Method for navigating a motor vehicle with an at least partly automated driving functionality by an electronic computing device, a computer program product, a computer-readable storage medium, and an electronic computing device

Abstract

A method of navigating a motor vehicle (12) by an electronic computing device (44) comprising the steps of: obtaining (S1) object detection data (42) concerning a vicinity (38) of the motor vehicle; g

Description

FIELD OF THE INVENTION

[0001] The present invention relates to the field of automobiles. More specifically, the present invention relates to a method for navigating a motor vehicle with an at least partly automated driving functionality by an electronic computing device of the motor vehicle. Furthermore, the present invention relates to a corresponding computer program product, a corresponding non-transitory computer-readable storage medium, and a corresponding electronic computing device. BACKGROUND INFORMATION

[0002] According to the state of the art as exemplified in document US 202310 041 498 A1, systems and methods may be employed for assisting a driver to travel through a drive-through lane of an establishment in an autonomous mode of operation. In an example method, a processor in a vehicle determines a location of the vehicle in a drive-through lane in various ways such as, for example, based on objects located in the vicinity of the drive-through lane, based on location coordinates, and / or based on a geofence defined around the establishment. The vehicle autonomously moves through the drive-through lane in a stop-and-go mode of movement at a controlled speed while executing lane-centering and collision avoidance. SUMMARY OF THE INVENTION

[0003] It is an object of the present invention to provide automated navigation of a motor vehicle in a narrow region such as a drive-through lane through which motor vehicles may only pass in single file.

[0004] The object of the present invention is solved by a method, a corresponding computer program product, a corresponding non-transitory computer-readable storage medium, and corresponding electronic computing device according to the independent claims. Advantageous embodiments are presented in the dependent claims, the following description, and the figures.

[0005] One aspect of the invention relates to a method for navigating a motor vehicle with an at least partly automated driving functionality by an electronic computing device of the motor vehicle. Object detection data concerning a vicinity of the motor vehicle are obtained by the electronic computing device. A topographic map of the vicinity is generated by the electronic computing device. A topographic content of the topographic map is derived from the object detection data. A drivable space of the topographic map is determined by the electronic computing device according to at least a height criterion. A virtual target destination rectangle is fitted within the drivable space by the electronic computing device according to a fitting loss function of the electronic computing device. A control signal is generated by the electronic computing device for the motor vehicle to move the motor vehicle to a position in the vicinity corresponding to the virtual target destination rectangle, depending on the fit.

[0006] Examples of object detection data detection include optical sensor data, such as photographs or videos from at least one optical sensor, such as a camera, operating at at least one wavelength; acoustic sensor data such as from an ultrasonic sensor; radar data; LIDAR data; information concerning the field of view and / or range of the at least one sensor; and geometric information concerning the extent and localization of a transmitting motor vehicle itself. Examples of object detection data furthermore include combinations of previous examples, also known in the prior art as data fusion products. Additionally, examples of data relevant to the task of object detection include the results of analyses performed on sensor data, regardless of whether those sensor data are still available. For example, an object detection system of a first motor vehicle may communicate to a second motor vehicle inferences about a detected object without communicating any sensor information related to the object. In this case, the standalone inferences are still object detection data.

[0007] The term “vicinity” here includes, but is not limited to, all physical locations with bearing on current trajectory planning of the motor vehicle. In particular, the vicinity comprises all physical features relevant to the task of vehicle navigation, such as other motor vehicles, pedestrians, bicyclists, poles, curbs, street markings, signs, buildings, and traffic cones.

[0008] The topographic map is a digital structure with spatially resolved height information. The topographic map may be an intermediate product of the method, one that is not accessible by a user of the motor vehicle.

[0009] However, the topographic may, as an example, be presented to a human occupant of the motor vehicle via a display device. In this case, the topographic map may contain at least one presentational feature designed to communicate information to or improve accessibility for a human observer but that is not necessary to the execution of the method by the electronic computing device. Examples of presentational features include a color map indicating height and / or visibility levels associated with various physical features of the vicinity, a fixed orientation of the map with respect to the motor vehicle, or a dashed box on the map indicating a physical extent of the motor vehicle.

