Lane detection system and lane keeping system

Abstract

Lane detection system (10) with - an optical lane detection system (12) and - a GNSS-based position detector (14), wherein - the GNSS-based position detector (14) is designed to determine information on a lane path from digital map data (16) depending on a detected position, a known position detection accuracy (44) and known accuracy limits (42) of the digital map data and to transmit data on the determined lane path to the optical lane detector, and wherein - the optical lane detection device (12) is configured to take into account the data on the determined lane course from the position detection device during lane detection, characterized in that the GNSS-based position detection device (14) is further configured to - depending on the detected position, information on the lane course can be determined from the digital map data (16), - to compare the determined lane path with the known position detection accuracy (44) and the known accuracy limits (42) of the digital map data, and - to transmit data on the determined lane path to the optical lane detector (12) if the determined lane path remains unchanged during a distance that is greater than the sum of the position detection accuracy (44) and the known accuracy limits (42) of the digital map data.

Description

[0001] The invention relates to a lane detection system and a lane keeping system.

[0002] Lane keeping systems currently installed in production vehicles use optical sensors to detect lane markings or lane-specific features such as road boundaries in images captured by a camera or other optical sensors. Based on information about detected lane markings or lane-specific features, their position relative to the vehicle, the detected further course of the lane, and a calculated future trajectory of the vehicle's position relative to the lane markings or lane-specific features, the system calculates when the vehicle is likely to leave the lane. Using this information, the lane keeping system issues a warning to the driver or even actively intervenes in the steering to keep the vehicle within the lane.

[0003] US patent 8,498,782 B2 discloses a lane-keeping system that, in addition to information from optical and distance sensors, also evaluates information from a GNSS (Global Navigation Satellite System). The GNSS information is used to determine the approximate position of a vehicle on a roadway. The precise vehicle position on the roadway is then determined using further information from the optical and distance sensors.

[0004] From DE 10 2006 040 334 A1 a method for lane detection with a driver assistance system and driver assistance system is known.

[0005] From DE 10 2013 003 216 A1 a method for determining a lane for steering control of an automated controlled vehicle is known.

[0006] Redundant lane detection systems for fault-tolerant vehicle transverse control units are known from DE 10 2012 100 164 A1.

[0007] From DE 10 2010 033 729 A1 a method and a device for determining the position of a vehicle on a roadway as well as a motor vehicle with such a device are known.

[0008] A disadvantage of these known lane-keeping systems, however, is that they rely on the visibility of lane markings or other features characteristic of lanes. The optical sensors only detect lane markings if they are also visible to the human eye. Therefore, road surface contamination, worn markings, poor lighting conditions, and weather conditions, etc., can prevent lane markings from being detected, rendering the lane-keeping system ineffective. Another disadvantage of GNSS-supported lane-keeping systems, such as the one described in US 8,498,782 B2, is that GNSS-based positioning is currently still too imprecise to guarantee a sufficiently accurate position for a lane-keeping system.The GNSS accuracy of navigation systems currently installed in production vehicles is around 10 meters, which is far from the accuracy required for a lane-keeping system. Another problem lies in the often inaccurate map data from production navigation systems. Therefore, the vehicle's position within the lane cannot be determined with sufficient accuracy using GNSS data. What is needed is an accuracy of a few centimeters for both the GNSS positioning and the map data.

[0009] The object of the present invention is therefore to propose a GNSS-supported lane keeping system which, despite the aforementioned disadvantages of GNSS support for lane keeping, can be advantageously used in vehicles.

[0010] This problem is solved by the subject matter of the independent claims. Further embodiments of the invention are described in the dependent claims.

[0011] The present invention utilizes the fact that digital GNSS map data now exhibits a high degree of accuracy, sometimes within a range of approximately 20 cm, with a tendency towards ever-increasing accuracy, to enable the determination of a lane's course based on GNSS information, taking into account the known accuracy limits of the digital map data and the known position detection accuracy in specific driving situations. The invention takes advantage of the fact that road segments in digital map data are generally stored in lists with attributes. For example, an entry in such a list can contain information about the length of a specific, constant road segment, such as 3 lanes, each 4.10 m wide, straight section, length 500 m, or 2 lanes, each 3.5 m wide, right-hand curve with a radius of 1000 m, and a curve segment length of 300 m.If the known position detection accuracy of the GNSS is approximately 10 m and the known accuracy limits of the digital map data are approximately 0.5 m, road sections that remain unchanged for longer than 21 m (each at 10.5 m at the road section boundary) can be determined as an interpolated lane path and transmitted to an optical lane detection system, which can then perform lane detection for the entire road section even in poor visibility conditions with only partially recognized road markings.

