DE102020118628A1 - Method and device for determining a position of a vehicle - Google Patents

Abstract

One aspect of the invention relates to a method (200) for determining a position of a vehicle (105), comprising the steps of: estimating (210) the position of the vehicle (105); and validating (220) the estimated position against information provided by a plurality of sources (115, 1151, 1152) from a pool of sources (115, 1151, 1152), the pool of sources (115, 1151, 1152) a first sensor (1151) and a second sensor (1152), and wherein first information provided by the first sensor (1151) is dependent on second information provided by the second sensor (1152) for validation (220) of the estimated position can be used.

Description

The invention relates to a method and a device for determining a position of a vehicle, and a vehicle with such a device. In particular, the invention relates to safeguarding the determination against possible errors. In this regard, reference is also made to the German patent application 10 2019 133 316.4 Reference is made to the content of which is hereby incorporated into this application.

A vehicle can have a driving assistant that is set up to influence a longitudinal or lateral control of the vehicle. For example, a lane assistant can be set up to keep the vehicle between lane markings. The markings can, for example, be scanned by a camera and recognized automatically.

Many driving assistants require knowledge of an exact position of the vehicle. The position can be determined in the longitudinal and/or transverse direction and expressed with respect to a predetermined reference point. An absolute geographic position can be determined, for example, with respect to a predetermined geodetic reference system such as the WGS84. A relative position of the vehicle can be specified, for example, in the transverse direction with respect to a recognized lane marking.

The determination of the position of the vehicle is usually subject to a number of errors and inaccuracies. For example, sensors provide noisy and/or corrupted information, or may occasionally fail altogether. Different measurement conditions or complex processing heuristics lead to different degrees of accuracy or reliability. If the vehicle is controlled based on such a determined position, the safety of the vehicle or an occupant may be compromised.

One of the objects on which the invention is based is to provide improved position determination for a vehicle. The invention solves the problem by means of the subject matter of the independent claims. Subclaims reflect preferred embodiments.

The object is solved by the features of the independent patent claims. Advantageous embodiments are described in the dependent claims.

It is pointed out that additional features of a patent claim dependent on an independent patent claim without the features of the independent patent claim or only in combination with a subset of the features of the independent patent claim can form a separate invention independent of the combination of all features of the independent patent claim, which can be made the subject of an independent claim, a divisional application or a subsequent application. This applies equally to the technical teachings described in the description, which can form an invention independent of the features of the independent patent claims.

According to a first aspect of the present invention, a method for determining a position of a vehicle comprises the steps of: estimating the position of the vehicle; and validating the estimated position based on information provided by multiple sources from a pool of sources, the pool of sources including a first sensor and a second sensor. In this case, first information provided by the first sensor is used depending on second information provided by the second sensor for validating the estimated position.

Estimating the vehicle position can be done, for example, in relation to map information, i.e. it can be estimated, for example, at which map position the vehicle is located.

For example, the vehicle position may be estimated using data from a satellite navigation system (e.g., GPS or DGPS). In addition or as an alternative, sensor data can also be used, which is provided, for example, by an environment sensor system of the vehicle. The environment sensors can include, for example, one or more radar and/or LiDAR sensors and/or one or more cameras.

In particular, the position can be estimated relative to one or more lane boundaries and, for example, make a statement about which of several lanes the vehicle is traveling in.

In order to be able to ensure adequate safety, for example in autonomous or highly automated driving, it is necessary for the vehicle position to be reliably determined relative to the surrounding lanes. It is therefore desirable within the framework of an overall safety concept to determine with a very high statistical reliability which lane the vehicle is in. In other words, a high level of localization certainty, particularly with regard to the lane, is desirable.

To ensure the integrity of the vehicle position estimate, it is validated. The validation is based on information that is provided by a number of sources from a total of available sources.

The totality of sources preferably includes a number of different and/or statistically independent sources, such as different types of sensors (e.g. one or more LiDAR sensors and/or one or more RADAR sensors and/or one or more optical cameras).

According to the method, the totality of sources that are available as information sources for the validation includes a first sensor (e.g. in the form of a LiDAR sensor) and a second sensor (e.g. in the form of a camera, in particular an optical one).

The validation can be performed by a processing device. The respective sources used for the validation can be assigned to a logical or data processing level, e.g. several software components called "validators".

Reliable lateral localization, including a determination of which lane the vehicle is traveling in, is generally particularly important. This can be checked, for example, using one or more so-called validators.