[0010] The drivable space includes any section of the vicinity determined by the electronic computing device to permit traversal by the motor vehicle, wherein traversal by the motor vehicle may not necessitate transgression of manufacturer-intended operation of the motor vehicle. For example, a rugged all-terrain road may not be determined by the electronic computing device to be a drivable space because a surface of the all-terrain road may be sufficiently uneven to present a contact risk to an undercarriage of the motor vehicle. As another example, a sidewalk may not be determined by the electronic computing device to be a drivable space because of road traffic laws in effect at a current position of the motor vehicle. Furthermore, the drivable space must be directly accessible by the motor vehicle from the current position of the motor vehicle. For example, if the motor vehicle is driving along a two-way road featuring a divider, a section of the road allocated to traffic facing in a direction opposite to one of the motor vehicle may not be determined by the electronic computing device to be a drivable space unless, for example, the motor vehicle performs a U-turn at an intersection.

[0011] The electronic computing device uses at least the height criterion in order to determine whether a space of the vicinity as represented by the topographic map may be included in the drivable space. For example, the electronic computing device may apply a height threshold according to which certain physical features are classified as too sudden and / or too large a change in elevation to be driven over. In this example, a sidewalk may be determined according to the height threshold not to represent a drivable space because a curb-street boundary associated with the sidewalk presents too large of a vertical step away from a current surface on which the motor vehicle is driving.

[0012] The drivable space may, for example, be derived by starting at a height value associated with a current surface on which the motor vehicle is situated. The derivation may then proceed by iteratively growing a drivable space radially outward from a current position of the motor vehicle. Upon each iteration, the electronic computing device may test an additional stretch of, for instance, 25 cm, extending in a particular direction from a boundary of the drivable space as updated in a most recent previous iteration of the derivation. If, for example, a detected height of the additional stretch deviates from that of the most recent previous boundary region by more than some predefined tolerance level, the derivation may stop for that particular direction and consider a different direction. For example, the derivation may discretize orientation of the motor vehicle into a predefined number of angles (e.g. 18 angles) separated from each other by e.g. twenty degrees, each of which may be considered in the manner of this derivation. In this way, regions too high or too low for manufacturer-intended traversal by the motor vehicle may be identified and excluded from the drivable space.

[0013] During this iterative derivation, the electronic computing device may require that the expansion of the drivable space not result in regions too narrow for the motor vehicle to access. For example, the electronic computing device may discard space to the left and to the right of a second motor vehicle further along a single lane road. However, the exclusion of overly narrow regions may also be accomplished during the step of fitting the virtual target destination rectangle.

[0014] Fitting the virtual target destination rectangle within the drivable space may involve a testing of at least one candidate virtual target destination rectangle by the electronic computing device. Each candidate virtual target destination rectangle is free to vary in offset (that is, position) and rotation, but must have fixed proportions corresponding to those of the motor vehicle. On the other hand, the candidate virtual target destination may also be slightly inflated, for example, by including buffer regions around its perimeter. The inclusion of these buffer regions may, for instance, facilitate identification of unfavorably cramped passages, because the buffer regions may not be contained within the drivable space.

[0015] The at least one candidate rectangle may be tested with the fitting loss function. The fitting loss function may output a loss value for each of the at least one current candidate rectangle that may be minimized in order to find a candidate rectangle most suitable for the method.

[0016] The at least one candidate rectangle may also be tested according to one or more filter, wherewith some undesired feature of the at least one candidate rectangle may result in its automatic disqualification, potentially without any loss value calculation. For example, one filter may require that all candidate rectangles be located entirely within the drivable space. This filter may, for example, automatically prevent the electronic computing device from attempting to path through an overly narrow region of the drivable space, because at least one side of the candidate rectangle will reside outside of the drivable space.

[0017] The generation and the testing of the at least one candidate rectangle may proceed in alternating fashion. For example, the electronic computing device may execute an iterative procedure in which a candidate rectangle is generated and tested before a successive candidate rectangle is generated. This procedure may, for instance, initially involve generation of candidate rectangles close to or at the position of the motor vehicle, followed by generation of candidate rectangles progressively further removed from the motor vehicle toward a navigational goal. This iteration may provide an advantage in automatically ensuring feasibility of reaching the candidate rectangle by the motor vehicle. By contrast, for instance, starting to generate candidate rectangles at the navigational goal may result in an evaluation of at least one candidate rectangle separated from a current position of the motor vehicle by non-drivable space.

