EP4208815A1 - Verfahren und vorrichtung zur umfeldwahrnehmung eines umfeldes eines zumindest teilautomatisiert fahrenden fahrzeugs - Google Patents
Verfahren und vorrichtung zur umfeldwahrnehmung eines umfeldes eines zumindest teilautomatisiert fahrenden fahrzeugsInfo
- Publication number
- EP4208815A1 EP4208815A1 EP21745257.2A EP21745257A EP4208815A1 EP 4208815 A1 EP4208815 A1 EP 4208815A1 EP 21745257 A EP21745257 A EP 21745257A EP 4208815 A1 EP4208815 A1 EP 4208815A1
- Authority
- EP
- European Patent Office
- Prior art keywords
- resolution
- sensor data
- evaluation
- vehicle
- focus region
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/254—Fusion techniques of classification results, e.g. of results related to same input data
- G06F18/256—Fusion techniques of classification results, e.g. of results related to same input data of results relating to different input data, e.g. multimodal recognition
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
Definitions
- the invention relates to a method and a device for perceiving the surroundings of a vehicle that is driving at least partially automatically.
- the invention is based on the object of improving a method and a device for perceiving the surroundings of a vehicle that is driving at least partially automatically, in particular with regard to the required computing power and accuracy.
- a method for perceiving the surroundings of an environment of an at least partially automated vehicle is made available, with the surroundings of the vehicle being recorded by means of at least one sensor, with sensor data recorded by the at least one sensor being evaluated by means of an evaluation device using at least one evaluation method in a first resolution, and wherein the detected sensor data is evaluated in at least one defined focus region by the evaluation device using the at least one evaluation method in a second resolution be, wherein the first resolution is smaller than the second resolution, and wherein the evaluation results are combined and output.
- a device for perceiving the surroundings of an at least partially automated vehicle comprising an evaluation device, wherein the evaluation device is set up to evaluate sensor data recorded by at least one sensor by at least one evaluation method in a first resolution, and the recorded sensor data by the at least to evaluate an evaluation method in at least one defined focus region in a second resolution, the first resolution being smaller than the second resolution, and to combine and output the evaluation results.
- the method and the device enable a compromise between the computing power required when evaluating the recorded sensor data and the accuracy of the evaluation. This is achieved by evaluating the sensor data in a first resolution that is lower than the original resolution.
- at least one focus region is defined in the captured sensor data, ie the scope of the sensor data is reduced compared to the original sensor data. For example, a detail in a two-dimensional camera image can form such a focus region.
- the sensor data are evaluated with a second resolution.
- the first resolution is smaller than the second resolution.
- the second resolution corresponds in particular to the original, that is to say the full, resolution of the recorded sensor data.
- the recording sensor data is evaluated overall with a lower resolution than the original resolution and only the at least one focus region is evaluated with a higher, in particular full, resolution, the computing power required for the evaluation can be reduced overall and still have a high level in the at least one focus region Accuracy can be achieved when evaluating.
- a sensor is in particular a camera that captures the surroundings of the vehicle.
- a sensor can also be a lidar sensor, a radar sensor or an ultrasonic sensor.
- the recorded sensor data can be one-dimensional or multi-dimensional.
- the recorded sensor data is two-dimensional.
- the captured sensor data are, in particular, captured camera images.
- the recorded sensor data can also be point clouds from a lidar sensor or a radar sensor or ultrasonic data.
- An evaluation method is in particular an evaluation method for the perception and/or interpretation of the environment, in which at least one perception function is carried out.
- Such a perception function can, for example, be one of the following: object recognition, determination of bounding boxes, semantic segmentation, instance segmentation or panoptical segmentation, etc.
- the evaluation method is provided and implemented in particular by means of a machine learning method.
- (trained) neural networks are used.
- a first trained neural network is provided by the evaluation device, which processes the recorded sensor data in the lower first resolution and executes one or more of the above evaluation methods.
- a second trained neural network which is provided by the evaluation device, evaluates the recorded sensor data in the at least one focus region in the higher, in particular in the original, second resolution.
- the evaluation method can also be provided and carried out in a different way and/or with other means, which means that classic methods of sensor data evaluation can also be used.
- the recorded sensor data is calculated down to the first resolution. This is done using methods known per se. For example, in the case of camera images as captured sensor data, a number of picture elements (pixels) is reduced so that the image is scaled down to a lower total number of picture elements.
