an order of the intermediate points in which they are travelled along the round trip, wherein the order is based on a clockwise or counterclockwise order of the respective auxiliary intermediate points on the outline of the basic route geometry. This provides an efficient way for setting the order for the intermediate points in which they shall be travelled. Also, this may accelerate the validation process as this order selection may avoid or reduce unwanted route forms, such as intersections or double travel of certain sections.

[0031] In some embodiments, the method further comprises defining at least one restriction area, the restriction area comprising at least an area including the center, wherein the restriction area defines a geospatial region that is restricted for route segments to pass therethrough when determining the route segments. Avoiding the center or a center region in the overall travel area (which is basically set by the basic route geometry) for route sections to pass therethrough can effectively avoid intersections or “unround” route shapes, such as “star shapes”, which include many unpleasant travels back and forth between the center and an outer area. The restriction area may also be referred to as “avoidance area” or “avoidance geometry”.

[0032] In some embodiments, the restriction area is defined to have a minimum distance from each of the one or more auxiliary intermediate points. This avoids cases, where for instance the center restriction area is chosen relatively large, to thereby allow sufficient room for determining the route segments. In other words, clearance spaces are held free around the intermediate points to ensure proper route finding between them.

[0033] In some embodiments, the determining the route segments comprises selecting the order of the intermediate points based on transport line characteristics, environmental criteria, or user preferences. In particular, selecting a clockwise or counterclockwise direction may depend on other factors, such as user preferences, daytime or forecasted weather, or even aspects like quality of landscape views when driving in a certain direction. The determining the route segments may also comprise including or excluding transport lines based on transport line characteristics, environmental criteria, or user preferences.

[0034] In some embodiments, the determining the one or more intermediate points comprises at least one of the following: receiving a user input that specifies at least one intermediate point, and selecting at least one point of interest from a collection (e.g., database) of points of interest. In particular, in addition to automatically setting the intermediate points as described above, manual input of intermediate points may be possible. Points of interest (POI) may be valuable for a round trip especially in a touristic context, and may be e.g., certain sights, restaurants or the like. Also in other contexts, POIs may be defined and chosen where appropriate.

[0035] In some embodiments, the obtaining the origin comprises obtaining a current user location, receiving a user input that specifies a desired geospatial location as the origin, or setting a default location as the origin.

[0036] In some embodiments, the route parameters comprise at least one of a travel direction, a travel time, a travel distance, a type of transport line, a type of intermediate point, and energy resources available for the round trip. The travel direction can be the travel direction as explained above. The travel time can be a duration of the trip, i.e., an envisioned total travel time, or a minimum travel time or maximum travel time. The travel distance (trip length) can be an envisioned travel distance, or a minimum travel distance, or maximum travel distance. The type of transport line can be defined as preferred types or types to be avoided (e.g., unpaved paths, highways, toll routes, etc.)

[0037] In some embodiments, the method further comprises obtaining route preferences, wherein the route segments are determined such that route segments are preferred that fulfil one or more of the obtained route preferences. The route preferences may comprise route intrinsic parameters or route extrinsic parameters. The route intrinsic parameters may specify characteristics of the transport lines, such as curviness, speed limits, surface conditions or the like. The route extrinsic parameters may specify characteristics of an environment along the transport lines, such as preferred environmental regions, e.g., forests or mountains, a vicinity to certain POIs, e.g., gas stations, restaurants, landmarks or even weather situations, e.g., avoidance of thunderstorms, or the like.

[0038] In some embodiments, the method further comprises obtaining transport line characteristics, wherein the route segments are determined such that route segments are preferred based on one or more of the obtained transport line characteristics. In this way, the round trip can be further optimized or customized for certain needs. As will also be described in more detail below, the transport line characteristics may comprising at least one of the following data: road sensor data, traffic monitoring data, public warning data, aerial inspection data (e.g., UAV sensor data obtained by an unmanned aerial vehicle (UAV) such as a drone), and charging availability data (e.g., available charging station for electric vehicle along the road).

[0039] In some embodiments of the present invention, which may be considered a second variant of constructing the round trip, the geospatial data is partitioned using a grid of a plurality of adjacent geospatial cells, wherein each of the geospatial cells is associated with information at least about points of interest and transport lines enclosed therein. In this variant, the validation process may be omitted, in particular because of the use of the grid of cells. However, it will be appreciated that the validation process and other features described above may also apply to this variant. Also, it will be appreciated that the use of the cell grid as described herein may also be applied to the first variant described above as far as possible.

[0040] In the method of this embodiment one of the cells in which the origin is located is determined as an origin cell. Cells surrounding the origin cell are evaluated based on their respectively associated information in view of the obtained route parameters. Some of the evaluated cells are selected based on evaluation results, wherein the selected cells together with the origin cell form a closed ring of adjacent cells. The one or more intermediate points are then determined based on the assigned information about the points of interest, wherein in each selected cell one point of interest is selected as a respective one of the intermediate points. The route segments are then determined as described above to obtain the round trip. In other words, in this variant, the round trip is oriented at the cell grid. Since the cells are assigned to information with regards to the transport lines and POIs available in each cell, an evaluation can be made to be able to select cells that are most promising for a round trip. In each cell, one POI is identified, e.g., a POI most relevant for a certain routing request, such that an easy way of finding intermediate points is provided. Because the cells are selected to form a closed ring, it is ensured that a closed round trip having a proper round shape is constructed. Then, for instance postprocessing as described above may be applied to optimize the route defined by the round trip.