[0018] The invention comprises further developments providing additional benefits.

[0019] One example of a fitting loss function penalizes candidate rectangles according to their distances from a programmed destination of the motor vehicle. By application of this fitting loss function, the electronic computing device may prioritize motion toward the programmed destination. A second example of a fitting loss function penalizes candidate rectangles according to a user ease heuristic. By application of this fitting loss function, the electronic computing device may prioritize retaining a current orientation over turning, especially when the physical extent of the region corresponding to the drivable space is narrow compared to a physical extent of the motor vehicle.

[0020] A fitting loss function may also be a composition of different fitting loss functions. For instance, the fitting loss functions described in the previous two examples may be summed in quadrature to create a fitting loss function amounting to a compromise between their two priorities.

[0021] As a consequence of application of the above method, the electronic computing device may optimize motion of the motor vehicle according to at least one navigational goal, as encoded in the fitting loss function, while satisfying constraints imposed by a topography of the vicinity. Therefore, automated navigation of the motor vehicle may be improved, fulfilling the objective as formulated earlier.

[0022] In a further development of the invention, the step of fitting the virtual target destination rectangle involves a discretization of at least part of the drivable space into at least one candidate rectangle whose virtual extent corresponds to a physical extent of the motor vehicle. In other words, each candidate rectangle generated according to this development lies entirely within the drivable space. However, at least one region of the drivable space may be excluded from a particular discretization, for instance if the region exceeds a stored maximum distance or falls below a stored minimum distance from the motor vehicle.

[0023] The positions of the candidate rectangles into which the part of the drivable space is discretized may be computed mathematically by means of, for example, a Minkowski difference algorithm, such as one involving a Minkowski difference between a rectangle bounding a current position of the motor vehicle and a polygon bounding the part of the drivable space.

[0024] Whenever application of the development results in at least two candidate rectangles, the candidate rectangles may at least partially overlap. For example, a first candidate rectangle may represent a westward translation of a second candidate rectangle by one quarter of a meter whereas an east-west extent of each rectangle corresponds to a physical distance of three meters. As a second example, the first candidate rectangle may represent a 180 degree rotation of the second candidate rectangle unaccompanied by any translation.

[0025] In a further development, the step of generating the topographic map and / or the step of fitting the virtual target destination rectangle is triggered upon a detection by the electronic computing device of a crossing by the motor vehicle over at least one geofence boundary.

[0026] A digital map used by the electronic computing device may make use of geofences to designate regions of the map as, for example, belonging to particular establishments. For instance, a drive-through lane of a fast food establishment may be geofenced order to group the lane with a primary building of the fast food establishment. By detecting geofences, a motor vehicle with an at least partly automated driving functionality may, for example, automatically switch to a different self-driving navigation procedure, such as one intended for navigation of a freeway.

[0027] In a further development, the steps of the method are performed in a continuous loop until the loop is terminated by the electronic computing device upon detection of a satisfaction of a stored loop completion criterion.

[0028] For example, if the electronic computing device and / or another onboard navigational system of the motor vehicle is programmed with a particular destination, the stored loop completion criterion may be arrival of the motor vehicle at the particular destination. Upon arrival, self-driving may terminate.

[0029] Alternatively, if the method is configured to be used by the electronic computing device alongside at least one other autonomous-driving navigation procedure, detection of the satisfaction of the stored loop completion criterion may trigger a switch to the at least one other self-driving navigation procedure. One example of a stored loop completion criterion appropriate to this case may be a driving environment criterion. Specifically, for example, if the electronic computing device is configured to use the method of the present invention for navigation of the motor vehicle in a narrow and / or slow-traffic region such as a drive-through area, the stored loop completion criterion may be a detection that the motor vehicle has exited the drive-through area.

[0030] In a further development, at least some of the object detection data are received by the electronic computing device from a map information server.

[0031] For example, the map information server may host at least one topographic map that may be downloaded by the electronic computing device. This development may present an advantage in providing the electronic computing device with object detection data that may not be perceptible by at least one sensor of the motor vehicle at a particular moment, for example if a line of sight of the at least one sensor to a particular object is blocked by another motor vehicle.

[0032] In a further development, at least some of the object detection data are received by the electronic computing device from at least one other motor vehicle in the vicinity.

