WO2006094585A1 - Method for estimating the course of a lane - Google Patents

Method for estimating the course of a lane Download PDF

Info

Publication number
WO2006094585A1
WO2006094585A1 PCT/EP2006/000895 EP2006000895W WO2006094585A1 WO 2006094585 A1 WO2006094585 A1 WO 2006094585A1 EP 2006000895 W EP2006000895 W EP 2006000895W WO 2006094585 A1 WO2006094585 A1 WO 2006094585A1
Authority
WO
WIPO (PCT)
Prior art keywords
lane
course
hypotheses
sensors
hypothesis
Prior art date
Application number
PCT/EP2006/000895
Other languages
German (de)
French (fr)
Inventor
Uwe Francke
Axel Gern
Carsten Knöppel
Original Assignee
Daimlerchrysler Ag
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Daimlerchrysler Ag filed Critical Daimlerchrysler Ag
Publication of WO2006094585A1 publication Critical patent/WO2006094585A1/en

Links

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • B60W40/02—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to ambient conditions
    • B60W40/06—Road conditions
    • B60W40/076—Slope angle of the road
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • B60W40/02—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to ambient conditions
    • B60W40/06—Road conditions
    • B60W40/072—Curvature of the road
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
    • G01C21/28—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network with correlation of data from several navigational instruments
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S11/00—Systems for determining distance or velocity not using reflection or reradiation
    • G01S11/12—Systems for determining distance or velocity not using reflection or reradiation using electromagnetic waves other than radio waves
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
    • G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR 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
    • G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W2050/0001—Details of the control system
    • B60W2050/0043—Signal treatments, identification of variables or parameters, parameter estimation or state estimation
    • B60W2050/0052—Filtering, filters
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2552/00—Input parameters relating to infrastructure
    • B60W2552/05—Type of road
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2552/00—Input parameters relating to infrastructure
    • B60W2552/20—Road profile
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2552/00—Input parameters relating to infrastructure
    • B60W2552/30—Road curve radius
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
    • G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles
    • G01S2013/932—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles using own vehicle data, e.g. ground speed, steering wheel direction
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
    • G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles
    • G01S2013/9321—Velocity regulation, e.g. cruise control

Definitions

  • the invention relates to a method for estimating the course of a lane according to the preamble of patent claim 1.
  • driver assistance systems are known, inter alia, as so-called ACC systems (automatic cruise control).
  • ACC systems automatic cruise control
  • ACC systems automatic cruise control
  • ACC systems automatic cruise control
  • This smoothing of the results has an effect on the control of the steering. Signal noise immediately results in a troubled steering wheel and, as a result, ineffective steering feel in lane assist systems such as systems referred to as Comfortable Lane Keeping. Furthermore, by smoothing the results measurement errors can be excluded.
  • multi-sensor systems are known with which the described approach can be meaningfully extended.
  • the signals of all sensors are thereby introduced in a common Kalman filter.
  • the reliability and robustness are improved.
  • the individual sensors it is examined whether their signals pass the 3 Sigma test. Any measurement within the 3Sig ⁇ na range is assumed to match the current track.