- the detected sensor data is correspondingly cropped to a section so that only the sensor data within the section of the at least one focus region is supplied as input data to the evaluation method, for example the second neural network. Provision can be made here for the second neural network to be instantiated again or provided and executed separately for each defined focus region.
- Parts of the device in particular the evaluation device and its components, can be designed individually or combined as a combination of hardware and software, for example as program code on a microcontroller or microprocessor is running. However, it can also be provided that parts are designed individually or combined as an application-specific integrated circuit (ASIC).
- ASIC application-specific integrated circuit
- a vehicle is in particular a motor vehicle.
- a vehicle can also be any other land, rail, water, air or space vehicle.
- At least one focal region is defined by means of a rule-based system depending on a position of the vehicle and/or a planned trajectory of the vehicle and/or a context of the surroundings.
- the at least one focus region can be directed specifically to particularly relevant areas in the area surrounding the vehicle.
- it can be provided that, depending on the position, particularly relevant areas in the environment are marked or stored in an environment map.
- an environment map For example, it can be stored in the environment map that there is a school near the current position of the vehicle. Based on this information, a focus region can be determined, for example, in such a way that it covers a roadside and a sidewalk in front of the school.
- a focus region can also be defined depending on the planned trajectory. For example, it can be provided that the planned trajectory provides for the vehicle turning left and crossing oncoming traffic. Based on the planned trajectory, a focus region can then be placed on the oncoming lane in order to improve the perception of approaching vehicles, i.e. with a higher resolution.
- An environment context can be derived both from detected sensor data and from an environment map, for example in the case of a game street, in particular to better perceive the areas on the game street and the edge areas via defined focus regions.
- links are stored in the rule-based system that link a position of the vehicle and/or a planned trajectory or a type or class of the planned trajectory (turning left, turning right, overtaking, etc.) and/or an environmental context with at least one focus region.
- the at least one focus region can be defined or defined, for example, via position information relative to the vehicle and/or in relation to the recorded sensor data.
- a region of focus can be specified by specifying pixel areas in captured camera images.
- it can be provided as an output link that a focus region covers an area around the middle of a road ahead.
- further focus regions can then be included Depending on set rules (e.g. school ahead, zebra crossing ahead, cycle path crossing etc.).
- the rule-based system can be provided, for example, by means of a logic module of the device set up for this purpose.
- changes over time in the captured sensor data are detected, with at least one focus region being defined as a function of at least one detected change over time in the captured sensor data.
- a focus region can be defined or newly created if something in the environment changes over time.
- a change over time should mean in particular that in the case of sensor data recorded consecutively in time, for example consecutive camera images (“frames”), these are different at least in partial areas. For example, in successive camera images, an imaged traffic light may change from a green signal to a red signal. Since changes represent a potential danger for the vehicle (or other road users who cause the change) and are therefore of greater interest in the perception of the environment, focus regions can be defined dynamically depending on the environment or a change in status in the environment.
- movements in the surroundings of the vehicle are detected in the acquired sensor data, with at least one focus region being defined as a function of at least one movement detected in the acquired sensor data.
- sensor data that depict moving objects in the environment can be evaluated with a higher resolution.
- the underlying idea is that moving objects in the environment are potentially more dangerous than static objects in the environment and therefore require more attention and greater accuracy in the evaluation.
- At least one focus region is defined by means of a machine learning method.
- a neural network is trained so that it estimates areas in which focus regions are defined depending on the recorded sensor data and/or a position of the vehicle and/or an environmental context (e.g. play street, freeway, zebra crossing, underground car park, etc.). should be.
- focus regions can also be estimated and defined for unknown environments.
- an uncertainty of the evaluation results estimated by the evaluation method is taken into account. As a result, the evaluation results can each be weighted with the estimated uncertainty.
- an uncertainty can be specified for each pixel, with which the assignment can be evaluated.
- a pixel-by-pixel assignment and associated specification of the uncertainty can be done, for example, by means of a neural network.
- An assignment with associated uncertainties for the entire camera image in a lower first resolution and an assignment with associated uncertainties for the (at least one) focus region in a second, higher resolution are then available as evaluation results.
- the assignments can be combined, weighted according to the respective uncertainties, in particular summed up.
- an assignment of the image elements to objects (or object classes) in the edge areas of the focus region is subject to greater uncertainty than in areas further inwards, since less information is available at the edges (since information beyond the cut-off edges is missing ).