[0041] In some related embodiments, the evaluation of the cells comprises calculating a score as the evaluation result represented by a normalized numeric value. Numeric values can be easily compared and ranked, which facilitates the selection of cells for the round trip.

[0042] In some embodiments, the score represents a combination of a POI score and a connectivity score, the POI score representing a relevance of the points of interest in a cell in view of the route parameters and the connectivity score representing a connectivity available by transport lines between adjacent cells. These two factors provide an efficient way for generating a pleasant round trip. As points of interest shall be focused on, it shall be found out which cells provide most interesting or most relevant points of interest. The POI score can also consider a number of points of interest per cell. If there is no point of interest in a cell, the center can be used as a “point of interest” if necessary. Furthermore, in terms of a pleasant round trip, cells should provide convenient connectivity. This may consider the number or density of transport lines in a cell and particularly those transport lines that reach out to adjacent cells. In this way, a closed ring of cells can be identified, which on the one hand has most relevant POIs and on the other hand provides appropriate connections for determining route segments that connected together result in a round trip.

[0043] In some embodiments, the cells are selected such that the closed ring of selected cells has the highest score among closed rings of cells surrounding the origin cell, wherein the score of a closed ring of cells is the sum of all scores of the cells in the closed ring.

[0044] In some embodiments, the route parameters comprise a thematic category of points of interest and a travel time. Specifically, these two parameters (along with an origin) may be sufficient to generate a round trip. This is advantageous because the user input can be kept at a minimum, while at the same time a pleasant round trip can be generated.

[0045] In some related embodiments, each cell is assigned with information about one or more of the following. One or more landcover classes may be provided per cell that affect the network of transport lines, such as landcover classes that are very likely to interfere with the route network such as water bodies, rivers, mountainous areas, or the like. Also, one or more landcover classes may be provided that specify an environmental cell character. This may be relevant as this may influence the thematic category of the round trip, such as urban areas or forests. As already mentioned, a cell-to-cell connectivity may be provided, which specifies a connection of a cell to its neighbor cells with respect to the network of transport lines. Last but not least, each cell is assigned information about contained points of interest, which in particular may be grouped by thematic category and ranked by a significance. The significance (relevance) and the respective ranking may depend on route request.

[0046] In some embodiments, the plurality of geospatial cells is formed by a plurality of adjacent hexagons. While other cell grids, such as squares may be used, a hexagonal grid is particularly suitable to cover the surface of the Earth without gaps. Also, a ring of hexagons may be suitable for describing a circle-like shape for a round trip.

[0047] A second aspect of the present invention is directed to a data processing system being configured to perform the method of the first aspect. The data processing system might specifically be configured by means of one or more computer programs to perform the method of the first aspect. In addition, or alternatively, the configuration may be implemented, in whole or in parts by respective hardware.

[0048] A third aspect of the present invention is directed to a computer program or a computer program product, comprising instructions, which when executed on a data processing system according to the second aspect of the invention cause the system to perform the method according to the first aspect of the invention.

[0049] The computer program (product) may in particular be implemented in the form of a data carrier on which one or more programs for performing the method are stored. Preferably, this is a data carrier, such as a CD, a DVD or other optical medium, or a flash memory module. This may be advantageous, if the computer program product is meant to be traded as an individual product independent from the processor platform on which the one or more programs are to be executed. In another implementation, the computer program product is provided as a file on a data processing unit, in particular on a server, and can be downloaded via a data connection, e.g., the Internet or a dedicated data connection, such as a proprietary or local area network.

[0050] The system of the second aspect may accordingly have a program memory in which the computer program is stored. Alternatively, the system may also be set up to access a computer program available externally, for example on one or more servers or other data processing units, via a communication link, in particular to exchange with it data used during the course of the execution of the computer program or representing outputs of the computer program.

[0051] The explanations, embodiments and advantages described above in connection with the method of the first aspect similarly apply to the other aspects of the invention.

[0052] BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Further advantages, features and applications of the present invention are provided in the following detailed description and the appended figures, wherein:

[0054] Fig. 1 schematically illustrates an exemplary embodiment of a method according to the present invention;

[0055] Fig. 2 schematically illustrates an exemplary geospatial region with a given geospatial location as the origin for a round trip;

[0056] Fig. 3 schematically illustrates the geospatial region of Fig. 2 with a travel direction;

[0057] Fig. 4 schematically illustrates examples for basic route geometries;

[0058] Fig. 5 schematically illustrates a circular basic route geometry with auxiliary intermediate points;

[0059] Fig. 6 to Fig. 8 schematically illustrate the construction of a restriction area;

[0060] Fig. 9 schematically illustrates the geospatial region of Fig. 3 with the basic route geometry of Fig. 5; Fig. 10 schematically illustrates the geospatial region of Fig. 9 with the restriction area of Fig. 8.

[0061] Figs. 11 to 15 schematically illustrate determination of route segments;

[0062] Fig. 16 schematically illustrates the resulting round trip;

[0063] Figs. 17 and 18 illustrate a first example of corrections of a round trip;

[0064] Figs. 19 and 20 illustrate a second example of corrections of a round trip;

[0065] Figs. 21 and 22 illustrate a third example of corrections of a round trip;

[0066] Figs. 23 to 25 illustrate examples for cell information;

[0067] Fig. 26 schematically illustrates an evaluation of four adjacent cells;

[0068] Fig. 27 schematically illustrates scores for and selection of cells surrounding an origin cell;

[0069] Fig. 28 schematically illustrates identifying invalid sections;

[0070] Fig. 29 schematically illustrates an example for removing invalid sections from a round trip; and

[0071] Figs. 30 to 32 illustrate an example for generating a round trip using a hexagonal cell grid.