  • the individual sensors are checked by means of the sensor associated reliability tests (reliability investigations) to determine whether the measurements can be trusted. For example, in the case of map systems, the distance of the current position measurement to the map (also with regard to the height data) and the correlation coefficient in the investigation of road-parallel structures can be used as a useful measure of confidence.
  • the present invention is based on the object to improve the estimation of the course of a lane.
  • the linear combination may optionally be weighted, wherein the weighting may depend on the plausibility of the respective hypothesis of the estimated lane course.
  • the generation of the hypotheses can be carried out in such a way that the respective subcombinations are formed from the existing sensors, a hypothesis being generated for each of these subcombinations by means of the Kalman filter and further calculated for the hypothesis of the individual subcombinations in the further measuring steps.
  • hypotheses can also be generated in such a way that the further calculation is carried out for each hypothesis in the next measurement step by deriving a further hypothesis from each of these hypotheses by deriving further hypotheses for the next measurement step for each of these hypotheses, By each of these hypotheses with all sub-combinations of the sensors is further calculated.
  • the hypothesis space is advantageously restricted in this procedure. This can be done, for example, by only pursuing a certain number of hypotheses, in which case the hypotheses can be evaluated for specific probabilities for their correctness.
  • the reliability of the statement is further improved.
  • certain theoretically resulting track courses can be eliminated. This can be done, for example, by comparing the hypotheses of estimated lanes with data from digitized maps. If a track pattern results that can not be reconciled with a road course due to the measured GPS position, this estimated track course can be completely discarded and furthermore no longer needs to be calculated.
  • MHT Multiple Hypothesis Tracking
  • the weighted linear combination of the individual estimated lanes, compared to a binary decision for the correctness of a particular result, combined with a corresponding erratic transition, when another model is judged correct, is achieved that no abrupt intervention in the steering of the vehicle.
  • the driving behavior of the vehicle is thus smoothed overall.
  • a "hard" switching can be provided in the sense that from one to the other moment the weighting to a sensor signal, the was detected as faulty, set to zero. Smoothing the transition is then not required.
  • the plausibility of the track course can be determined, for example, via the mean residual (the mean prediction error) or the so-called likelihood.
  • not only a weighting of the individual results for the resulting lane course from the individual systems can be allowed, but optionally also a hard switchover.
  • Each of these hypotheses can be tracked with its own Kalman filter.
  • the method can use a combination of different sensors.
  • These sensors can be, for example:
  • driving maneuvers are taken into account in the generation of the hypotheses.
  • These driving maneuvers can be, for example:
  • Road geometry change speed such as the difference between a winding country road and a typically relatively straight highway.
  • a hypothesis is generated for each sensor combination for different models, each of which takes into account other driving maneuvers.
  • a multi-filter Systems can be estimated a resulting hypothesis.
  • Such a multi-filter system may for example be an IMM (Interacting Multiple Model). It is also possible to evaluate all resulting hypotheses using the HMT method.
  • the weighted linear combination is used to assess the plausibility of the individual hypotheses.
  • the deviation of the individual hypotheses is evaluated in order to assess their plausibility as a function of their deviation from the linear combined estimate.