Such a validator, which validates a lateral localization, can, for example, also check a longitudinal localization (possibly with lower accuracy requirements compared to the lateral localization).

For example, one or more validators can also be provided, which check that the vehicle is actually on the right road.

In general, it is preferred that a number of different validators check whether assigned sensor data confirm an estimated position or not. For example, based on LiDAR data, a first validator can make a statement as to whether the vehicle is in a specific lane according to the estimated position. In addition, a second validator, e.g. based on camera track detection, can provide a corresponding evaluation of the position estimate. In general, if both validators confirm the position estimate, there is a higher certainty that the position estimate is correct, compared to a situation in which there is only confirmation by one validator, or in comparison to a situation in which even one or even both Validators expressly do not confirm the position estimate. A non-confirmation of the position estimate by a validator can occur, for example, in cases where the data available to the validator is insufficient to reliably confirm the position and/or when an actual deviation in the position is detected.

In advantageous embodiments, the validation concept is therefore based on a number of preferably statistically independent validators that can confirm a position estimate. When enough validators confirm an estimate, the position estimate is considered secure.

With such a validation concept, the determined position of the vehicle can, in principle, be arbitrarily safely within a predetermined error range. If the position of the vehicle is validated on the basis of three validators, and the probabilities of an incorrect validation by the individual validators are determined in such a way that only one in 100 cases is incorrectly validated by a validator, then a validated with a positive result can be validated The position of the vehicle can still be wrong in only one out of 100 * 100 * 100 = 1,000,000 cases. For this determination it was assumed that the validators are independent of each other. Overall, the probability of an unnoticed erroneous determination of the vehicle position can be made so improbable that the position can be determined with a degree of certainty that satisfies a required safety in use.

According to the above, it is in principle desirable that the individual validators fail as rarely as possible. In this context, failure can mean in particular that a validator confirms a position estimate that is actually incorrect. For example, a safety-critical situation could result if, for example, a LiDAR-based validator, possibly together with other validators, incorrectly comes to the conclusion that a position estimate is correct.

According to the invention it is therefore provided that first information provided by the first sensor is used (or not) depending on second information provided by the second sensor for validating the estimated position.

The formulation that the first information is used during validation depending on the second information is in to be understood in such a way that this also includes in particular the possibility that the first information (depending on the second information available) is not used in the validation.

According to a preferred embodiment, the second information can indicate in particular the reliability of validating the estimated position using the first information.

The invention is based on the idea that it is possible in certain situations, for example, to use the second information from the second sensor (e.g. a camera) to predict that a validator, which is based on the first information from the first sensor (e.g a LiDAR sensor) will probably fail.

Such a prediction can be implemented, for example, as part of a software component that can also be referred to as a "predictor". If the predictor predicts such a failure, e.g. a confirmation of the estimated vehicle position by the LiDAR-based validator can be considered less reliable. In this case, it can make sense not to use the first information (i.e. the LiDAR data) to validate the estimated position at all.

According to one embodiment, a prediction regarding the expected reliability of a LiDAR-based validator is based on camera data, i.e. the first sensor in this case is a LiDAR sensor and the second sensor is an (optical) camera.

A prediction of a failure of the LiDAR-based validator can, for example, proceed as follows: For example, the camera can be used to recognize certain types of plants in the vehicle environment, which can change their appearance (shape, color) significantly over time. This can affect deciduous trees, for example, which change their appearance significantly with the seasons. In such a case, the camera-based predictor can determine that the LiDAR-based validator is likely to fail because it will not detect an expected landmark in the shape of a leafy tree. It is also conceivable that the camera will detect that a roadside bush has been cut back and consequently changed shape, so the LiDAR-based validator will likely have trouble recognizing the bush as a landmark. Alternatively or additionally, the camera can be used, for example, to detect a truck parked next to the freeway, with the result that the LiDAR-based validator could incorrectly assume a crash barrier.

According to some advantageous embodiments, such aspects that can be recognized and evaluated by the predictor do not necessarily have to be known in advance and, for example, stored in a data memory in the sense of an explicit listing. Rather, they can also be determined using a data-based approach using machine learning. Thus, a developer who configures a device according to the invention does not have to be aware that, for example, young deciduous trees can pose a problem for the LiDAR-based validator, while old conifers may not.

In an advantageous embodiment, the type and/or the number of the remaining sources (i.e. the sources other than the first sensor) whose information is used for validation are determined depending on the second information. For example, in the event of a predicted failure of a LiDAR-based validator, another validator can be used instead of the LiDAR-based validator. It is also conceivable that in this case a larger set of validators is required overall in order to nevertheless consider a position estimate to be secure.