[0072] Fig. 1 illustrates a flow chart of an exemplary embodiment 100 of a method of the first aspect of the present invention. The method 100 is particularly described also referring to Figs. 2 to 16. The method is configured to calculate a round trip 10 for a given location on a network of transport lines on geospatial data.

[0073] For further improvement of the method, it may be envisioned to augment transport lines with additional input that serves as supplementary information for generating the round trip 10. The input data may be selected according to the respective context and used to optimize the result in terms of transport lines that should be avoided or preferred. In the following a non-exhaustive list of possible sources for augmentation input data is provided. The data may be useful to further refine the determination of the route segments, e.g., to better fit certain needs or better fulfil requirements of a certain routing request.

[0074] Firstly, road sensor networks may be utilized that monitor the condition of a road. For instance, data about temperature or wetness may influence the selection or avoidance of certain sections. Traffic monitoring systems that monitor the vehicle flow and traffic jams may deliver respective data that may also be useful. Satellite data may be considered as well, e.g., for providing information on local or regional weather situations, or changes of the terrain as indication, e.g., for possible landslides or other obstacles that may have an impact on transport lines. It may also be envisioned to consider public warning announcements, e.g., warnings about natural disasters such as flooded areas or wildfires. In particular, if an electric vehicle shall be used, occupancy rates of electric chargers may be relevant, e.g., to favor transport lines that are closer to less occupied chargers. This may also lead to better distribution of charger usage. UAV based sensors, e.g., optical RGB or infrared cameras, or radar sensors, may provide inspection data of bridges, road surface conditions, or indications of possible activity in restricted areas. Based on that information, transport lines nearby may be favored to enforce inspection round trips automatically.

[0075] The method can respect a general direction (travel direction) from the given location (origin 2) and a travel time and / or travel distance as base customization parameters (route parameters). A round trip 10 can be constructed, which is not just round in shape but contain sections that are supposed to be interesting and efficient for travelers, maintenance workers, patrols, or other users, i.e. , it is possible to avoid repetition and to take a section of the transport lines just once during the round trip.

[0076] Optionally, a user may define preferred attractions (viewpoints, beaches, archeological sites, castles, ...), or road features (road classes, smoothness, surface types, ...), or preferred regions (regions with recent severe weather, not inspected for a long time) that may further alter the characteristics of the round trip.

[0077] Fig. 2 illustrates an exemplary geospatial region 1 , i.e., a region on the surface of the Earth, in which a round trip 10 shall be generated in accordance with the method 100 of Fig. 1. In a first phase, user defined parameters and preferences are obtained. Specifically, a start location (origin 2) is set in step S1 that can be matched on a geospatial network of transport lines 11 . This is identical to the end point of the round trip 10. The origin 2 can be the current location of the user, e.g., defined by a geoposition system such as GPS. Alternatively, the origin 2 can be set manually by the user. The origin 2 may also be set to a POI, e.g., hotel, to serve as default location for multiple users.

[0078] Route parameters (or user parameters) can then be obtained (cf. database D1). In particular, as shown in Fig. 3 a preferable travel direction 3 with respect to the start location can be set (step S2). This can be manual input of the user, e.g., if a user wants to go south. It may also be envisioned to automatically determine the travel direction 3 by using a direction that offers the best potential to meet fuzzy user constraints, e.g., if a user indicates to get a round trip “with the most forest areas” or the like.

[0079] Furthermore, round trip constraints can be set, such as a desired trip length, available time, available fuel I battery charge. Also, general route preferences may be set that are used to adapt the actual route that is generated to the custom preferences of a user. These can be route intrinsic parameters like curviness, speed limits, or surface conditions and / or route extrinsic parameters such as preferred environmental regions, e.g., forests or mountains, vicinity to POIs, e.g., gas stations, restaurants, landmarks or weather situations, e.g., avoidance of thunderstorms or the like.

[0080] Then, in step S3, intermediate points 8 are determined. This is particularly done with the help of auxiliary intermediate points 6. The auxiliary intermediate points 6 serve as basic control of the route of the round trip 10 on the transport lines 11 of the geospatial network. Other routing constraints may be introduced to achieve the preferred routing result. Further route parameters (database D2) may be considered. First, a basic route geometry 5 (5’, 5”) is defined as a round trip “control shape”. The shape could have any fixed size geometric shape, e.g., circle, rectangle, ellipse as shown in Fig. 4. The basic route geometry 5 extends around a center 4 that is given by the origin 2 and the travel direction 3, wherein the origin 2 is located on the shape (the outline) of the basic route geometry 5. It will be appreciated that the number of auxiliary intermediate points 6 may be at least three to actually span a region and enable a proper round trip creation. The size of the geometric shape 5 could be static or derived from an average speed setting, e.g., 40km / h. The average speed could be calculated in advance for a geographic region (e.g., a county, a state, a country) by sampling random points and calculating the average speed to connect these locations. For further optimizations, the round trip shape could depend on the shape of land and water masses (e.g., by calculating the overlap of land mass and requested round trip shape).