- a greater weight can then be placed on the evaluation results of the evaluation in the first, lower resolution, weighted by the respective associated uncertainties, so that an overall result in the edge areas of the focus region is improved overall.
- the focus regions are assigned or have been assigned a priority, with the focus regions being processed in the order of the assigned priorities when evaluating the respectively associated sensor data.
- the priority can be assigned, for example, as a function of an object class assigned to an object as part of the environment perception. For example, the vulnerability of other road users can be prioritized, with pedestrians and cyclists being given a higher priority than other vehicles. Provision can also be made, for example, for children to be given a higher priority than adults, since the behavior of children is generally less predictable than the behavior of adults. This makes it possible to concentrate limited computing power on the focus regions with the highest priority.
- an evaluation result is retained in the second resolution at least for a predetermined period of time.
- the evaluation results generated for the higher second resolution can continue to be used at least in the specified period of time and therefore increase accuracy despite the abandonment of this focus region in the perception of the surroundings.
- Abandoning a focus region is intended to mean in particular that the sensor data in the abandoned focus region are no longer evaluated in the second, higher resolution, but only in the first, lower resolution.
- the specified period of time can, in particular, be a few hundred milliseconds to seconds.
- a vehicle comprising at least one device according to one of the described embodiments.
- FIG. 1 shows a schematic representation of an embodiment of the device for perceiving the surroundings of an at least partially automated vehicle
- 3a shows a schematic representation to clarify a procedure according to the method described in this disclosure
- 3b shows a schematic representation of a section of sensor data belonging to a defined focus region
- 3c shows a schematic representation of reduced-resolution sensor data
- 4 shows a schematic representation to clarify an overall result combined from the evaluation results according to the method.
- Device 1 shows a schematic illustration of an embodiment of the device 1 for perceiving the surroundings of an at least partially automated vehicle 50 .
- Device 1 carries out the method described in this disclosure for perceiving the surroundings of a vehicle 50 that is driving at least partially automatically.
- the device 1 comprises an evaluation device 2 .
- the evaluation device 2 has a resolution reduction module 3 , a cutting module 4 , a first neural network 5 , a second neural network 6 and a merging module 7 .
- Parts of the device 1, in particular the evaluation device 2 and its components, can be designed individually or combined as a combination of hardware and software, for example as program code that is executed on a microcontroller or microprocessor.
- Detected sensor data 10 of at least one sensor 51 of vehicle 50 is fed to device 1 .
- the sensor 51 is in particular a camera and the sensor data 10 are in particular camera images captured from the surroundings.
- a resolution of the captured sensor data 10, in particular the captured camera images, is reduced to a first resolution by the resolution reduction module 3, and reduced-resolution sensor data 10- are provided.
- the cutting module 4 cuts out a section 9 from the recorded sensor data 10, in particular from the camera images, as a function of a defined focus region 8.
- a second resolution of section 9 corresponds to a resolution of original sensor data 10. The first resolution is therefore smaller than the second resolution.
- the sensor data 10 in the first resolution are fed to the first neural network 5 .
- the section 9 in the second resolution is fed to the second neural network 6 .
- the first neural network 5 and the second neural network 6 carry out the same evaluation method, in particular a perception function, for example a semantic segmentation, in which an object class is assigned to individual image elements in the captured camera images, i.e. for each image element it is estimated which object the picture element depicts in each case.
- Evaluation results 11 , 12 are output by the neural networks 5 , 6 and fed to the merging module 7 .
- the procedure is described using neural networks 5, 6, other methods of machine learning or classic sensor data evaluation can also be used to convert the sensor data 10 in the first resolution and the section 9 associated with the focus region 8 from the sensor data 10 in the second resolution evaluate.
- the evaluation method i.e. the perception function for the perception or interpretation of the environment (e.g. object recognition, semantic segmentation, semantic instance segmentation, bounding boxes, etc.) is the same for both ways.
- the combination module 7 combines the evaluation results 11 , 12 into an overall result 13 .
- the evaluation result 11 is upscaled to the original resolution (which corresponds to the second resolution), for example, and the evaluation result 12 for section 9 is inserted into the upscaled evaluation result 11 at the correct position, pixel by pixel.
- it can be provided in particular that an average value of the evaluation results 11, 12 is formed.
- the neural networks 11, 12 estimate the evaluation results 11, 12, for example in the form of object classes assigned to individual picture elements, and at the same time additionally supply a respective uncertainty for the estimates per picture element.