[0012] In the aforementioned example of the curve, which is 300 m long, as soon as a small part of the lane is detected by the optical sensors, the remaining part of the section, minus the limits due to the inaccuracies of the digital map data and the GNSS, i.e. the inner 279 m of the section, could be interpolated relative to the vehicle, and the lane detection system would also work here if the remaining part of the section is not optically detected.

[0013] According to the invention, a lane detection system is further provided with an optical lane detector and a GNSS-based position detector, wherein the GNSS-based position detector is configured to determine information on a lane path from digital map data depending on a detected position, a known position detection accuracy and known accuracy limits of the digital map data and to transmit data on the determined lane path to the optical lane detector, and wherein the optical lane detector is configured to take the data on the determined lane path from the position detector into account during lane detection.Unlike conventional GNSS-based lane detection systems, this system does not simply output a position detected by GNSS to an optical lane detector, but rather determines, or more precisely interpolates, a lane path taking into account the known position detection accuracy and known accuracy limits of the digital map data.

[0014] Likewise, the invention does not attempt to directly determine the absolute position of the vehicle within the lane based on GNSS positioning and map data, which fails due to accuracy limitations, in order to provide data to the lane recognition system.

[0015] The GNSS-based position detector can further be configured to determine lane-path information from the digital map data depending on the detected position, to compare the determined lane-path with the known position detection accuracy and the known accuracy limits of the digital map data, and to transmit lane-path data to the optical lane-detector if the determined lane-path remains unchanged over a distance greater than the sum of the position detection accuracy and the known accuracy limits of the digital map data.

[0016] The optical lane recognition system can also be designed to perform lane recognition based on image data, to compare the lane recognized based on image data with the data on the determined lane path from the position recognition system, and to signal deviations between the lane recognized based on image data and the data on the lane path determined by the position recognition system if these deviations exceed a predetermined value.

[0017] The optical lane recognition system can also be designed to perform lane recognition using image data and, in case of problems with lane recognition using image data, to use the data on the determined lane path from the position recognition system for lane recognition.

[0018] The optical lane recognition system can also be designed to use the data on the determined lane path from the position recognition system for lane recognition and to perform lane recognition based on image data if no data on the determined lane path from the position recognition system is available.

[0019] The lane detection system may also include an interface for outputting data from the optical lane detector and / or an interface from the GNSS-based position detector to a database containing digital map data.

[0020] In a further embodiment, the invention relates to a lane keeping system for a vehicle comprising a lane detection system according to the invention and as described herein, a comparison unit for comparing lane detection information with current driving information and generating an output signal when the current driving information deviates from the lane detection information, an output interface for issuing warning signals to users depending on the output signal of the comparison unit, and / or a steering assistance device configured to actively intervene in the vehicle's steering system depending on the output signal of the comparison unit.

[0021] Further advantages and possible applications of the present invention will become apparent from the following description in conjunction with the embodiment(s) shown in the drawing(s).

[0022] The terms and reference numerals used in the list of reference numerals at the end are used in the description, claims, summary and drawing(s).

[0023] The drawing(s) show / show in Fig. 1 a block diagram of an embodiment of a lane keeping system with a lane detection system according to the invention; and Fig. 2 an example of a lane layout and its determination according to the invention.

[0024] In the following description, identical, functionally equivalent, and functionally related elements may be designated with the same reference numerals. Absolute values ​​are given below only as examples and are not to be understood as limiting the invention.