[0081] Referring now to Fig. 5 auxiliary intermediate points 6 on the round trip shape (the basic route geometry 5; here the circle from the examples of Fig. 4 was selected) are picked. This can be done by for instance by sampling equidistant points (four points in the example of Fig. 5) or by finding the most prominent points within a buffer around the round trip shape, e.g., touristic attractions or POI.

[0082] In a step S4 (possibly using geospatial network data, see database D3) a restriction area 7 is constructed that allows finer control over the round trip route and may help to circumvent unwanted situations. A first portion 7’ of the restriction area (“avoidance geometry”) is defined in the center 4 of the round trip control shape 5 (see Fig. 6). Its purpose is to prevent that the resulting round trip 10 collapses and that trips to intermediate points 8 are served from the center 4 using the same transport lines 11 . The shape of this avoidance geometry 7’ may be based on the round trip control shape (i.e. , the basic route geometry 5, which is a circle here). Additional avoidance geometries can be defined by vegetation zones, protected zones, military zones, unrouteable geographic topologies, disaster zones, remote locations without gas stations, etc. in order to increase the quality of the resulting round trip.

[0083] The restriction area can, thus, be constructed as follows. As explained above and shown in Fig. 6 a circle geometry is chosen as base avoidance geometry 7’ and placed in the center 4 of the control shape 5 to avoid round trip routes collapsing into the center 4 of the control shape 5. Then, as illustrated in Fig. 7 clearance buffers 7” are constructed around the (auxiliary) intermediate points 6 to ensure sufficient space around those points for proper route finding. The final avoidance geometry 7 (i.e., the final restriction area) is then constructed by the difference of the base avoidance geometry 7’ with each clearance buffer geometries 7” (result shown in Fig. 8)

[0084] Based on the auxiliary intermediate points 6, intermediate points 8 are determined, i.e., locations that shall actually be travelled along the round trip 10. For this purpose, the auxiliary intermediate points 6 are projected onto meaningful locations based on the transport line grid 11 that form the intermediate points 8. The projection can be done in various ways, e.g., to the closest road segment as shown in Fig. 9. Fig. 10 shows a result by combined views of Fig. 8 and Fig. 3.

[0085] The auxiliary intermediate points 6 may also be projected to the closest POI selected from a group of static or user defined POI, taken from a digital map or a database, or to a nearby point that the user has not visited in the past (i.e., enable round trips to differ every time), or to the closest point on land if the intermediate location is on water, or to a nearby road segment with a specific road profile, e.g., with a specific surface, curvature or otherwise classified as interesting, or to a point representing a cluster of multiple nearby locations of the same type, e.g., touristic attractions, or to the most important POI nearby (e.g., a viewpoint). Importance can be derived from the amount of information provided about that point or any other ranking scheme such as the number of photos taken by users of this location. It will be appreciated that intermediate points (locations) may be restricted to a specific subset of road classes (e.g., primary, secondary, tertiary roads), exclude ferry lines, exclude long road segments where no U-turns are possible, e.g., motorways.

[0086] It may be useful to check plausibility of the determined intermediate points 8, in particular to avoid expensive route calculation as early as possible. For example, intermediate points 8 may be discarded if a specific percentage is located on water (e.g., when trips at the coast are requested into the direction of the open sea) or other difficult to route topology, e.g., sparse mountain passes, fjords, etc. If discarded, a valid round trip may still be generated by adaptation of parameters, e.g., changing the travel direction 3 by + / -30 degrees to stay close to the original direction but allow some flexibility, and restart with constructing the (auxiliary) intermediate points.

[0087] After the intermediate points 8 have been determined, the round trip 10 can be calculated (step S5). Fig. 11 to 15 show how route segments 9 are determined between the intermediate points 8 including the origin 2 (i.e., the starting and end point). The direction of travel of the intermediate points 8 (clockwise or counterclockwise) may be arbitrarily chosen. However, if there are issues in areas with one-ways the direction may be changed. The direction can also be chosen with respect to weather, e.g., to avoid driving through a rain area that moves away later, or, to avoid driving against low angle sun.

[0088] An optimal path from the start point to the end point (which is the same as the start point) including the intermediate locations 8 can be found as shown in Figs. 11 to 15. It will be appreciated that transport lines 11 within the restriction area 7 (avoidance geometries) are excluded. Specific road features based on user preferences may be included or excluded. If no path was found, a valid round trip may still be generated by adaptation of parameters, e.g., changing the travel direction 3 by + / -30 degrees to stay close to the original direction but allow some flexibility, and restart with constructing intermediate points 8.

[0089] The generated round trip 10 is then validated (step S6) using certain validation parameters (see database D4). A valid round trip should have a minimal roundness, i.e., the ratio of the perimeter and the area should be in a valid range. Also, it should be ensured that the “tentacle factor”, i.e., the length of road segments taken twice, is below a threshold. Further, it should be ensured that the deviation of certain parameters of the resulting round trip, e.g., distance or time, to the input parameters is below a threshold. It may be applied an “illness rule”, which may be defined as a linear combination of all other rules with individual weights and a normalization of rules to a unified domain, which could be a user specific setting and be learned from past trips.

[0090] If the round trip 10 is determined to not be valid, the method returns by adjusting parameters (step S7) and restarts from step S3 with constructing intermediate points 8. If a valid round trip 10 is obtained (possibly by iteratively repeating the process) or if iterative round trip generation is enabled and a timeout has been reached, the round trip 10 is outputted. An example result is shown Fig. 16.