Abstract

The invention relates to a method for estimating the course of a lane, in which the output signals of several sensors are evaluated, the characteristics determined from several sensors are fed to a common Kaiman filter in order to generate a hypothesis about the course of the lane therefrom, the characteristics of sub-combinations of the sensors are fed to a Kaiman filter in order to generate one respective hypothesis about the course of the lane for the individual sub-combinations, and a resulting course of the lane is determined by linearly combining the individual hypotheses.

Description

DaimlerChrysler AG Herr BöppleDaimlerChrysler AG Mr. Böpple
Verfahren zum Schätzen des Verlaufs einer FahrspurMethod for estimating the course of a lane
Die Erfindung betrifft ein Verfahren zum Schätzen des Verlaufs einer Fahrspur nach dem Oberbegriff des Patentanspruchs 1.The invention relates to a method for estimating the course of a lane according to the preamble of patent claim 1.
Die Erkennung des Verlaufes einer Fahrspur spielt eine vergleichsweise bedeutende Rolle für Fahrerassistenzsysteme. Bei derartigen Systemen ist das Bestreben, eine automatische Steuerung des Fahrzeugs zu erreichen, zumindest jedoch den Fahrzeugführer durch entsprechende Informationsausgabe bei der Führung des Kraftfahrzeugs zu unterstützen. Derartige Fahrerassistenzsysteme sind u. a. als sogenannte ACC-Systeme bekannt (automatic cruise control) . Im Rahmen der Spurerkennung ist es beispielsweise bekannt, den Verlauf von Spurmarkierungen zu verfolgen. Diese werden durch Auswertung kleiner Messfenster erkannt und mittels Kaiman Filtern verfolgt. Jede aktuelle Messung wird dabei bewertet, indem der Messwert mit dem Erwartungswert verglichen wird. Der Erwartungswert ist der Wert, der sich aus der Schätzung des Spurverlaufs auf Grund vorher gehender Messungen ergibt . Messpunkte, die außerhalb des sogenannten 3Sigma-Bereichs liegen, können dann von der weiteren Auswertung ausgeschlossen werden. Entsprechend der Filtereinstellung ergibt sich dabei ein Kompromiss zwischen der Erfassung schneller Änderungen des Spurverlaufs wie beispielsweise auch der Erkennung von Manövern und einer Glättung der Ergebnisse. Diese Glättung der Ergebnisse hat Auswirkungen auf die Ansteuerung der Lenkung. Ein Signalrauschen führt unmittelbar zu einem unruhigen Lenkrad und in der Folge zu einem nicht optimalen Lenkgefühl bei Spur-unterstützenden Systemen wie beispielsweise Systemen, die als Comfortable Lane Keeping bezeichnet werden. Weiterhin können durch die Glättung der Ergebnisse Messfehler ausgeschlossen werden.The detection of the course of a traffic lane plays a comparatively important role for driver assistance systems. In such systems, the endeavor to achieve automatic control of the vehicle, but at least to assist the driver by appropriate information output in the management of the motor vehicle. Such driver assistance systems are known, inter alia, as so-called ACC systems (automatic cruise control). In the context of lane detection, it is known, for example, to track the course of lane markings. These are detected by evaluating small measurement windows and tracked using Kalman filters. Each current measurement is evaluated by comparing the measured value with the expected value. The expected value is the value resulting from the estimation of the lane course based on previous measurements. Measuring points that lie outside the so-called 3 Sigma range can then be excluded from further evaluation. According to the filter setting, this results in a compromise between the detection of rapid changes in the track course such as the detection of maneuvers and a smoothing of Results. This smoothing of the results has an effect on the control of the steering. Signal noise immediately results in a troubled steering wheel and, as a result, ineffective steering feel in lane assist systems such as systems referred to as Comfortable Lane Keeping. Furthermore, by smoothing the results measurement errors can be excluded.
Weiterhin sind Multisensorsysteme bekannt, mit denen der beschriebene Ansatz sinnvoll erweitert werden kann. Die Signale aller Sensoren werden dabei in einem gemeinsamen Kaiman- Filter eingebracht. Indem mehrere Informationen ausgewertet werden, ergibt sich eine Verbesserung der Zuverlässigkeit und Robustheit. Bezüglich der einzelnen Sensoren wird untersucht, ob deren Signale den 3Sigma-Test bestehen. Zu jeder Messung, die innerhalb des 3Sigτna-Bereiches liegt, wird angenommen, dass diese zum aktuellen Spurverlauf passt. Zusätzlich werden die einzelnen Sensoren mittels dem Sensor zugehörigen Konfi- denzuntersuchungen (Zuverlässigkeitsuntersuchungen) dahin gehend überprüft, ob den Messungen vertraut werden kann. Als brauchbares Konfidenzmaß ist beispielsweise bei Kartensystemen der Abstand der aktuellen Positionsmessung zur Karte (auch hinsichtlich der Höhendaten) und der Korrelations- koeffizient bei der Untersuchung fahrbahnparalleler