[0033] For example, the at least one other motor vehicle in the vicinity may share sensor data and / or sensor data analysis products with the motor vehicle. This development may be advantageous for a similar reason as that of a previously mentioned development. In particular, with a greater amount of sensor data available to the electronic computing device, object detection and / or classification accuracy may improve.

[0034] In a further development, the method comprises the additional step of generating a control signal by the electronic computing device to complete at least one stop at at least one interaction point. The term “interaction point” is used here to refer to a location at which a human occupant of the motor vehicle interacts with at least one entity that is external to the motor vehicle and that is located at the interaction point.

[0035] For instance, a drive-through area of a fast food establishment may contain multiple interaction points. Examples include microphones at which to place orders, windows at which to pay for the orders, and windows at which to collect food associated with the orders. This development may be advantageous in allowing a user of the motor vehicle to pause the navigation method without having to take over navigation of the motor vehicle. An interaction point may be automatically detected, for example if the object detection system of the motor vehicle detects a payment window. Alternatively or in addition, and interaction point may be specified by the user. For example, the user may specify a stop at a particular gas station during a road trip spanning several hours.

[0036] In particular, the method is a computer-implemented method. Therefore, one aspect of the invention relates to a computer program product consisting of computer-readable instructions for performing a method according to the preceding aspect.

[0037] One aspect of the invention relates to a non-transitory computer-readable storage medium comprising the computer program product according to the preceding aspect.

[0038] One aspect of the invention relates to an electronic computing device of a motor vehicle, wherein the electronic computing device is configured for performing a method according to the preceding aspect. In particular, the method is performed by the electronic computing device.

[0039] An “electronic computing device” may in particular be understood as a data processing device, which comprises processing circuitry. The electronic computing device may therefore in particular process data to perform computing operations. This may also include operations to perform indexed accesses to a data structure, for example a look-up table, LUT.

[0040] In particular, the electronic computing device may include one or more computers, one or more microcontrollers, and / or one or more integrated circuits, for example, one or more application-specific integrated circuits, ASIC, one or more field-programmable gate arrays, FPGA, and / or one or more systems on a chip, SoC. The electronic computing device may also include one or more processors, for example one or more microprocessors, one or more central processing units, CPU, one or more graphics processing units, GPU, and / or one or more signal processors, in particular one or more digital signal processors, DSP or neural processing units, NPU. The electronic computing device may also include a physical or a virtual cluster of computers or other of said units.

[0041] In various embodiments, the electronic computing device includes one or more hardware and / or software interfaces and / or one or more memory units.

[0042] A memory unit may be implemented as a volatile data memory, for example a dynamic random access memory, DRAM, or a static random access memory, SRAM, or as a non-volatile data memory, for example a read-only memory, ROM, a programmable read-only memory, PROM, an erasable programmable read-only memory, EPROM, an electrically erasable programmable read-only memory, EEPROM, a flash memory or flash EEPROM, a ferroelectric random access memory, FRAM, a magnetoresistive random access memory, MRAM, or a phase-change random access memory, PCRAM.

[0043] Furthermore, the present invention relates to a vehicle comprising the electronic computing device according to the preceding aspect.

[0044] Further advantages, features, and details of the invention derive from the following description of preferred embodiments as well as from the drawings. The features and feature combinations previously mentioned in the description as well as the features and feature combinations mentioned in the following description of the figures and / or shown in the figures alone can be employed not only in the respectively indicated combination but also in any other combination or taken alone without leaving the scope of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The novel features and characteristics of the disclosure are set forth in the appended claims. The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate exemplary embodiments and together with the description, serve to explain the disclosed principles. The same numbers are used throughout the figures to reference like features and components. Some embodiments of system and / or methods in accordance with embodiments of the present subject matter are now described below, by way of example only, and with reference to the accompanying figures.

[0046] The drawings show in:

[0047] Fig. 1 an example of a topographic map that may be generated during an execution of the method;

[0048] Fig. 2 a three-part schematic top-down view of motion of a motor vehicle governed by an electronic computing device performing an embodiment of the method;

[0049] Fig. 3 a schematic top-down view of a motor vehicle comprising an electronic computing device according to an aspect of the invention;

[0050] Fig. 4 a companion view to Fig. 3, outlining various regions referred to during execution of an embodiment of the method;

[0051] Fig. 5 a companion view to Fig. 3 and Fig. 4, giving an example of a discretization of a drivable space into multiple candidate virtual target destination rectangles; and

[0052] Fig. 6 a schematic flowchart according to an embodiment of the method.