[0091] The result can be cleaned to achieve a more pleasant round trip 10. For instance, construction of the round trip 10 may cause loops. Such loops can be discarded by identifying route sections that are taken twice in the same direction. In addition to cutting loops, there might still be short sections where a road segment is taken twice. Those edges can be excluded from the final route, but only if no dead end is formed. It may be required to relocate locations, i.e., if many maneuvers are close to an intermediate point, it is an indication for a complex topology that may malform the round trip. In this case the intermediate point is simplified by putting it on the first (local) maneuver and omitting the set of complex maneuvers. Examples are shown in Figs. 17 to 22 and described in more detail below.

[0092] Fig. 17 shows a first example, where the round trip 10 has a loop 12 around an intermediate location. Loops 12 should generally be suppressed because they deteriorate the visual appearance of the round trip, they extend the round trip time without increasing attractiveness since the loop occurs in a small area around the intermediate location, and they lead over the same road twice which is boring for the touristic driver or unproductive for an inspection ride. The loops 12 can be removed by performing the following steps. The double driven edges are identified. The route is then reconstructed, wherein all edges are iterated. Passing the first double driven edge is the indicator for skipping. All following edges are skipped and not written to the response (the new route) until the double driven edge is passed again. Then skipping is stopped and the edges are written to the response again. The result without loops is shown in Fig. 18.

[0093] Fig. 19 shows another example in which the round trip 10 shows a “tentacle” 13 around an intermediate location. Tentacles 13 should generally be suppressed because they deteriorate the visual appearance of the round trip, and they extend the round trip time without increasing attractiveness since the same road is driven twice which is boring for the touristic driver or unproductive for an inspection ride. This is fixed by avoiding double driven streets with following mechanism. The edges that already occur in the route may be included in an extra cost edges vector and then generally avoided during path finding. The result without tentacles is shown in Fig. 20.

[0094] Fig. 21 shows another example in which the round trip 10 includes unnecessary turning maneuvers (i.e., a “cluster” 14) around an intermediate location. Such behavior should generally be suppressed because it increases the fuel / battery consumption. It may also be exhausting for the touristic driver or unproductive for an inspection ride, since many maneuvers take a longer time and therefore fewer regions can be covered per time. This can be fixed by moving intermediate locations with the following steps. Clusters 14 are identified around an intermediate location and the intermediate location is moved to the first maneuver of a cluster. The round trip is then constructed again with the moved intermediate location. The result without the cluster is shown in Fig. 22.

[0095] As a variant, the calculation of round trips could be parallelized with a variation of (single) parameters within reasonable ranges, e.g., reduction or inflation of the round trip control shape yields a set of round trips with different distances and durations. From this set, the round trip matching best to the user preferences can be chosen.

[0096] A variant can pre-calculate a set of round trips for often requested locations or POI, e.g., hotels, transport depots, central stations.

[0097] A variant can present round trips with variations of parameters, extrinsic and intrinsic route preferences to a user, who indicates whether they like or dislike the round trip without knowledge of underlying parameters and preferences. After a sufficient number of round trips has been presented, the system can derive a profile for the user to present best matching round trips for future requests.

[0098] A variant can adjust the round trip control shape by respecting average waiting I resting times at certain POI to better match the preferred round trip time given by the user.

[0099] For finer round trip control, preference geometries can be incorporated into the generation process similar to avoidance geometries. In contrast to avoidance geometries (restriction areas) they attract the route to certain areas. The above-described variant of the method of generating a round trip 10 is based on a basic route geometry, such as a round shape that is easy to obtain, but, however, does not necessarily respect the shape of land masses. For instance, intermediate points could end up on water areas or in unreachable mountainous areas. The variant described below (which may be an extension of the first variant) takes into account further land cover information and POI locations. For this to work, additional pre-computed geospatial datasets are used, particularly based on a hexagonal grid.

[0100] Road networks can be very diverse in density and shape. In some regions, the density of road networks might not be sufficient to provide round trips in all directions or not at all. Therefore, it is not possible to reliably create round trips that meet given criteria without having extensive knowledge about the network. Yet, collecting comprehensive information with every round trip request is costly and tends to exceed the self-imposed latency of a few seconds processing time. Thus, a method is proposed to obtain an abstract and simple data structure of the influencing factors of a round trip before attempting to actually calculate the route itself.

[0101] For this purpose, a hexagonal grid 20 is introduced with information on the aforementioned factors. For instance, the uber / H3 hexagonal hierarchical geospatial indexing system may be used. It will be appreciated, however, that other grid or indexing systems may be used where appropriate. The information assigned to each cell 21 of the hexagonal grid may comprise the following. Landcover classes are provided that are very likely to interfere with the route network such as water bodies, rivers and mountainous areas. Landcover classes may also be provided that influence the thematic category of the round trip such as urban areas or forests. Also, a cell-to-cell connectivity factor is provided, e.g., the connection of a cell to its neighbors with respect to the network of transport lines 11. Cells 21 are also assigned with information about contained points of interest that represent the routing locations of the round trip, grouped by thematic category, and ranked by significance.

[0102] Generally, a hexagonal grid 20 has certain mathematical advantages over any other grids. Hexagons are optimally space-filling, i.e. , hexagons are the most circle-like regular polygon that is known to cover a plane densely. The sphere (the Earth globe) can be tessellated with hexagons more regularly than with a rectangular grid. In a hexagonal grid the neighbors of a grid cell all have the same distance (unlike square or triangular grids).