Strukturen verwendbar .Furthermore, multi-sensor systems are known with which the described approach can be meaningfully extended. The signals of all sensors are thereby introduced in a common Kalman filter. By evaluating more information, the reliability and robustness are improved. With regard to the individual sensors, it is examined whether their signals pass the 3 Sigma test. Any measurement within the 3Sigτna range is assumed to match the current track. In addition, the individual sensors are checked by means of the sensor associated reliability tests (reliability investigations) to determine whether the measurements can be trusted. For example, in the case of map systems, the distance of the current position measurement to the map (also with regard to the height data) and the correlation coefficient in the investigation of road-parallel structures can be used as a useful measure of confidence.
Der vorliegenden Erfindung liegt die Aufgabe zu Grunde, die Schätzung des Verlaufes einer Fahrspur zu verbessern.The present invention is based on the object to improve the estimation of the course of a lane.
Diese Aufgabe wird nach der vorliegenden Erfindung gemäß Anspruch 1 gelöst, indem die Merkmale von Unterkombinationen der Sensoren einem Kaiman-Filter zugeführt werden, um zu den einzelnen Unterkombinationen jeweils eine Hypothese für den Verlauf der Fahrspur zu generieren, wobei ein sich ergebender Verlauf der Fahrspur bestimmt wird, indem die einzelnen Hypothesen linear kombiniert werden.This object is achieved according to the present invention according to claim 1, by the features of sub-combinations of the sensors are supplied to a Kalman filter to each of the individual sub-combinations to generate a hypothesis for the course of the lane, with a resulting Lane is determined by the individual hypotheses are linearly combined.
Es ist dabei möglich, nur bestimmte Unterkombinationen der vorhandenen Sensoren zu betrachten. Es können aber auch alle Unterkombinationen der vorhandenen Sensoren betrachtet werden. In diesem Falle werden also auch die aus den einzelnen Sensoren für sich abgeleiteten Hypothesen bei der Linearkombination berücksichtigt.It is possible to consider only certain subcombinations of the existing sensors. However, all subcombinations of the existing sensors can also be considered. In this case, therefore, the hypotheses derived from the individual sensors for the linear combination are also taken into account.
Die Linearkombination kann gegebenenfalls gewichtet erfolgen, wobei die Gewichtung abhängen kann von der Plausibilität der jeweiligen Hypothese des geschätzten Spurverlaufs.The linear combination may optionally be weighted, wherein the weighting may depend on the plausibility of the respective hypothesis of the estimated lane course.
Die Generierung der Hypothesen kann derart erfolgen, dass aus den vorhandenen Sensoren die jeweiligen Unterkombinationen gebildet werden, wobei für jede dieser Unterkombinationen mittels des Kaiman-Filters eine Hypothese generiert und jeweils für die Hypothese der einzelnen Unterkombinationen bei den weiteren Messschritten weitergerechnet wird.The generation of the hypotheses can be carried out in such a way that the respective subcombinations are formed from the existing sensors, a hypothesis being generated for each of these subcombinations by means of the Kalman filter and further calculated for the hypothesis of the individual subcombinations in the further measuring steps.
Die Generierung der Hypothesen kann- aber auch derart erfolgen, dass zu jeder Hypothese im nächsten Messschritt die weitere Berechnung erfolgt, indem aus jeder dieser Hypothesen eine weitere Hypothese abgeleitet wird, indem die zu jeder dieser Hypothesen wiederum weitere Hypothesen für den nächsten Messschritt abgeleitet werden, indem jede dieser Hypothesen mit allen Unterkombinationen der Sensoren weiter gerechnet wird. Um den Verarbeitungsaufwand zu beschränken, wird bei dieser Vorgehensweise vorteilhaft der Hypothesenraum eingeschränkt. Dies kann beispielsweise erfolgen, indem nur eine bestimmte Zahl von Hypothesen weiter verfolgt, wobei dazu die Hypothesen nach bestimmten Wahrscheinlichkeiten für deren Richtigkeit bewertet werden können. Vorteilhaft wird dabei die Zuverlässigkeit der Aussage weiter verbessert .However, the hypotheses can also be generated in such a way that the further calculation is carried out for each hypothesis in the next measurement step by deriving a further hypothesis from each of these hypotheses by deriving further hypotheses for the next measurement step for each of these hypotheses, By each of these hypotheses with all sub-combinations of the sensors is further calculated. In order to limit the processing effort, the hypothesis space is advantageously restricted in this procedure. This can be done, for example, by only pursuing a certain number of hypotheses, in which case the hypotheses can be evaluated for specific probabilities for their correctness. Advantageously, the reliability of the statement is further improved.