The LiDAR-based validator and the camera-based predictor together can also be considered conceptually as a single more robust validator.

It should be noted that the validator that works with the first information (e.g. the LiDAR validator) preferably only uses the first information for the validation itself, since it should be as independent as possible from the other validators. An idea according to the invention, however, is precisely to introduce a specific cross-reference between the validators, which are as separate as possible and are based on two independent sensors - namely in such a way that the first information (from the first sensor) depends on the second information (from the second sensor, eg a camera) are used in the validation.

In accordance with the examples described above, the method according to an embodiment comprises the following steps: determining, depending on the second information, whether the estimated position can be validated with sufficient reliability based on the first information; validating the estimated position using the first information if the determining indicates that a sufficiently reliable the estimated position can be casually validated based on the first information; Validating the estimated position without using the first information if the determination reveals that a sufficiently reliable validation of the estimated position based on the first information is not possible.

Depending on the second information, the determination of whether a sufficiently reliable validation of the estimated position is possible using the first information can also be a probability-based estimate, for example, i.e. the determination does not have to deliver an accurate result in every case.

The following step can optionally also be provided: If the determination shows that a sufficiently reliable validation of the estimated position based on the first information is not possible, validating the estimated position using information from at least one or more additional sources (apart from the first sensor ) from the totality of sources.

For example, a desired level of accuracy can be achieved in this way, which can depend, for example, on a number of statistically independent sources used.

For example, determining the position of the vehicle may be performed at least in part using machine learning. For example, the position can be estimated and/or the position validated using an artificial neural network or another machine-learning-capable system. Although the position determination cannot necessarily be formally validated in this case, the invention can enable protection against erroneous determination, which also enables machine learning to be used in a safety-relevant environment.

In particular, machine learning methods can be used to determine, depending on the second information, whether the estimated position can be validated sufficiently reliably using the first information. Alternatively, classic methods that do not require machine learning can also be implemented.

If machine learning is used, a sufficient database and suitable ground truth are required for good validation. For example, the data from test drives covering millions of kilometers can be used as a data set. This data can e.g. be intentionally linked to fake position estimates that the LiDAR-based validator has to reject, as well as to ground truth positions (e.g. received from a reference DGPS sensor) that the validator has to confirm as correct. If the LiDAR-based validator incorrectly confirms a falsified position estimate for a data sample, the validator is considered faulty. In this case, the predictor must predict that the validator will fail.

The method can be further improved by determining a pose of the vehicle, i.e. estimating and validating it in the manner proposed by the invention. A pose includes an orientation of the vehicle (e.g., orientation relative to one or more lane boundaries) in addition to the position of the vehicle. The position can be specified, for example, in Cartesian coordinates along a right-hand system of two or three axes, and the orientation as an angle of rotation around these axes.

According to a second aspect of the invention, an apparatus for determining a position of a vehicle comprises: a first sensor and a second sensor; and a processing device arranged to carry out the following steps: estimating the position of the vehicle; and validating the estimated position based on information provided by multiple sources from a pool of sources, the pool of sources including the first sensor and the second sensor, and wherein first information provided by the first sensor is dependent of second information provided by the second sensor to validate the estimated position.

The processing device can be set up to carry out a method described herein in whole or in part. For this purpose, the processing device can comprise a programmable microcomputer or microcontroller and the method can be present in the form of a computer program product with program code means. The computer program product can also be stored on a computer-readable data carrier. Features or advantages of the method can be transferred to the device or vice versa.

Different processors may also be used, such as a dedicated processor to estimate the position and another processor to validate the position.

In general, the first sensor and the second sensor and optionally one or more other sources set up from the totality of sources to collect and provide information about the environment of the vehicle. The sensors mentioned and possibly one or more other sources can in particular include a camera, a depth camera, a radar sensor, a LiDAR sensor, a receiver of a satellite-based navigation system or a similar sensor. Such a sensor preferably works without contact, for example by means of detection of electromagnetic waves or ultrasonic waves, and can be imaging.

In a preferred embodiment, the first sensor is a LiDAR sensor.

In a preferred embodiment, the second sensor is a camera, in particular an optical camera.

For example, the first sensor and/or the second sensor and/or one or more other sources from the totality of sources can provide information about a lane boundary detected (e.g. optically or using LiDAR), an orientation point detected (e.g. optically or using LiDAR) or a ( For example, provide an object recognized optically or by means of LiDAR.