[0025] Fig. Figure 1 shows the block diagram of a lane keeping system 22, which is primarily designed for optical lane detection by processing image data generated by a camera 18 that captures the area in front of a vehicle equipped with the system 22 and can, for example, be located in the vehicle's central rearview mirror. For this purpose, the system 22 includes a lane detection system 10, a comparison unit 24, and an output interface 30 for warning signals and / or a steering assistance device 32. The lane detection system 10 includes an optical lane detector 12 and a GNSS-based position detector 14. The system 22 is also connected via corresponding data communication interfaces to a GNSS receiver 20, for example, for GPS, GLONASS, or Galileo signals, and to a database 16 containing digital map data for vehicle navigation.Via a further interface to a vehicle bus, the system 22, more precisely the comparator unit 24, can receive current driving information 26 such as data from wheel sensors, steering angle sensors, yaw rate sensors, acceleration sensors, etc. The steering assistance device 32 of the system 22 outputs data to steering actuators, which can influence the steering of the vehicle, in particular, can perform active steering interventions.

[0026] The following section explains how the lane keeping system 22 works, in particular how the lane detection system 10 functions according to the invention. For this purpose, reference is made to the information contained in the Fig. The lane layout shown in section 2 is used for explanation.

[0027] The image data from camera 18 are fed to the optical lane recognition system 12 of the lane recognition system 10 and processed by it in order to determine a lane course based on lane markings or lane-characteristic features as described above.

[0028] The GNSS-based position detector 14 receives GNSS data from the GNSS receiver 20 via the corresponding interface, which it uses to calculate or detect the current position of the vehicle. The position thus detected can only be determined with a known position detection accuracy of, for example, 10 m. Based on the detected position, the position detector 14 reads a data record from the database 16 containing data for the section of the route corresponding to the detected position. Fig. 2. For example, vehicle 34 is located on a track section 1, a straight 300 m long section of road 38 of a two-lane roadway 36. The data set therefore contains, in simplified terms, the following data: 2 lanes Lane width: 4.10 m each straight section of track Length 300 m

[0029] The position detector 14 determines, based on the data for the route segment and the known position detection accuracy 44 ( Fig. 2) and the known accuracy limits 42 ( Fig. 2) The lane path is determined from the digital map data as follows: first, the values ​​of the known position detection accuracy 44 (e.g., 10 m) and the known accuracy limits 42 (e.g., 0.5 m) are added. The sum corresponds approximately to a radius of 10.5 m around the detected position. In other words, the vehicle 34 is located approximately within a circle with a radius of 10.5 m around the known position. It follows that road segments that remain unchanged for more than 21 m can be determined as the lane path.In the present case, the position detector 14 therefore compares the route contained in the above data set, more precisely the specified length of 300 m, with twice the sum of the position detection accuracy 44 (10 m) and the known accuracy limits 42 (0.5 m) and outputs the data from the data set as the lane path, since, according to the data set, the lane path remains unchanged over a distance corresponding to twice the sum. If, on the other hand, the data set contained a specification such as "straight section of track, length 20 m", this lane path would not be output, since the length of the straight section would not exceed the inaccuracies of position detection and digital maps, and therefore this lane path, due to its shortness, would not be suitable for interpolation with respect to the inaccuracies.The above procedure is not limited to examples with straight sections of track, but can also be applied to curves.

[0030] The position sensor 14 outputs data on the determined lane alignment to the optical lane detector 12 (e.g., straight section of road length 279 m = 300 m - 2 * 10.5 m). The optical lane detector 12 can take this data into account in various ways during lane detection: For example, if optical lane detection fails, particularly in poor visibility conditions, the optical lane recognition system 12 can interpolate the lane path based on the received data and therefore also operate without optical lane detection. For example, if poor visibility conditions cause the system to operate in a curve section 40 ( Fig.2) If the optical lane detection fails or does not provide usable results on a road with a length of 300 m, a curve radius of 1000 m, and a two-lane carriageway with a lane width of 4.10 m each, the optical lane detection system 12 can switch directly to processing the data received from the position detection system 14 and interpolate the further lane path based on this data. As soon as visibility improves, the optical lane detection system 12 can switch back to optical lane detection. In this case, the position detection system 14 is used as a replacement for optical lane detection when problems arise with lane detection based on image data.

[0031] The optical lane detection unit 12 can also be configured to compare a lane detected using image data with the data on the determined lane path from the position detection unit 14 and to signal deviations if these exceed a predefined value. In this case, the position detection unit 14 is used to verify or validate the optical lane detection.

[0032] Note: The lane detection system can only interpolate if the optical sensor has already detected a small portion of the road. Therefore, as far as I understand, what is described here is not possible – or am I misunderstanding something?