Um nicht zu viele Möglichkeiten berechnen zu müssen, können bestimmte sich theoretisch ergebende Spurverläufe ausgeschieden werden. Dies kann beispielsweise erfolgen, indem die Hypothesen geschätzter Spurverläufe mit Daten digitalisierter Karten verglichen werden. Wenn sich ein Spurverlauf ergibt, der auf Grund der gemessenen GPS-Position nicht mit einem Straßenverlauf in Einklang zu bringen ist, kann dieser geschätzte Spurverlauf vollständig verworfen werden und braucht weiterhin nicht mehr berechnet zu werden.In order not to have to calculate too many possibilities, certain theoretically resulting track courses can be eliminated. This can be done, for example, by comparing the hypotheses of estimated lanes with data from digitized maps. If a track pattern results that can not be reconciled with a road course due to the measured GPS position, this estimated track course can be completely discarded and furthermore no longer needs to be calculated.
In Kenntnis der Erfindung sei auf die "Multiple Hypothesis Tracking" (MHT) verwiesen, die eine aus der Objektverfolgung bekannte Assoziationstechnik darstellt.With reference to the invention, reference is made to the "Multiple Hypothesis Tracking" (MHT), which represents an association technique known from object tracking.
Durch die gewichtete Linearkombination der einzelnen geschätzten Spurverläufe wird gegenüber einer binären Entscheidung für die Richtigkeit eines bestimmten Ergebnisses, verbunden mit einem entsprechend sprunghaften Übergang, wenn ein anderes Modell als richtig bewertet wird, erreicht, dass keine abrupten Eingriffe in die Lenkung des Fahrzeugs erfolgen. Das Fahrverhalten des Fahrzeugs wird also insgesamt geglättet.The weighted linear combination of the individual estimated lanes, compared to a binary decision for the correctness of a particular result, combined with a corresponding erratic transition, when another model is judged correct, is achieved that no abrupt intervention in the steering of the vehicle. The driving behavior of the vehicle is thus smoothed overall.
Sofern kein Eingriff in eine Lenkung abgeleitet werden soll, sondern beispielsweise lediglich eine Spurzuordnung in einem ACC-System vorgenommen werden soll, kann eine "harte" Umschaltung vorgesehen werden in dem Sinne, dass von einem auf den anderen Moment die Gewichtung zu einem Sensorsignal, das als fehlerhaft erkannt wurde, auf Null gesetzt wird. Eine Glättung des Übergangs ist dann nicht erforderlich. Die Plausibilität des Spurverlaufs kann beispielsweise über das mittlere Residuum (den mittleren Prädiktionsfehler) oder die sogenannte Likelihood bestimmt werden. Darüber kann dann wie gerade beschrieben nicht nur eine Gewichtung der Einzel- ergebnisse zum resultierenden Spurverlauf aus den Einzel- Systemen erlaubt werden sondern gegebenenfalls auch eine harte Umschaltung.If no intervention is to be derived in a steering, but for example, only a track assignment to be made in an ACC system, a "hard" switching can be provided in the sense that from one to the other moment the weighting to a sensor signal, the was detected as faulty, set to zero. Smoothing the transition is then not required. The plausibility of the track course can be determined, for example, via the mean residual (the mean prediction error) or the so-called likelihood. In addition, as just described, not only a weighting of the individual results for the resulting lane course from the individual systems can be allowed, but optionally also a hard switchover.
Es kann also jede Spurverlaufshypothese hinsichtlich der Plausibilität bewertet werden. Wie ausgeführt kann dabei unter bestimmten Bedingungen eine Schranke definiert werden, bei deren Überschreiten (fehlender Plausibilität) der geschätzte Spurverlauf gar nicht weiter in die Auswertung eingeht und auch nicht mehr weiter berechnet wird. Um den erwähnten glatten Übergang zu erreichen, kann die Gewichtung dieses geschätzten Spurverlaufes dann entsprechend zügig bis auf 0 reduziert werden.It is therefore possible to evaluate each lane course hypothesis in terms of plausibility. As stated, under certain conditions, a barrier can be defined which, if exceeded (lack of plausibility), does not include the estimated lane course in the evaluation and is no longer calculated. In order to achieve the aforementioned smooth transition, the weighting of this estimated lane course can then be correspondingly reduced rapidly to 0.
Jede dieser Hypothesen kann mit einem eigenen Kaiman-Filter verfolgt werden.Each of these hypotheses can be tracked with its own Kalman filter.
Es kann bei dem Verfahren eine Kombination verschiedener Sensoren Verwendung finden. Diese Sensoren können beispielsweise sein:The method can use a combination of different sensors. These sensors can be, for example:
• ein optisches Spurerkennungssystem auf Basis der lokalen Spurverfolgung,An optical tracking system based on local tracking,