In accordance with the above, a camera and/or a LiDAR sensor can be provided for scanning an area surrounding the vehicle, for example. In an image provided by the camera and/or by the LiDAR sensor, a lane boundary and/or an orientation point and/or an object in the vehicle environment can be detected, for example.

According to one embodiment, one of the sources provides information based on received signals from a satellite-based navigation system. For example, the device can include a receiving device for receiving signals from a satellite-supported navigation system, with the estimation and/or validation of the position taking place using the received signals.

The device preferably includes a map memory for providing map data, with the estimation and/or the validation of the position taking place using the map data. The map data can in particular include a position or orientation of an object, a point of orientation (also called a landmark) and/or a route on which the vehicle can drive.

The map data can be based, for example, at least in part on data recorded by sensors during one or more reconnaissance trips by a reconnaissance vehicle. An environment sensor system used here can include, for example, a receiver of a global satellite navigation system (e.g. GPS or DGPS), one or more optical cameras, one or more RADAR sensors and/or one or more LiDAR sensors.

Accordingly, a digital map stored in the map memory may contain multiple layers, e.g., where one layer is based on data from a global navigation satellite system, another layer is based on optical camera data, and another layer is based on LiDAR data. The different layers can contain features that can be recognized by the respective sensors.

According to one embodiment, the device comprises an odometer for providing odometry data, with the estimation and/or the validation of the position taking place using the odometry data. The odometry data can, for example, express a movement that has taken place in relation to a previously occupied point and can be determined, for example, on the basis of signals which are provided by yaw rate sensors on the vehicle's wheels.

According to another aspect, a vehicle includes a device as described herein. The vehicle preferably includes a drive motor and is a motor vehicle, in particular a road-bound motor vehicle. The motor vehicle can be controlled in the longitudinal direction, for example by influencing the drive motor or a braking device.

The vehicle is preferably set up for at least partially automated driving, up to highly automated or even autonomous driving.

For example, a driving function of the vehicle can be controlled depending on the determined position. In particular, the driving function can bring about a longitudinal and/or lateral control of the vehicle, for example in the form of a speed assistant or a lane departure warning system. The determined position can be a safety-relevant parameter of the vehicle and can be measured, for example, in the longitudinal and/or transverse direction of the vehicle.

The above statements on the method according to the invention according to the first aspect of the invention also apply in a corresponding manner to the device according to the invention according to the second aspect of the invention. At this point and advantageous exemplary embodiments of the device according to the invention that are not explicitly described in the patent claims correspond to the advantageous exemplary embodiments of the method according to the invention described in the description or in the patent claims, and vice versa.

• 1A-C jeweils schematisch und beispielhaft ein Fahrzeug mit einer Vorrichtung zum Bestimmen einer Position des Fahrzeugs; und
• 2A-B jeweils ein schematisches Ablaufdiagramm eines Verfahrens zum Bestimmen einer Position eines Fahrzeugs.

The invention is described below on the basis of exemplary embodiments with the aid of the accompanying drawings. In these show:

• 1A-C in each case, schematically and by way of example, a vehicle with a device for determining a position of the vehicle; and
• 2A-B each shows a schematic flow chart of a method for determining a position of a vehicle.

the 1A-C illustrate schematically (ie in particular not necessarily true to scale) some example scenarios in which a position of a motor vehicle 105 is determined using different landmarks 3, 4, 5.

In the following, the invention is explained by way of example using these example scenarios, with reference also being made immediately to the 2A and 2 B Reference is made which show schematic flowcharts of embodiments of a method according to the invention.

The vehicle 105 is equipped with a device 110 for determining a position of a vehicle 105 . In other words is in the 1A-C So in each case a system 100 is shown, which comprises the vehicle 105 and the device 110 .

The device 110 comprises a plurality of sources 115 for providing information, with a first sensor 1151 and a second sensor 1152 being among the sources in particular. In the present embodiment, the first sensor 1151 is a LiDAR sensor and the second sensor 1152 is a camera.

Next to the in the 1A-C Sources 115 that are explicitly illustrated can also be provided in addition to one or more other sources, which together form a total of information sources that are available for determining the position of vehicle 105 . For example, a receiving device for receiving signals from a satellite-based navigation system (eg DGPS) and/or an odometer for providing odometry data can be provided as further sources of the totality of sources 115 . Furthermore, for example, one or more RADAR sensors can be provided as additional source(s).