[0103] In order to identify cells to be selected for an area of the round trip 10, cell content is evaluated as will be explained below. Fig. 23 illustrates evaluation of cell content with respect to landcover classes. Landcover is classified into types, wherein the area adds up to 100% of the cell’s surface area. Then a ranking of landcover may be done by area proportion and type. In particular, this procedure can be further utilized to evaluate if a cell is urban, rural, mountainous, etc. An output could look like shown in Fig. 23 (dark = high density of urban areas; light = low density of urban areas).

[0104] Fig. 24 illustrates evaluation of cell content with respect to points of interest 23 (POI). POIs 23 are classified into categories. The number of occurrences, i.e., the number of POI 23 of a certain category is variable and can be zero. It may be envisioned to evaluate a popularity. For this purpose, POIs 23 are connected to additional category dependent metadata such as photos taken nearby for touristic destinations or geographic variables such as a peak’s altitude. A ranking with respect of the significance (relevance) of POIs depends on a thematic category of the round trip. For example, high mountain peaks are more important to a touristic panorama themed round trip through the alps while popular architecture or historic buildings may be more important for a sightseeing tour through a city center. If a cell 21 does not hold any POI 23, the center of the cell may be considered a routing location (intermediate point). The single most significant POI 23 per cell 21 is then chosen as a routing location (intermediate point). By this, it is ensured that POI 23 are not clustered very close to each other which is likely to cause complicated round trip geometries. An output could look like shown in Fig. 24 (dark = high density of viewpoints; light = low density of viewpoints)

[0105] Fig. 25 illustrates evaluation of cell content with respect to cell-to-cell connectivity. Cells 21 may be ranked by number and type of transport line 11 between each other. This could look like shown in Fig. 25. Each triangle within a hexagon indicates that a transport line 11 (e.g., motorway) connects to the neighboring hexagon.

[0106] The described evaluation options refer to a non-exhaustive list of geospatial parameters. Others like population density or measures of road quality, weather information, security risks, travel or civil warnings may be made available in the same way. Such data may also be obtained by respective sensors as described above.

[0107] It may be advantageous to individually normalize the values to obtain a comparable score, e.g., linearly between 0 and 1 or 0% and 100% or exponentially (1 1 ; 10 2; 100 3; and so on) or any other normalization schema. An overall score may then be obtained from a weighted average of the scores. The weighting schema may be fixed or user defined. The scores are updated if the raw data changes.

[0108] Fig. 26 shows an example with four cells 21 . Dots in the cells indicate POIs 23, while arrows between the cells 21 indicate respective connectivity between cells 21 via transport lines 11. The right part shows a respective example for composite scores for single cells. More specifically, in cell A, the POI score is 100 % as the most popular POI of the cell selection is contained in this cell. The connectivity score of cell A is also 100 % as the cell meets the required connectivity prerequisites. Thus, an overall score of cell A is (100 + 100) / 2 = 100 %. In cell C, the POI score is 40 %. The POI score is normalized towards the best POI of the selection. The connectivity score of cell C, however, is 0 % because the cell does not meet the required connectivity prerequisites. Thus, the overall score of cell C is (0 + 40) 12 = 20 %.

[0109] Referring now to Fig. 27, it is exemplarily explained how a round trip 10 may be generated according to this variant using the hexagonal grid 20. The middle figure of Fig. 27 is enlarged for better readability of the numbers in the cells 21. Generally, the needed input for round trip creation is advantageously supposed to be as small as possible. This may especially apply to cases where the user has no further knowledge about the geographic surrounding. Thus, it is possible to generate well composed and thematically delineated round trips without any local knowledge. The needed input may be limited to the following parameters: the origin 2, e.g., the starting location (and end point) of the round trip, the desired category of points of interest to visit (a category corresponds to a thematic collection of types that subjectively or objectively describes characteristics of points and areas of interest, e.g., in a touristic context: 'historic', 'scenic', 'maritime'), and the desired duration of the trip.

[0110] First, the user inputs a location that can be mapped to a cell, i.e., the origin cell 22 (left in Fig. 27). Next, scores of the first and second ring of hexagons around the origin cell 22 are evaluated (middle of Fig. 27). By this, a closed ring 24 of cells 21 with the best score for a particular location category combination can be identified. Having identified the optimal ring 24, the single most significant POI 23 from each cell 21 is chosen. The POI collection is then ordered clockwise or counterclockwise and passed to the routing engine (right of Fig. 27).

[0111] Finally, a route is returned that connects all given points of interest while attempting to maintain an approximate circular shape. Yet, no detailed information about the street network was considered in this first routing step. Thus, the first routing request is optimized towards a circular layout of POIs 23 and a reduction of interfering street network influences, still geometric irregularities may occur in the first result. This is due to the fact, that network data was greatly reduced and abstracted in favor of computation performance and latency.

[0112] As also explained for the first variant (method of Fig. 1), a validation process is applied to avoid returning irregular shapes of round trips. Proportions of route segments considered as invalid decrease the overall quality. It will be appreciated that the methods below could also be added to the validation step S6 described above. In particular, invalid segments of a route are parts of the route driven twice (“tentacles”), enclosed loops that are connected to the main route, zig-zag patterns (if avoidable), or self-intersections.