• ein DGPS-basiertes Trajektorien bzw. Kartensystem,• a DGPS-based trajectory or map system,
• eine optische Spurerkennung mit Bildern einer zweiten Kamera nach vorne mit anderem Blickwinkel und/oder anderem Fokus ,An optical lane detection with images of a second camera forward with a different angle of view and / or different focus,
• eine optische Spurerkennung mit Bildern einer Rückraumkamera, • eine Spurerkennung mit Totwinkelkameras, die in die Außenspiegel integriert sein können,• an optical track recognition with pictures of a backspace camera, • lane detection with blind cameras, which can be integrated into the exterior mirrors,
• die durch ein Objekterkennungssystem erkannten Verkehrsteilnehmer,The road users recognized by an object recognition system,
• eine globale Bildanalyse zur Spurerkennung,• a global image analysis for track recognition,
• weitere Merkmale im Kamerabild wie fahrbahnparallele Strukturen .• additional features in the camera image such as lane-parallel structures.
Dabei ist es möglich, diese Sensoren in unterschiedlichen denkbaren Varianten miteinander zu kombinieren. Es kann auch die Zahl der verwendeten Sensoren verändert werden.It is possible to combine these sensors in different conceivable variants. It is also possible to change the number of sensors used.
Bei der Ausgestaltung nach Anspruch 2 werden bei der Generierung der Hypothesen Fahrmanöver berücksichtigt.In the embodiment according to claim 2, driving maneuvers are taken into account in the generation of the hypotheses.
Diese Fahrmanöver können beispielsweise sein:These driving maneuvers can be, for example:
• ein Nicken des Fahrzeugs auf Grund von Bodenwellen,A pitching of the vehicle due to bumps,
• dynamische Spurwechsel bzw. Ausweichmanöver des eigenen Fahrzeugs, die unter anderem eine Wankbewegung des Fahrzeugs mit sich bringen,• dynamic lane changes or avoidance maneuvers of the own vehicle, which, among other things, bring about a rolling motion of the vehicle,
• ein Spurwechsel von voraus fahrenden Fahrzeugen,A lane change of vehicles in front,
• sich schnell ändernde Spurbreiten, die beispielsweise an Spurgabelungen, Zusammenführungen und Ausfahrten auftreten können,• rapidly changing track widths, which can occur, for example, on track cabling, mergers and exits,
• Geschwindigkeit der Straßengeometrieänderung wie beispielsweise der Unterschied zwischen einer kurvigen Landstraße und einer typischerweise vergleichsweise gerade verlaufenden Autobahn.• Road geometry change speed, such as the difference between a winding country road and a typically relatively straight highway.
Es wird also nochmals zu jeder Sensorkombination zu unterschiedlichen Modellen, die jeweils andere Fahrmanöver berücksichtigen, eine Hypothese generiert. Aus diesen Hypothesen zu den Fahrmanövern kann mittels eines Mehrfilter- Systems eine resultierende Hypothese geschätzt werden. Ein solches Mehrfiltersystem kann beispielsweise ein IMM (Interacting Multiple Model) sein. Ebenso ist es möglich, alle sich so ergebenden Hypothesen mittels des HMT-Verfahrens zu bewerten.Thus, once again, a hypothesis is generated for each sensor combination for different models, each of which takes into account other driving maneuvers. From these hypotheses to the driving maneuvers, by means of a multi-filter Systems can be estimated a resulting hypothesis. Such a multi-filter system may for example be an IMM (Interacting Multiple Model). It is also possible to evaluate all resulting hypotheses using the HMT method.
Es wird beispielsweise möglich, durch die Interpretation der Änderungsgeschwindigkeit weiterhin eine Klassifikation der Straße vorzunehmen dahin gehend, ob es sich um eine Autobahn, eine Bundesstraße oder eine Landstraße handelt. Diese Unterscheidung kann durch eine Erkennung der unterschiedlichen und für die jeweiligen Straßenarten typischen Dynamik von Fahrmanövern vorgenommen werden .It becomes possible, for example, to continue to classify the road by interpreting the rate of change as to whether it is a highway, a main road or a country road. This distinction can be made by recognizing the different dynamics of driving maneuvers typical of the respective road types.
Bei der Ausgestaltung des Verfahrens nach Anspruch 3 wird die gewichtete Linearkombination heran gezogen, um die Plausibi- lität der einzelnen Hypothesen zu bewerten.In the embodiment of the method according to claim 3, the weighted linear combination is used to assess the plausibility of the individual hypotheses.
Anhand des Ergebnisses der bisherigen linear kombinierten Schätzung wird die Abweichung der einzelnen Hypothesen bewertet, um deren Plausibilität abhängig von deren Abweichung zur linear kombinierten Schätzung zu bewerten. Based on the result of the previous linear combined estimate, the deviation of the individual hypotheses is evaluated in order to assess their plausibility as a function of their deviation from the linear combined estimate.