[0053] In the figures, the same elements or elements having the same function are indicated by the same reference signs. DETAILED DESCRIPTION

[0054] In the following detailed description of the example embodiments of the disclosure, references are made to the accompanying drawings that form parts hereof and in which are shown by way of illustration specific embodiments in which the disclosure may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the disclosure, and it is to be understood that other embodiments may be utilized and that changes may be made without departing from the scope of the present disclosure. The following description is, therefore, not to be taken in a limiting sense.

[0055] While the disclosure is susceptible to various modifications and alternative forms, specific embodiments thereof have been shown by way of example in the drawings and have been described in detail below. It should be understood, however, that it is not intended to limit the disclosure to the particular forms disclosed, but on the contrary, the disclosure is to cover all modifications, equivalents, and alternatives falling within the scope of the disclosure.

[0056] Fig 1 shows an example of a topographic map 10 that may be generated during an execution of the method. A rectangular outline at the center indicates a motor vehicle 12 comprising an electronic computing device that generates the topographic map 10 but that is not shown here. The motor vehicle 12 may currently be positioned along a flat, slightly curving road that may, in this example, represent a drivable space 14.

[0057] The topographic map 10 may, in this example, indicate variations in surface altitude by means of a series of contour lines 16, 18, 20, 22, 24, 26. The two lines 16 bounding the drivable space 14 may, in this example, correspond to a curb representing a six inch elevation difference compared to the drivable space 14. In addition, the lines 18 and 20 may, in this example, represent elevation differences associated with physical features of a drive-through area of a fast food establishment. Examples of such features include menu signs, buildings, and trash cans.

[0058] On the other side of the motor vehicle 12, the contour lines 22, 24, and 26 may represent various altitudes along a sloping hill feature. The density of the contour lines 22, 24, and 26 across the feature may be given by a stored spacing rule. For example, the electronic computing device may automatically place a new contour line for each elevation change amounting to six inches compared to a previous contour line.

[0059] In the example of Fig. 2, the steps of the method are performed in a continuous loop until the loop is terminated by the electronic computing device upon detection of a satisfaction of a stored loop completion criterion. Specifically, in Fig. 2, the topographic maps 10 from three successive executions of the method are shown, where Fig. 2a describes a first time step and Fig. 2c describes a third time step.

[0060] At the center of each topographic map 10 is the rectangular outline that, as in Fig. 1, indicates the motor vehicle 12, which in this example will also be referred to as the first motor vehicle 12. In this example, the first motor vehicle 12 may be navigating through a drive-through area of a pharmacy.

[0061] In this example, the topographic maps 10 may be simplified compared to that of Fig. 1 in the sense that only two topographic levels may be indicated: the drivable space 14 and the non-drivable space 28. In this example, the definition of the drivable space 14 may include spaces, such as those at the top of Fig. 2a, that are nevertheless unreachable by the motor vehicle 12 on account of their narrowness. Unreachably narrow spaces may, in this example, be automatically excluded during a later step, such as while fitting a virtual target destination rectangle within the drivable space 14.

[0062] Some sections of the non-drivable space 28 may be dynamic, changing in extent and / or position relative to the first motor vehicle 12 when it is stationary. For example, a rectangle at the top of Fig. 2a may, in this example, correspond to a second motor vehicle 30. In Fig. 2b, motion of the second motor vehicle 30 may be appreciated as the corresponding rectangle moves relative to a stationary feature 32, here a pair of protrusions.

[0063] In each topographic map 10, a virtual target destination rectangle 34 is indicated by a second outline 34. The virtual target destination rectangle 34 represents a planned navigational step for the electronic computing device controlling the motor vehicle 10 in moving toward a longer-term navigational goal. In this example, the longer-term navigational goal may be a point of egress from the drive-through area. Fig. 2b shows the motor vehicle 10 having moved to approximately where the destination rectangle 34 of Fig. 2a was, and similarly Fig. 2c shows the motor vehicle 10 having moved to approximately where the destination rectangle 34 of Fig. 2b was.