[0113] Fig. 28 illustrates a way of identifying invalid segments 25. In particular, sections that are too close together are potentially invalid, such as (small) loops, tentacles, or self-intersections. Thus, (with the exception of large loops) the aforementioned patterns can be detected by checking predefined segments of the route for clearance. Fig. 28 in a) to d) show how to identify and remove invalid sections, which are the two vertical lines. First, the route’s shape is simplified and split into segments 26 of fixed size (Fig. 28a). These segments 26 are then rotated by 90 degrees (dotted lines in Fig. 28b). Intersections of rotated segments with parts of the simplified route shape are detected (Fig. 28c). The basic idea behind this is to detect road segments that are nearby in relation to the segment length, i.e., for longer segments the roads need to be further apart than for just a small breakout. These detections lead to classifying the linked segment as invalid, e.g., a member of a geometric irregularity (dotted segments in Fig. 28d are invalid). If invalid segments 25 split the overall geometry into multiple parts (see Fig. 28e), smaller geometries 27 are also marked as invalid (Fig. 28f)

[0114] The length of invalid segments 25 in proportion to valid segments represents the geometric quality. If the geometric quality is below a given threshold, improvement attempts are made such to correct the irregular shape as in the simplified exemplary route shown in Fig. 29.

[0115] The route shape (Fig. 29a) is divided into valid and invalid segments 25 as described above (see Fig. 29b). All routing locations are then mapped onto the valid segments (Fig. 29c). By this, POIs 23 that were formerly reached via invalid segments are mapped further away from their original location. POIs 23 reached by valid segments are connected to the round trip in the same way. With a second routing request using the optimized locations, invalid segments 25 are very likely to be excluded from the overall shape. At this point, average distance to the original set of POI 23 increased while geometric quality increased as well. The result is shown in Fig. 29d. 5. These steps of “quality assessment” and “quality improvement” can be repeated iteratively until the geometric quality reaches the desired threshold or locations cannot be mapped to a better location.

[0116] An example of the method using the hexagonal grid, including respective corrections is shown in Figs. 30 to 32. As illustrated in Fig. 30 an optimal ring of grid cells 21 close to the center location is obtained (in this case the leftmost cell was selected as the origin of travel; shown are the cells and the highest ranked points of interest within each cell are connected by a dashed line). The route segments that connect highest ranked POIs 23 are determined. A respectively obtained round trip 10 is shown in Fig. 31. Then geometric quality is evaluated, and invalid route segments are identified (shown as white areas in the lower three grid cells in Fig. 31). POIs 23 are mapped to the nearest valid segments of the route, and the generation of the round trip 10 is (iteratively) repeated to obtain a new route. This procedure terminates as soon as a geometry quality threshold is reached (otherwise the mentioned steps are reiterated until the threshold is reached). A result is shown in Fig. 32, i.e. , a round trip 10 without invalid sections that cause the irregularities, thereby achieving a pleasant round trip.

[0117] While above at least one exemplary embodiment of the present invention has been described, it has to be noted that a great number of variations thereto exists. Furthermore, it is appreciated that the described exemplary embodiments only illustrate non-limiting examples of how the present invention can be implemented and that it is not intended to limit the scope, the application or the configuration of the herein-described apparatus’ and methods. Rather, the preceding description will provide the person skilled in the art with constructions for implementing at least one exemplary embodiment of the invention, wherein it has to be understood that various changes of functionality and the arrangement of the elements of the exemplary embodiment can be made, without deviating from the subject-matter defined by the appended claims and their legal equivalents.

[0118] LIST OF REFERENCE SIGNS

[0119] 1 geospatial region

[0120] 2 origin

[0121] 3 travel direction

[0122] 4 center

[0123] 5 basic route geometry (control shape)

[0124] 6 auxiliary intermediate point

[0125] 7 restriction area

[0126] 8 intermediate points

[0127] 9 route segments

[0128] 10 round trip

[0129] 11 transport lines

[0130] 12 loop

[0131] 13 tentacle

[0132] 14 cluster

[0133] 20 hexagonal grid

[0134] 21 cell

[0135] 22 origin cell

[0136] 23 point of interest (POI)

[0137] 24 ring

[0138] 25 invalid segment

[0139] 26 segment

[0140] 27 disconnected portion

Claims

CLAIMS1 . A method of generating a round trip (10) in a network of transport lines (11) on geospatial data, wherein the geospatial data is partitioned using a grid (20) of a plurality of adjacent geospatial cells (21), wherein each of the geospatial cells (21) is associated with information at least about points of interest (23) and transport lines (11) enclosed therein, the method comprising:- obtaining an origin (2) for the round trip (10), wherein the origin (2) is a geospatial location, and setting a starting point and an end point for the round trip (10) as a common point on the network of transport lines (11) based on the obtained origin (2);- obtaining route parameters defining at least one of a requirement and a constraint for the round trip (10);- determining one of the cells (21) in which the origin (2) is located as an origin cell (22);- evaluating cells surrounding the origin cell (22) based on their respectively associated information in view of the obtained route parameters;- selecting some of the evaluated cells (21) based on evaluation results, wherein the selected cells (21) together with the origin cell (22) form a closed ring (24) of adjacent cells (21);- determining one or more intermediate points (8) based on the assigned information about the points of interest (23), wherein in each selected cell (21) one point of interest (23) is selected as a respective one of the intermediate points (8); and- determining route segments (9) between the starting point, the one or more intermediate points (8) and the end point and connecting the determined route segments (9) to obtain a round trip (10) from the starting point, via the one or more intermediate points (8) to the end point.