Claims

DaimlerChrysler AG Herr BöpplePatentansprüche DaimlerChrysler AG Mr. BöpplePatent claims
1. Verfahren zum Schätzen des Verlaufs einer Fahrspur, wobei zur Schätzung des Spurverlaufs die Ausgangssignale mehrerer Sensoren ausgewertet werden, wobei die aus mehreren Sensoren ermittelten Merkmale einem gemeinsamen Kaiman- Filter zugeführt werden, um daraus eine Hypothese für den Verlauf der Fahrspur zu generieren, dadurch gekennzeichnet, dass die Merkmale von Unterkombinationen der Sensoren einem Kaiman-Filter zugeführt werden, um zu den einzelnen Unterkombinationen jeweils eine Hypothese für den Verlauf der Fahrspur zu generieren, wobei ein sich ergebender Verlauf der Fahrspur bestimmt wird, indem die einzelnen Hypothesen linear kombiniert werden.1. A method for estimating the course of a lane, wherein the output signals of several sensors are evaluated to estimate the lane, wherein the determined from a plurality of sensors features are fed to a common Kalman filter to generate a hypothesis for the course of the lane, characterized characterized in that the characteristics of sub-combinations of the sensors are supplied to a Kalman filter in order to generate a hypothesis for the course of the lane to the individual sub-combinations, wherein a resulting course of the lane is determined by linearly combining the individual hypotheses.
2. Verfahren nach Anspruch 1 , dadurch gekennzeichnet, dass bei der Generierung der Hypothesen Fahrmanöver berücksichtigt werden.2. The method according to claim 1, characterized in that driving maneuvers are taken into account in the generation of the hypotheses.
3. Verfahren nach Anspruch 1 oder 2 , dadurch gekennzeichnet, dass die gewichtete Linearkombination heran gezogen wird, um die Plausibilität der einzelnen Hypothesen zu bewerten. Verfahren nach einem der vorstehenden Ansprüche, dadurch gekennzeichnet, dass die Geschwindigkeit mit der sich der Spurverlauf ändert herangezogen wird, um damit den Typ der befahrenen Straße zu klassifizieren. 3. The method according to claim 1 or 2, characterized in that the weighted linear combination is used to assess the plausibility of the individual hypotheses. Method according to one of the preceding claims, characterized in that the speed with which the lane course changes is used in order to classify the type of the road being traveled.
PCT/EP2006/000895 2005-03-08 2006-02-02 Method for estimating the course of a lane WO2006094585A1 (en)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
DE102005011030.4 2005-03-08
DE102005011030 2005-03-08
DE102005038314A DE102005038314A1 (en) 2005-03-08 2005-08-11 Method for estimating the course of a lane
DE102005038314.9 2005-08-11

Publications (1)

Publication Number Publication Date
WO2006094585A1 true WO2006094585A1 (en) 2006-09-14

Family

ID=36294272

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/EP2006/000895 WO2006094585A1 (en) 2005-03-08 2006-02-02 Method for estimating the course of a lane

Country Status (2)

Country Link
DE (1) DE102005038314A1 (en)
WO (1) WO2006094585A1 (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110245995A1 (en) * 2009-01-23 2011-10-06 Daimler Ag Method for determining a road profile of a lane located in front of a vehicle
CN109313032A (en) * 2016-06-24 2019-02-05 高通股份有限公司 It defines in dynamic lane
CN111194459A (en) * 2017-10-10 2020-05-22 大众汽车有限公司 Evaluation of automatic driving function and road recognition in different processing stages

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102006057276B4 (en) 2006-12-05 2023-09-28 Robert Bosch Gmbh Method and device for object tracking in a driver assistance system of a motor vehicle
CN102209658B (en) * 2008-11-06 2014-01-15 沃尔沃技术公司 Method and system for determining road data
DE102016214045A1 (en) * 2016-07-29 2018-02-01 Bayerische Motoren Werke Aktiengesellschaft Method and device for determining a roadway model for a vehicle environment

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4970653A (en) * 1989-04-06 1990-11-13 General Motors Corporation Vision method of detecting lane boundaries and obstacles
WO2004008648A2 (en) * 2002-07-15 2004-01-22 Automotive Systems Laboratory, Inc. Road curvature estimation and automotive target state estimation system
US6718259B1 (en) * 2002-10-02 2004-04-06 Hrl Laboratories, Llc Adaptive Kalman filter method for accurate estimation of forward path geometry of an automobile

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4970653A (en) * 1989-04-06 1990-11-13 General Motors Corporation Vision method of detecting lane boundaries and obstacles
WO2004008648A2 (en) * 2002-07-15 2004-01-22 Automotive Systems Laboratory, Inc. Road curvature estimation and automotive target state estimation system
US6718259B1 (en) * 2002-10-02 2004-04-06 Hrl Laboratories, Llc Adaptive Kalman filter method for accurate estimation of forward path geometry of an automobile