[0064] In Fig. 3, the first motor vehicle 12 is shown queuing behind the second motor vehicle 30 in a drive-through area of a fast food establishment 36. A vicinity 38 of the first motor vehicle 12 is outlined. In this example, an optical sensor 40, here a camera, of the first motor vehicle 12 may capture object detection data 42 from which the electronic computing device 44 may determine the vicinity 38. Alternatively or in addition, the vicinity 38 may be defined using, for example, a location of the first motor vehicle 12, for example as encoded in global positioning system coordinates (GPS coordinates), within a map of road and terrain features.

[0065] In this example, at least some of the object detection data 42 may be received by the electronic computing device 44 from a map information server 46. In this example, the object detection data 42 may be a set of polygons describing the drivable space 14 and the non-drivable space 28 of the drive-through area and the map information server 46 may be a manufacturer-provided backend database accessed automatically by a network device, such as a Wi-Fi card, of the electronic computing device 44.

[0066] For simplicity, the map information server 46 is illustrated close to the first motor vehicle 12. However, the map information server 46 is not necessarily physically situated within the vicinity 38. For example, the map information server 46 may be located at a headquarters of a manufacturer of the first motor vehicle 12 and may communicate with the first motor vehicle 12 via cell tower.

[0067] In this example, at least some of the object detection data 42 may be received by the electronic computing device 44 from at least one other motor vehicle in the vicinity 38. In this example, the electronic computing device 44 may receive some object detection data 42 from the second motor vehicle 30 concerning an area of the vicinity 38 not currently perceptible by the optical sensor 40, such as an area directly in front of the second motor vehicle 30. In this example, the second motor vehicle 30 may transmit at least some of the object detection data 42 over a cell network.

[0068] Using the object detection data 42, in this example from the optical sensor 40, the electronic computing device 44 generates the topographic map 10. Furthermore, the electronic computing device 44 determines the drivable space 14 according to at least a height criterion 48, in this example one stored in the electronic computing device 44.

[0069] The drivable space 14 in which the first motor vehicle 12 is queuing is bounded by a pair of curbs 50. The curbs may be identified by the electronic computing device 44 as representing regions of the non-drivable space 28. However, in this example, the vicinity 38 may be defined so as to extend, for example, twenty meters beyond a closest approach of non-drivable space 28, in order to monitor for bikers and pedestrians who may be approaching the drivable space 14.

[0070] The electronic computing device 44 fits the virtual target destination rectangle 34 within the drivable space 14 according to a fitting loss function 52 of the electronic computing device 44. The electronic computing device 44 generates a control signal, here referred to as a first control signal 54, to move the first motor vehicle 12 to a position in the vicinity 38 corresponding to the virtual target destination rectangle 34.

[0071] In this example, the fitting loss function 52 may prioritize proximity to at least one interaction point 56, 58. Specifically, in this example, the fast food establishment 36 may be a sufficiently high-volume fast food establishment 36 that two customer interaction windows 56, 58 are staffed, a first window 56 for accepting payments and a second window 58 for giving out ordered food. The fitting loss function 52 may therefore result in preferential selection, by the electronic computing device 44 and at an instant illustrated in Fig. 2, of a virtual target destination rectangle candidate that is not only further down the road but also further to the left.

[0072] In this example, the electronic computing device 44 may generate a stopping control signal 60 to complete at least one stop at at least one interaction point 56, 58. At the instant illustrated in Fig. 3, the electronic computing device 44 may in particular have generated the stopping control signal 60 to pause at the first interaction point 56 until, for example, an occupant of the first motor vehicle 12 interacts with a resume button of the electronic computing device 44.

[0073] Fig. 4 shows a companion view to Fig. 3 including various regions defined for the method. In this example, the drivable space 14 may correspond to a section of a road of the drive-through area, wherein the section may start close to a bumper of the second motor vehicle 30 and extend upward out of view. In this example, the drivable space 14 may not include regions too narrow to permit traversal by the first motor vehicle 12.

[0074] The virtual target destination rectangle 34 selected after application of the fitting loss function 52 by the electronic computing device 44 to various candidate virtual target destination rectangles is shown near the first interaction point 56. As part of the first control signal 54, the first motor vehicle 12 may be controlled in various ways, such as braking, turning, and activating of an acceleration mechanism.