- a weighting factor for merging can be defined or can be defined via the respectively estimated uncertainty.
- the overall result 13 can be improved in this way, particularly at the edges of the section 9, where estimates of the second neural network 6 have greater uncertainty due to a lack of additional information from areas beyond the edge, since the evaluation result 11 of the first neuron is in these areas Network 5 can be considered more weighted.
- the device 1 can have, for example, a logic module 19 in which the rule-based system 18 is made available and used.
- a focus region 8 can be placed on oncoming traffic (eg when turning and crossing an oncoming lane).
- An environmental context can, for example, relate to a play street, a settlement area with families with small children or a deer crossing, etc., with focus regions 8 then being set, for example, to areas near vehicles parked at the edge of the play street, to oncoming traffic or to the edge of the road adjacent to the forest will.
- the position 53 and the planned trajectory 54 are supplied to the logic module 19 by a navigation system and/or a vehicle controller (not shown) of the vehicle 50, for example.
- the device 1 can have, for example, a change detection module 14, which is used to detect areas in the recorded sensor data 10 in which a change (e.g. a change in a traffic light phase or a blinking indicator of another vehicle, etc.) is detected.
- a change e.g. a change in a traffic light phase or a blinking indicator of another vehicle, etc.
- the device 1 can have a motion detection module 15, for example, which detects movements in the captured sensor data 10, in particular captured camera images, and defines an associated area, in particular image element area in at least one camera image, as the focus region 8.
- another neural network 16 can be trained to estimate and define focus regions 8 as a function of captured sensor data 10, in particular captured camera images, and/or other information (an environmental context and/or a position 53 of vehicle 50 etc.).
- the priorities 17 can be selected, for example, depending on the vulnerability of another road user included in the focus region 8 .
- a prioritization module (not shown) of the device 1 can be provided for this purpose, for example.
- an evaluation result 12 is retained in the second resolution at least for a predetermined period of time. This takes place, for example, in the merging module 7.
- the device 1 makes it possible to find a compromise when carrying out the evaluation method with regard to the required computing power and accuracy.
- computing power can be reduced, since the sensor data 10 recorded are processed completely only in a reduced resolution, relevant areas can be evaluated with a higher resolution, in particular with the original resolution, by defining focus regions 8 .
- the overall result 13 can, for example, be fed to a maneuver planner 52 of the vehicle 50 and taken into account there in the maneuver planning for automated driving.
- FIG. 2a shows a schematic illustration to clarify the method.
- Captured sensor data 10 is shown in the form of a camera image.
- 2b shows a semantic segmentation as an evaluation result 11, in which an object class (indicated as different hatching) is assigned to each picture element of the camera image.
- the sensor data 10 shown in FIG. 2a ie the camera image
- a child 30 standing at the end of the row of vehicles visible on the right in the camera image is not detected by the semantic segmentation due to the reduced resolution and therefore does not appear in the evaluation result 11 .
- FIG. 3a The procedure according to the method described in this disclosure is shown in FIG. 3a.
- the captured sensor data 10 that is, in the camera image (At least) one focus region 8 is defined.
- a simple case is shown, in which the focus region 8 encompasses the middle of the road ahead.
- a section 9 (FIG. 3b) corresponding to the focus region 8 is evaluated in a second resolution, which corresponds to the original resolution of the sensor data 10 .
- the complete sensor data 10 that is to say the complete camera image, is scaled down to a lower first resolution to reduced-resolution sensor data 10 (FIG. 3c).
- the focus region 8 and the resolution-reduced sensor data 10 are each evaluated by the evaluation method, which again includes a semantic segmentation, for example, in which an object class is assigned to the individual picture elements.
- Fig. 4 shows a schematic representation of the combined overall result 13.
- the overall result 13 clearly shows that the image elements associated with the child 30 are correctly classified and the object type “person” or “child” was assigned to these image elements, since a resolution could be selected that was higher in this area by defining a focus region 8 than in the remaining areas.
- the child 30 has detected or recognized the semantic segmentation in the detected sensor data 10 .