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
DICKMANNS E D ET AL: "RELATIVE 3D-STATE ESTIMATION FOR AUTONOMOUS VISUAL GUIDANCE OF ROAD VEHICLES*", ROBOTICS AND AUTONOMOUS SYSTEMS, ELSEVIER SCIENCE PUBLISHERS, AMSTERDAM, NL, vol. 7, no. 2 / 3, 1 August 1991 (1991-08-01), pages 113 - 123, XP000219094, ISSN: 0921-8890 *

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110245995A1 (en) * 2009-01-23 2011-10-06 Daimler Ag Method for determining a road profile of a lane located in front of a vehicle
US8676508B2 (en) * 2009-01-23 2014-03-18 Daimler Ag Method for determining a road profile of a lane located in front of a vehicle
CN102292249B (en) * 2009-01-23 2014-10-08 戴姆勒股份公司 Method for determining a road profile of a lane located in front of a vehicle
CN109313032A (en) * 2016-06-24 2019-02-05 高通股份有限公司 It defines in dynamic lane
CN109313032B (en) * 2016-06-24 2020-08-14 高通股份有限公司 Dynamic lane definition
CN111194459A (en) * 2017-10-10 2020-05-22 大众汽车有限公司 Evaluation of automatic driving function and road recognition in different processing stages
CN111194459B (en) * 2017-10-10 2024-01-09 大众汽车有限公司 Evaluation of autopilot functions and road recognition in different processing phases

Also Published As

Publication number Publication date
DE102005038314A1 (en) 2006-09-14

Similar Documents

Publication Publication Date Title
EP3584663B1 (en) Method for automatic transverse guidance of a follow vehicle in a vehicle platoon
EP1680317B1 (en) Driver assist method and device based on lane information
EP2565106B1 (en) Method for monitoring lanes and lane monitoring system for a vehicle
EP2793045B1 (en) Method for testing an environment detection system of a vehicle
DE19749086C1 (en) Device for determining data indicating the course of the lane
EP1690730B1 (en) Driver assistance system comprising redundant decision unit
DE102017009435B4 (en) Evaluation of components of automatic driving functions and lane recognition at different processing levels
DE102006040334A1 (en) Lane recognizing method for use with driver assistance system of vehicle i.e. motor vehicle, involves reconstructing characteristics of lane markings and/or lanes from position of supporting points in coordinate system
EP2756264B1 (en) Method for determining position data of a vehicle
WO2020229002A1 (en) Method and device for multi-sensor data fusion for automated and autonomous vehicles
DE102016209232B4 (en) Method, device and computer-readable storage medium with instructions for determining the lateral position of a vehicle relative to the lanes of a roadway
DE102020107349A1 (en) METHOD AND APPARATUS FOR DYNAMIC YEAR RATE DEVIATION ESTIMATION
WO2018019454A1 (en) Method and device for determining a roadway model for the surroundings of a vehicle
WO2018019464A1 (en) Method, device and computer-readable storage medium with instructions for determining the lateral position of a vehicle relative to the lanes of a road
WO2006094585A1 (en) Method for estimating the course of a lane
DE102017117593A1 (en) Vehicle driving assistance device
DE102016011366A1 (en) Method for determining the position of a vehicle
WO2009013052A2 (en) Method and device for sensing a lane with a driver assistance system
EP3857532A1 (en) Method for evaluating an effect of an object in the surroundings of a means of transport on a driving manoeuvre of the means of transport
DE102004028404A1 (en) Method for estimating the course of a lane of a motor vehicle
DE102010003375B4 (en) Environment evaluation system in a vehicle with sensor means for detecting objects in the vicinity of the vehicle
DE102008021380B4 (en) Method and device for predicting a course of a roadway and driver assistance system
DE102010062322A1 (en) Method for calibrating system for assisting driver of motor car during driving maneuvers in e.g. narrow road, involves averaging error, and correcting distance detected before starting of maneuvers to objects by error average value
DE102020113418B4 (en) AUTONOMOUS DRIVING DEVICE AND METHOD
DE102007015227A1 (en) Method for determining active drive lane of roadway from vehicle, involves determining active drive lane from vehicle by artificial neuronal network that has input layer, activation layer and output layer

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application
NENP Non-entry into the national phase

Ref country code: DE

NENP Non-entry into the national phase

Ref country code: RU

WWW Wipo information: withdrawn in national office

Country of ref document: RU

122 Ep: pct application non-entry in european phase

Ref document number: 06706567

Country of ref document: EP

Kind code of ref document: A1