[0075] In this example, the step of generating the topographic map 10 and / or the step of fitting the virtual target destination rectangle 34 is triggered upon a detection by the electronic computing device 44 of a crossing by the first motor vehicle 12 over at least one geofence 62 boundary. In this example, the geofence 62 may encompass the entire fast food establishment 36, including the drive-through area.

[0076] In this example, the geofence 62 may be specified as a range of global positioning coordinates (GPS coordinates) downloaded by the electronic computing device 44 from the map information server 46. The electronic computing device 44 may, in this example, detect a crossing of a boundary of the geofence 62 by comparing GPS coordinates of the first motor vehicle 12 with the range associated with the geofence 62.

[0077] Fig. 5 shows an example of how the electronic computing device 44, in the situation portrayed in Fig. 3 and Fig. 4, may discretize the drivable space 14 into candidate virtual target destination rectangles. In this illustration, for the sake of simplicity, each candidate rectangle is visualized by a point corresponding to a center of a candidate rectangle as well as an arrow corresponding to an orientation of that candidate rectangle. Each arrow corresponds to the direction in which the front of the first motor vehicle 12 may point if the first motor vehicle 12 were positioned according to that candidate rectangle. Therefore, the points do not extend all the way to the edges of the drivable space 14, because points near the edges of the drivable space 14 would correspond to candidate rectangles not entirely contained by the drivable space 14.

[0078] In this example, the electronic computing device 44 may attempt to conserve computational resources by sampling regions of the drivable space 14 at a density proportional to proximity to the first interaction point 56. For instance, there are more candidate rectangles in the upper left area of Fig. 5 than there are in any other area of Fig. 5.

[0079] Figure 6 shows a schematic flowchart describing an exemplary control flow of the method.

[0080] In a first step S1, the electronic computing device 44 obtains the object detection data 42 from at least one sensor. Examples of sensors include optical sensors such as cameras, LIDAR sensors, and radar sensors.

[0081] In a second step S2, the electronic computing device 44 generates the topographic map 10, here referred to as the local topographic map 10. In this example, the electronic computing device 44 may identify interaction points 56, 58 as part of a labeling step in the generation of the local topographic map 10.

[0082] In a first part S2A of the second step S2, the electronic computing device 44 perceives obstacles in the vicinity 38 by computing heights of various regions of the vicinity 38 from the object detection data 42. In a second part S2B of the second step S2, the electronic computing device 44 determines the drivable space 14 according to at least the height criterion 48, in this example by detecting at least one lane along which the first motor vehicle 12 may drive.

[0083] In a third part S2C of the second step S2, the electronic computing device 44 may, in this example, identify three interaction points. In addition to the payment window 56 and the food pick-up window 58 introduced in a previous example, the electronic computing device 44 may further identify a location at which the occupant of the first motor vehicle 12 may place an order for specific food items. This location may, for example, feature a microphone positioned near a menu sign.

[0084] In a fourth part S2D of the second step S2, the electronic computing device 44 may, in this example, perform an additional computer vision step of detecting and reading at least one sign in the vicinity 38. For example, the at least one sign may be a drive-through speed limit sign indicating a maximum permissible speed of eight km / h. The electronic computing device 44 may then, in this example, incorporate this speed limit in its motion planning.

[0085] In a third step S3, at least some of the object detection data 42 may be received by the electronic computing device 44 from a map information server 46. Specifically, in this example, the electronic computing device 44 may download an auxiliary topographic map from the map information server 46. The electronic computing device 44 may then proceed to compare the auxiliary topographic map to the local topographic map 10 in order to identify potential blind spots and inaccuracies affecting the local topographic map 10.

[0086] In a fourth step S4, the electronic computing device 44 plans the path of the first motor vehicle 12 by fitting the virtual target destination rectangle 34 within the drivable space 14 according to the fitting loss function 52.

[0087] In a fifth step S5, the electronic computing device 44 generates the first control signal 54 for the first motor vehicle 12 to move the first motor vehicle 12 to a position in the vicinity 38 corresponding to the virtual target destination rectangle 34, depending on the fit.

[0088] In a sixth step S6, the electronic computing device 44 may generate a stopping control signal 60 by the electronic computing device 44 to complete at least one stop at the at least one interaction point 56, 58.