In the present exemplary embodiment, the device 110 further comprises a map memory 125 for providing map data, which can also be viewed as an information source that is part of the set of sources 115 .

In general, the first sensor 1151 and the second sensor 1152 and optionally one or more other sources from all of the sources 115 are set up to collect and provide information relating to the environment of the vehicle 105 . For example, the first sensor 1151 and/or the second sensor 1152 and/or one or more further sources 115 from the totality of sources can receive information about a recognized lane boundary, a landmark 3, 4, 5 and/or a recognized object 3, 4, 5 deploy.

In the present exemplary embodiment, the camera 1152 and the LiDAR sensor 1151 are designed to capture or scan the surroundings of the vehicle 105 . One or more lane boundaries and/or an orientation point in the form of a building 3 (cf. 1A) , of a tree 4 (cf. 1B) be recognized.

Furthermore, the device 110 comprises a processing device 120, which is set up to perform the following in 2A carry out the steps shown schematically: estimating 210 a position of the vehicle 105; and validating 220 the estimated position against information provided by multiple sources in the set of sources 115 .

Estimating 210 and/or validating 220 the position of the vehicle can be done, for example, using received signals from a satellite-based navigation system and/or using odometry data and/or using map data from the map memory 125 .

The validation 220 of the estimated position can be done, for example, by means of a number of statistically independent validators, which relate certain landmarks detected by sensors to the estimated position.

For example, in the in 1A illustrated situation, the position estimate 210 correctly result in that the vehicle 105 as turned on stands out on the right lane. This estimate may then be validated using, for example, a LiDAR-based validator using data from the LiDAR sensor 1151 and a camera-based validator using data from the camera 1152 . This can be done, for example, on the basis of a successful detection of the building 3 (contained in the map and/or in other sensor-specific reference data) both by means of the LiDAR sensor 1151 and by means of the camera 1152 .

It is now proposed that first information provided by the first sensor 1151 (LiDAR) is used to validate 220 the estimated position depending on second information provided by the second sensor 1152 (camera). In this case, the second information preferably indicates the reliability of validating the estimated position using the first information.

In this way, with reference to the present exemplary embodiment, it is possible, for example, to use the second information provided by the camera 1152 to predict in certain situations that a validator, based on the first information provided by the LiDAR sensor 1151 works, will probably fail.

Such a prediction can be implemented, for example, as part of a software component that can be executed by the processing device 120, which can also be referred to as a “predictor”. If the predictor predicts such a failure, e.g. a confirmation of the estimated vehicle position by the LIDAR-based validator can be considered less reliable. In this case, it can make sense not to use the first information (i.e. the LiDAR data) to validate the estimated position at all.

A prediction of a failure of the LiDAR-based validator should now be carried out as an example based on the 1B-C illustrated example situations.

The camera 1152 can be used, for example, to identify specific types of plants in the area surrounding the vehicle, which change their appearance (shape, color) significantly over time. This can e.g. as in 1B shown, concern deciduous trees 4, which can change their appearance greatly according to the seasons.

For example, the LiDAR sensor 1151 may not be able to detect certain aspects of a deciduous tree 4, or only to a limited extent. This relates in particular to the colors of the deciduous tree 4. In this respect, camera-based detection can compensate for certain weaknesses of the LiDAR sensor 1151. Aspects such as the colors or also the age of the deciduous tree 4 can be better recognized by means of the camera 1152 or a deciduous tree 4 can be recognized as such at all. As already mentioned above, machine learning can be used, for example, so that these specific aspects do not have to be determined in advance when developing a device 110 according to the invention.

In such a case, the camera-based predictor can determine, for example, that the LiDAR-based validator will probably fail because it will not reliably detect an expected landmark in the form of a leafy tree 4 and will not take it into account in the validation 220 . It is also conceivable, for example, for the camera 1152 to recognize that a roadside bush has been cut back and has consequently changed its shape, so that the LiDAR-based validator will probably have problems recognizing the bush as a landmark

Referring to 1C In another example situation, a truck 5 parked next to the freeway can be detected by the camera 1152, with the result that the LiDAR-based validator could incorrectly assume a crash barrier. Therefore, the LiDAR-based validator could assume, for example, that a correctly estimated lateral position (according to which the vehicle 105 is located approximately in the middle of the right lane, as shown) is wrong, since it places the vehicle 105 closer to the supposed left guardrail ( the truck 5) and is therefore located at least partially in the left lane. The LiDAR-based validator can therefore be used in the 1C shown situation will fail with a certain probability. For example, the LiDAR-based validator may not confirm an accurate position estimate or, potentially posing a greater security concern, may confirm an inaccurate position estimate. Therefore, it may be advantageous not to use the LiDAR-based validator to validate 220 the estimated position in the first place if the camera-based predictor assumes that the LiDAR-based validator is likely to fail.