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
Description
Claims
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| DE102020211023.9A DE102020211023B4 (de) | 2020-09-01 | 2020-09-01 | Verfahren und Vorrichtung zur Umfeldwahrnehmung eines Umfeldes eines zumindest teilautomatisiert fahrenden Fahrzeugs |
| PCT/EP2021/068720 WO2022048811A1 (de) | 2020-09-01 | 2021-07-06 | Verfahren und vorrichtung zur umfeldwahrnehmung eines umfeldes eines zumindest teilautomatisiert fahrenden fahrzeugs |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| EP4208815A1 true EP4208815A1 (de) | 2023-07-12 |
Family
ID=77021310
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| EP21745257.2A Pending EP4208815A1 (de) | 2020-09-01 | 2021-07-06 | Verfahren und vorrichtung zur umfeldwahrnehmung eines umfeldes eines zumindest teilautomatisiert fahrenden fahrzeugs |
Country Status (5)
| Country | Link |
|---|---|
| US (1) | US12412363B2 (de) |
| EP (1) | EP4208815A1 (de) |
| CN (1) | CN116134492A (de) |
| DE (1) | DE102020211023B4 (de) |
| WO (1) | WO2022048811A1 (de) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102020106967A1 (de) * | 2020-03-13 | 2021-09-16 | Valeo Schalter Und Sensoren Gmbh | Festlegen eines aktuellen Fokusbereichs eines Kamerabildes basierend auf der Position der Fahrzeugkamera an dem Fahrzeug und einem aktuellen Bewegungsparameter |
Family Cites Families (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6324532B1 (en) | 1997-02-07 | 2001-11-27 | Sarnoff Corporation | Method and apparatus for training a neural network to detect objects in an image |
| US9275308B2 (en) | 2013-05-31 | 2016-03-01 | Google Inc. | Object detection using deep neural networks |
| US9511767B1 (en) | 2015-07-01 | 2016-12-06 | Toyota Motor Engineering & Manufacturing North America, Inc. | Autonomous vehicle action planning using behavior prediction |
| DE102015216352A1 (de) * | 2015-08-27 | 2017-03-02 | Bayerische Motoren Werke Aktiengesellschaft | Verfahren zum Erkennen einer möglichen Kollision eines Fahrzeuges mit einem Fußgänger auf Grundlage hochaufgelöster Aufnahmen |
| US9549125B1 (en) * | 2015-09-01 | 2017-01-17 | Amazon Technologies, Inc. | Focus specification and focus stabilization |
| US10229329B2 (en) * | 2016-11-08 | 2019-03-12 | Dedrone Holdings, Inc. | Systems, methods, apparatuses, and devices for identifying, tracking, and managing unmanned aerial vehicles |
| US9983591B2 (en) | 2015-11-05 | 2018-05-29 | Ford Global Technologies, Llc | Autonomous driving at intersections based on perception data |
| DE102016213494A1 (de) | 2016-07-22 | 2018-01-25 | Conti Temic Microelectronic Gmbh | Kameravorrichtung sowie Verfahren zur Erfassung eines Umgebungsbereichs eines eigenen Fahrzeugs |
| US10452068B2 (en) | 2016-10-17 | 2019-10-22 | Uber Technologies, Inc. | Neural network system for autonomous vehicle control |
| US20180306905A1 (en) * | 2017-04-20 | 2018-10-25 | Analog Devices, Inc. | Method of Providing a Dynamic Region of interest in a LIDAR System |
| DE102017127592A1 (de) | 2017-11-22 | 2019-05-23 | Connaught Electronics Ltd. | Verfahren zum Klassifizieren von Bildszenen in einem Fahrunterstützungssystem |
| US11500099B2 (en) * | 2018-03-14 | 2022-11-15 | Uatc, Llc | Three-dimensional object detection |
| US11592818B2 (en) * | 2018-06-20 | 2023-02-28 | Zoox, Inc. | Restricted multi-scale inference for machine learning |
| US10467503B1 (en) | 2018-09-05 | 2019-11-05 | StradVision, Inc. | Method and device for generating image data set to be used for learning CNN capable of detecting obstruction in autonomous driving circumstance |
-
2020
- 2020-09-01 DE DE102020211023.9A patent/DE102020211023B4/de active Active
-
2021
- 2021-07-06 EP EP21745257.2A patent/EP4208815A1/de active Pending
- 2021-07-06 WO PCT/EP2021/068720 patent/WO2022048811A1/de not_active Ceased
- 2021-07-06 US US18/023,683 patent/US12412363B2/en active Active