[0089] In a seventh step S7, the electronic computing device 44 may check for a satisfaction of a stored loop completion criterion. If the electronic computing device 44 does not find that the stored loop completion criterion is satisfied, the electronic computing device 44 may commence a new iteration of the method. Otherwise, the method may, in this example, terminate.

[0090] In this example, the seventh step S7 may be placed at the end, but in a different example the electronic computing device 44 may also perform the check at any other point in the method. In addition, in a different example, the electronic computing device 44 may perform multiple checks throughout the method for satisfaction of the stored loop completion criterion.

[0091] Taken together, the examples show how a procedure for iteratively evaluating virtual spots for drive-through assist path planning may be performed.

[0092] In the present document, the word "exemplary" is used herein to mean "serving as an example, instance, or illustration." Any embodiment or implementation of the present subject matter described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments.

[0093] The terms “comprises,” “comprising,” or any other variations thereof, are intended to cover a non-exclusive inclusion so that a setup, device or method that comprises a list of components or steps does not include only those components or steps but may include other components or steps not expressly listed or inherent to such setup or device or method. In other words, one or more elements in a system or apparatus preceded by “comprises” or “comprise” does not or do not, without more constraints, preclude the existence of other elements or additional elements in the system or method. Reference Signs topographic map (first) motor vehicle drivable space contour lines contour lines contour lines contour lines contour lines contour lines non-drivable space second motor vehicle stationary feature virtual target destination rectangle fast food establishment vicinity optical sensor object detection data electronic computing device map information server height criterion curbs fitting loss function first control signal first interaction point second interaction point stopping control signal geofence S1-S7 steps of the method

Claims

1. A method for navigating a motor vehicle (12) with an at least partly automated driving functionality by an electronic computing device (44) of the motor vehicle (12), comprising the steps of:- obtaining (S1) object detection data (42) concerning a vicinity (38) of the motor vehicle (12) by the electronic computing device (44);- generating (S2A) a topographic map (10) of the vicinity (38) by the electronic computing device (44), wherein a topographic content of the topographic map (10) is derived from the object detection data (42);- determining (S2B) a drivable space (14) of the topographic map (10) by the electronic computing device (44) according to at least a height criterion (48);- fitting (S4) a virtual target destination rectangle (34) within the drivable space (14) by the electronic computing device (44) according to a fitting loss function (52) of the electronic computing device (44); and- generating (S5) a control signal (54) by the electronic computing device (44) for the motor vehicle (12) to move the motor vehicle (12) to a position in the vicinity (38) corresponding to the virtual target destination rectangle (34), depending on the fit.

2. The method according to claim 1, whereinthe step of fitting (S4) the virtual target destination rectangle (34) involves a discretization of at least part of the drivable space (14) into at least one candidate rectangle whose virtual extent corresponds to a physical extent of the motor vehicle (12).

3. The method according to any one of the preceding claims, whereinthe step of generating the topographic map (10) and / or the step of fitting the virtual target destination rectangle (34) is triggered upon a detection by the electronic computing device (44) of a crossing by the motor vehicle (12) over at least one geofence (62) boundary.

4. The method according to any one of the preceding claims, characterized in that the steps of the method are performed in a continuous loop until the loop is terminated by the electronic computing device (44) upon detection (S7) of a satisfaction of a stored loop completion criterion.

5. The method according to any one of the preceding claims, whereinat least some of the object detection data (42) are received (S3) by the electronic computing device (44) from a map information server (46).

6. The method according to any one of the preceding claims, wherein at least some of the object detection data (42) is received by the electronic computing device (44) from at least one other motor vehicle (30) in the vicinity (38).

7. The method according to any one of the preceding claims, comprising the additional step ofgenerating (S6) a stopping control signal (60) by the electronic computing device (44) to complete at least one stop at at least one interaction point (56, 58), that is, is a location at which a human occupant of the motor vehicle (12) interacts with at least one entity that is external to the motor vehicle (12) and that is located at the interaction point (56, 58).

8. A computer program product containing computer-readable instructions for performing a method according to any one of claims 1 to 7.

9. A non-transitory computer-readable storage medium comprising the computer program product according to claim 8.

10. An electronic computing device (44) of a motor vehicle (12), wherein the electronic computing device (44) is configured to perform a method according to any one of claims 1 to 7.

Images