[0033] The lane detection system 10 outputs information on the determined lane path to the comparison unit 24, which compares the received information with the current lane path information 26, i.e., data from wheel sensors, steering angle sensors, yaw rate sensors, acceleration sensors, etc. Comparison here means that the comparison unit determines whether the current lane path information 26 matches the information on the determined lane path.If the current driving path information 26 deviates from the information on the determined lane path, for example if a strong steering input over a longer period is compared as current driving path information and a straight section of the route as information on the determined lane path, the comparison unit 24 outputs an output signal 28 via an output interface, which signals the deviation of the current driving path information from the information on lane recognition, for example indicating the degree or type of deviation.

[0034] Depending on the output signal 28 of the comparator unit 24, the output interface 30 outputs warning signals, for example, a warning indicator on a display device, an audible signal, or a haptic signal. The output signal 28 is also supplied to the steering assistance device 32, which actively intervenes in the vehicle's steering system accordingly. For example, in the case of a sharp steering input signaled by the output signal 28 over a longer period on a straight section of road, the steering assistance device makes an active steering intervention to counteract the sharp steering input.

[0035] The invention makes lane keeping systems more stable because the relatively frequent failures of conventional purely optical lane detection due to lane marking recognition problems can be reduced by supplementing it with GNSS-based lane path determination according to the invention. The improved stability of lane detection and lane keeping systems according to the invention also increases safety. Reference sign 10 Lane Detection System 12 optical lane recognition systems 14 GNSS-based position detectors 16 Database with digital map data 18 Camera 20 GNSS receivers 22 Lane Keeping System 24 comparison units 26 current journey information 28 Output signal of the comparator unit 24 30 Output interface for warning signals 32 Steering support device 34 Vehicle with lane keeping system 22 36 two-lane roadway 38 straight section of road 40 Curve section 42 Accuracy of digital map data 44 GNSS position detection accuracy

Claims

[1] Lane detection system (10) with - an optical lane detection system (12) and - a GNSS-based position detector (14), wherein - the GNSS-based position detector (14) is designed to determine information on a lane path from digital map data (16) depending on a detected position, a known position detection accuracy (44) and known accuracy limits (42) of the digital map data and to transmit data on the determined lane path to the optical lane detector, and wherein - the optical lane detection system (12) is designed to take into account the data on the determined lane course from the position detection system during lane detection, characterized by , that the GNSS-based position detector (14) is further trained to - depending on the detected position, information on the lane course can be determined from the digital map data (16), - to compare the determined lane path with the known position detection accuracy (44) and the known accuracy limits (42) of the digital map data, and - to transmit data on the determined lane path to the optical lane detector (12) if the determined lane path remains unchanged during a distance that is greater than the sum of the position detection accuracy (44) and the known accuracy limits (42) of the digital map data. [2] Lane detection system according to claim 1, characterized by , that the optical lane detection system (12) is further designed to - to perform lane detection using image data, - to compare the lane detected using image data with the data on the determined lane course from the position detector (14), and - To signal deviations between the lane detected using image data and the lane path data determined by the position detector (14) when these deviations exceed a predetermined value. [3] Lane detection system according to claim 1 or 2, characterized by , that the optical lane detection system (12) is further designed to - to perform lane detection using image data and - in case of problems with lane detection using image data, use the data on the determined lane course from the position detector (14) for lane detection. [4] Lane detection system according to claim 1, 2 or 3, characterized by , that the optical lane detection system (12) is further designed to - to use the data on the determined lane path from the position detector (14) for lane detection and - to perform lane detection using image data if no data on the determined lane path is available from the position detector (14). [5] Lane detection system according to any one of the preceding claims, characterized by , that an interface for outputting data from the optical lane detection system (12) and / or an interface from the GNSS-based position detection system (14) to a database containing the digital map data is or are provided. [6] Lane keeping system (22) for a vehicle with - a lane detection system (10) according to one of the preceding claims, - a comparison unit (24) for comparing lane detection information with current driving path information (26) and generating an output signal (28) when the current driving path information deviates from the lane detection information, - an output interface (30) for outputting warning signals to users depending on the output signal of the comparator unit and / or - a steering assistance device (32) which is designed to actively intervene in the steering system of the vehicle depending on the output signal of the comparator unit.

Images