2. The method of claim 1 , wherein evaluating the cells (21) comprises calculating a score as the evaluation result represented by a normalized numeric value.

3. The method of claim 2, wherein the score represents a combination of a POI score and a connectivity score, the POI score representing a relevance of the points of interest (23) in a cell (21) in view of the route parameters and the connectivity score representing a connectivity available by transport lines (11) between adjacent cells4. The method of claim 2 or 3, wherein the cells (21) are selected such that the closed ring (24) of selected cells has the highest score among closed rings (24) of cells surrounding the origin cell (22), wherein the score of a closed ring (24) of cells (21) is the sum of all scores of the cells (21) in the closed ring (24).

5. The method of any one any one of the preceding claims, wherein the route parameters comprise a thematic category of points of interest (23) and a travel time.

6. The method of any one any one of the preceding claims, wherein each cell (21) is assigned with information about one or more of the following:- one or more landcover classes affecting the network of transport lines (11);- one or more landcover classes specifying an environmental cell character;- a cell-to-cell connectivity, which specifies a connection of a cell (21) to its neighbor cells (21) with respect to the network of transport lines (11);- contained points of interest (23) grouped by thematic category and ranked by a significance.

7. The method of any one any one of the preceding claims, wherein the plurality of geospatial cells (21) is formed by a plurality of adjacent hexagons.

8. The method of any one any one of the preceding claims, further comprising determining a validation value for the obtained round trip (10) using one or more validation parameters and- if the validation value is below the predetermined threshold, adjusting the route parameters and repeating the determining steps, or- if the validation value is equal to or above a predetermined threshold, outputting the obtained round trip (10).

9. The method of claim 8, wherein the validation value is determined to comprise one or more of the following:- a shape factor, the shape factor representing a ratio of a perimeter and an area defined by the obtained round trip (10) on the geospatial data; and- a deviation factor, the deviation factor representing a deviation of actual properties of the obtained round trip (10) from respective ones of the obtained route parameters.

10. The method of claim 9, wherein the validation value is determined by calculating a weighted and normalized linear combination of at least the shape factor and the deviation factor.11 . The method of any one of the preceding claims, further comprising postprocessing the obtained round trip (10) to identify and remove invalid sections (25) of the obtained round trip (10), the invalid sections (25) comprising at least one of the following:- loops (12), wherein the loops (12) are removed by deleting transport lines (11) that form the loop (12), such that each point on the network of transport lines (11) is travelled through only once;- multiple travels (13) of transport lines (11), wherein the multiple travels (13) are removed by indicating transport lines (11) that are travelled more than once and at least partially recalculating the round trip (10) avoiding the indicated transport lines (11) such that each transport line is travelled only once; and- clusters (14) of transport lines (11) around an intermediate point, wherein the clusters (14) are removed by moving a respective intermediate point to a location nearby on the network of transport lines (11) and at least partially recalculating the round trip (10).

12. The method of claim 11 , wherein the invalid sections (25) are identified by determining a lateral distance between sections (26) of the round trip (10), wherein sections (26) are identified as invalid sections (25) if the determined lateral distance is below a predetermined threshold.

13. The method of any one of the preceding claims, wherein determining the route segments (9) comprises determining an order of the intermediate points (8) in which they are travelled along the round trip (10), wherein the order is based on a clockwise or counterclockwise order.

14. The method of any one of the preceding claims, wherein determining the route segments (9) comprises at least one of the following:- selecting the order of the intermediate points (8) based on transport line characteristics, environmental criteria, or user preferences; and- including or excluding transport lines (11) based on transport line characteristics, environmental criteria, or user preferences.

15. The method of any one of the preceding claims, wherein determining the one or more intermediate points (8) comprises at least one of the following:- receiving a user input that specifies at least one intermediate point (8);- selecting at least one point of interest from a collection of points of interest (23).

16. The method of any one of the preceding claims, wherein obtaining the origin (2) comprises obtaining a current user location, receiving a user input that specifies a desired geospatial location as the origin, or setting a default location as the origin.

17. The method of any one of the preceding claims, wherein the route parameters comprise at least one of a travel direction, a travel time, a travel distance, a type of transport line, a type of intermediate point, and energy resources available for the round trip (10).

18. The method of any one of the preceding claims, further comprising obtaining route preferences, wherein the route segments (9) are determined such that route segments (9) are preferred that fulfil one or more of the obtained route preferences, the route preferences comprising at least one of the following:- route intrinsic parameters, the route intrinsic parameters specifying characteristics of the transport lines (11); and- route extrinsic parameters, the route extrinsic parameters specifying characteristics of an environment along the transport lines (11).

19. The method of any one of the preceding claims, further comprising obtaining transport line characteristics, wherein the route segments (9) are determined such that route segments (9) are preferred based on one or more of the obtained transport line characteristics, the transport line characteristics comprising at least one of the following:- road sensor data;- traffic monitoring data;- public warning data;- aerial inspection data; and- charging availability data.

20. A data processing system being configured to perform the method of any one of claims 1 to 19.1 . A computer program or a computer program product, comprising instructions, which when executed on one or more processors of a system according to claim 20 cause the system to perform the method according to any one of claims 1 to 19.

Images