- 2021-07-06 CN CN202180053638.XA patent/CN116134492A/zh active Pending
Also Published As
| Publication number | Publication date |
|---|---|
| US12412363B2 (en) | 2025-09-09 |
| WO2022048811A1 (de) | 2022-03-10 |
| US20240029389A1 (en) | 2024-01-25 |
| DE102020211023B4 (de) | 2026-01-15 |
| CN116134492A (zh) | 2023-05-16 |
| DE102020211023A1 (de) | 2022-03-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP1634243B1 (de) | Verfahren und vorrichtung zur objektortung für kraftfahrzeuge | |
| EP3765927B1 (de) | Verfahren zum erzeugen eines trainingsdatensatzes zum trainieren eines künstlichen-intelligenz-moduls für eine steuervorrichtung eines fahrzeugs | |
| DE102006057552B4 (de) | System und Verfahren zur Messung des Abstands eines vorausfahrenden Fahrzeugs | |
| WO2020048669A1 (de) | Verfahren zum bestimmen einer spurwechselangabe eines fahrzeugs, ein computerlesbares speichermedium und ein fahrzeug | |
| DE102014207802B3 (de) | Verfahren und System zum proaktiven Erkennen einer Aktion eines Verkehrsteilnehmers | |
| DE10138641A1 (de) | Verfahren zur Bestimmung einer Modellfahrbahn | |
| DE102011055458A1 (de) | Adaptierungstechnik zur Erkennung einer freien Fahrbahn mittels verlässlicher lokaler Modellkalibrierung | |
| EP1298454A2 (de) | Verfahren zur Erkennung und Verfolgung von Objekten | |
| DE102009007342A1 (de) | Verfahren und Vorrichtung zum Ermitteln einer geltenden Fahrspurmarkierung | |
| DE102016210534A1 (de) | Verfahren zum Klassifizieren einer Umgebung eines Fahrzeugs | |
| DE102021119871B4 (de) | Verfahren und Prozessorschaltung zum Betreiben einer automatisierten Fahrfunktion mit Objektklassifikator in einem Kraftfahrzeug, sowie Kraftfahrzeug | |
| DE102011055459A1 (de) | Adaptierungstechnik zur Erkennung einer freien Fahrbahn mit zusätzlichen Klassifikatoren | |
| WO2019057252A1 (de) | Verfahren und vorrichtung zum erkennen von fahrspuren, fahrerassistenzsystem und fahrzeug | |
| DE102016202594A1 (de) | Verfahren und Vorrichtung zum Interpretieren einer Fahrzeugumgebung eines Fahrzeugs sowie Fahrzeug | |
| DE102017212175A1 (de) | Verfahren und Vorrichtung zum Ermitteln eines optischen Flusses anhand einer von einer Kamera eines Fahrzeugs aufgenommenen Bildsequenz | |
| DE102018211368A1 (de) | Verfahren zur Beschreibung einer Umgebung eines Fahrzeugs durch die Topologie der befahrenen Straße | |
| DE102024129386A1 (de) | Verfahren und vorrichtung zur fahrerabsichtsprognose | |
| WO2023237424A1 (de) | Verfahren und assistenzsystem zum vorhersagen eines fahrschlauches und kraftfahrzeug | |
| WO2021122338A1 (de) | Verfahren und vorrichtung zum robustifizieren von sensordaten gegen adversariale störungen | |
| EP4208815A1 (de) | Verfahren und vorrichtung zur umfeldwahrnehmung eines umfeldes eines zumindest teilautomatisiert fahrenden fahrzeugs | |
| DE102008050456B4 (de) | Verfahren und Vorrichtung zur Fahrspurerkennung | |
| DE102006037600B4 (de) | Verfahren zur auflösungsabhängigen Darstellung der Umgebung eines Kraftfahrzeugs | |
| DE112020002733T5 (de) | Elektronische Steuereinheit | |
| DE102021213756B3 (de) | Verfahren zum Fusionieren von Sensordaten im Kontext eines künstlichen neuronalen Netzwerks | |
| DE102019204187A1 (de) | Klassifizierung und temporale Erkennung taktischer Fahrmanöver von Verkehrsteilnehmern |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: UNKNOWN |
|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE |
|
| PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE |
|
| 17P | Request for examination filed |
Effective date: 20230403 |
|
| AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR |
|
| DAV | Request for validation of the european patent (deleted) | ||
| DAX | Request for extension of the european patent (deleted) | ||
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: EXAMINATION IS IN PROGRESS |
|
| 17Q | First examination report despatched |
Effective date: 20250721 |