For example, in the case of a predicted failure of a LiDAR-based validator, at least one other validator can be used instead. It is also conceivable that in this case an overall larger set of validators is used in order to still be able to consider the position estimate to be reliable.

In accordance with the examples described above, in addition to estimating 210 the position, the method according to one embodiment comprises the following steps, which are schematically illustrated in 2 B illustrated are: determining 215, depending on the second information, whether a sufficiently reliable validation of the estimated position based on the first information is possible; Validating 220a the estimated position using the first information if the determination 215 shows that a sufficiently reliable validation of the estimated position is possible using the first information; and validating 220b the estimated position without using the first information if the determination 215 shows that a sufficiently reliable validation of the estimated position based on the first information is not possible.

Optionally, the following (in 2 B not shown) step can be provided: if the determination 215 shows that a sufficiently reliable validation of the estimated position based on the first information is not possible, validating 220b the estimated position using information from at least one or more additional sources (apart from the first sensor 1151 ) from the totality of sources 115.

In this way, for example, a desired accuracy can be achieved overall, which can depend, for example, on a number of the statistically independent sources used.

Reference List

100

system

105

motor vehicle

110

contraption

115

Sources, sources of information

1151

first sensor; LiDAR

1152

second sensor; camera

120

processing facility

125

map storage

200

proceedings

210

estimate position

215

Determine the reliability of initial information (LiDAR data).

220

Validate position

220a

Validate position using first information

220b

Validate position without using the first information

3

building

4

deciduous trees

5

truck

QUOTES INCLUDED IN DESCRIPTION

This list of documents cited by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent Literature Cited

• DE 102019133316 [0001]

Claims (11)

Method (200) for determining a position of a vehicle (105), the method (200) comprising the following steps: - estimating (210) the position of the vehicle (105); and - validating (220) the estimated position using information provided by a plurality of sources (1151, 1152, 125) from a set of sources (115), the set of sources (115) comprising a first sensor (1151) and a second sensor (1152), and wherein first information provided by the first sensor (1151) is used in dependence on second information provided by the second sensor (1152) to validate (220) the estimated position . Method (200) according to claim 1 , wherein the type and/or number of the other sources (115), apart from the first sensor (1151), whose information is used for validation (220), are defined as a function of the second information. Method (200) according to any one of the preceding claims, wherein the second information indicates a reliability of validating the estimated position based on the first information. Method (200) according to claim 3 , comprising: - determining (215), depending on the second information, whether a sufficiently reliable validation of the estimated position is possible on the basis of the first information; - Validating (220a) the estimated position using the first information if the determination (215) shows that a sufficiently reliable validation of the estimated position using the first information is possible; - validating (220b) the estimated position without using the first information if the determination (215) shows that a sufficiently reliable validation of the estimated position based on the first information is not possible. Method (200) according to claim 4 , further comprising: - if the determination (215) shows that a sufficiently reliable validation of the estimated position based on the first information is not possible, validating (220b) the estimated position using information from at least one or more additional sources (115). the set of sources (115). Method (200) according to one of the preceding claims, wherein a pose of the vehicle, which in addition to the position of the vehicle comprises an orientation of the vehicle, is determined in the manner described (105). Device (110) for determining a position of a vehicle (105), the device (110) comprising: - a set of sources (115) comprising a first sensor (1151) and a second sensor (1152); - a processing device (120) arranged to carry out the following steps: o estimating (210) the position of the vehicle (105); and o validating (220) the estimated position using information provided by multiple sources (1151, 1152, 125) from a set of sources (115), the set of sources (115) including the first sensor (1151) and the second sensor (1152), and wherein first information provided by the first sensor (1151) is used in dependence on second information provided by the second sensor (1152) to validate (220) the estimated position . Device (110) according to claim 7 , wherein the first sensor (1151) is a LiDAR sensor. Device (110) according to claim 7 or 8th , wherein the second sensor (1152) is a camera. Device (110) according to any one of Claims 7 until 9 , further comprising a map memory for providing map data, wherein the estimation (210) and/or the validation (220) of the position takes place using the map data. Vehicle (105) comprising a device (110) according to one of Claims 7 until 10 .

Images