TWI723679B - Lane positioning system and lane positioning method - Google Patents

Abstract

A lane positioning system and a lane positioning method are provided. The lane positioning system includes a plurality of image capturing units, a vehicle positioning module, and a processor. The image capturing units acquire a first image of a road on the right side of the vehicle, the second image of a road on the left side of the vehicle, and the front lane image of the vehicle. The vehicle positioning module obtains a geographic location of the vehicle. The processor determines a number of lanes of the located road according to the geographic location, generates a plurality of lane marking values of the lanes and a positioning marking value of the vehicle according to at least one of the first image, the second image, and the front lane image, and performs a lane position to the vehicle according to the positioning marking value and the lane marking values.

Description

Lane positioning system and lane positioning method

The present invention relates to a lane positioning system and a lane positioning method, and particularly relates to a lane positioning system and a lane positioning method.

Currently, vehicles are driving on highways or fast roads, and the precise positioning of vehicles on the lanes requires a relatively high-level positioning system to assist the driver in deciding when to switch lanes. In particular, autonomous vehicles need to rely on high-resolution map data and expensive positioning systems, such as the use of optical radar (light detection and ranging, LiDAR). Therefore, how to accurately locate the vehicle lane and reduce the cost of the positioning system is one of the key development issues of the vehicle positioning system.

The invention provides a lane positioning system and a lane positioning method for accurate lane positioning of vehicles and low cost.

The lane positioning system of the present invention is used for lane positioning of vehicles. The lane positioning system includes a plurality of image capturing units, a vehicle positioning module, and a processor. A plurality of image capturing units are respectively arranged in the vehicle. The multiple image capturing units are used to obtain a first image on the right side of the vehicle, a second image on the left side of the vehicle, and an image of the front lane of the vehicle. The vehicle positioning module is installed on the vehicle. The vehicle positioning module is used to obtain the geographic location of the vehicle. The processor is coupled to the plurality of image capturing units and the vehicle positioning module. The processor is used for judging the number of multiple lanes of the positioning road where the vehicle is located based on the lane data corresponding to the geographic location, and marking the multiple lanes based on at least one pair of the first image, the second image, and the front lane image to generate Corresponding to the multiple lane marking values of the multiple lanes and the positioning mark value of the vehicle, and performing lane positioning of the vehicle according to the positioning mark value. The positioning road is at least one of an expressway and an expressway.

The lane positioning method of the present invention is suitable for lane positioning systems. The lane positioning system is used for lane positioning of the vehicle. The lane positioning system includes a plurality of image capturing units and a vehicle positioning module. The lane positioning method includes: obtaining a first image of the right side of the vehicle, a second image of the left side of the vehicle, and an image of the front lane of the vehicle by the plurality of image capturing units, and obtaining the geographic location of the vehicle by the vehicle positioning module Location; judging the number of multiple lanes of the positioning road where the vehicle is located based on the lane data corresponding to the geographic location; marking the multiple lanes based on at least one of the first image, the second image, and the front lane image to generate a correspondence A plurality of lane marking values of the plurality of lanes and a positioning mark value of a vehicle, wherein the positioning road is at least one of a freeway and an expressway; and the vehicle is positioned according to the positioning mark value.

Based on the foregoing, the lane positioning system and the lane positioning method of the present invention determine the number of lanes based on geographic location, and generate multiple lane marking values and vehicle positioning marks based on at least one of the first image, the second image, and the front lane image. Value, and locate the vehicle's lane based on the location mark value and the lane mark value. In this way, the present invention can realize accurate lane positioning. In addition, the lane positioning system and the lane positioning method of the present invention obtain the number of lanes, lane marking values, and positioning marking values through a plurality of image capturing units and vehicle positioning modules. Therefore, the lane positioning system and the lane positioning method of the present invention can be implemented at a lower cost.

In order to make the above-mentioned features and advantages of the present invention more comprehensible, the following specific embodiments are described in detail in conjunction with the accompanying drawings.

Please refer to FIG. 1 and FIG. 2 at the same time. FIG. 1 is a system diagram of a lane positioning system according to an embodiment of the present invention. Fig. 2 is a method flowchart of a lane positioning method according to an embodiment of the present invention. In this embodiment, the lane positioning system 100 is suitable for lane positioning of the vehicle 10. The vehicle 10 may be an autonomous vehicle or a non-autonomous vehicle of any vehicle type. In this embodiment, the lane positioning system 100 includes image capturing units CM1 to CM3, a vehicle positioning module 110, and a processor 120. The image capturing units CM1 to CM3 are respectively installed in the vehicle 10. In step S110, the image capturing units CM1 to CM3 obtain the first image MS1 of the road on the right of the vehicle 10, the second image MS2 of the road on the left of the vehicle 10, and the image MS3 of the front lane of the vehicle 10. For example, the image capturing unit CM1 is arranged on the right side of the vehicle 10. The image capturing unit CM1 obtains the first image MS1 of the road on the right of the vehicle 10. The image capturing unit CM2 is installed on the left side of the vehicle 10. The image capturing unit CM2 obtains the second image MS2 of the road on the left of the vehicle 10. The image capturing unit CM3 is, for example, installed on a rear-view mirror inside the vehicle 10. The image capturing unit CM3 obtains the image MS3 of the front lane of the vehicle 10. The image capturing unit CM3 can capture all lanes of the field of view in front of the vehicle 10. The image capturing units CM1 to CM3 can be implemented by multiple cameras or at least one set of driving recorders. For ease of description, this embodiment takes three image capturing units as an example. However, there may be multiple image capturing units in the present invention, which is not limited to this embodiment.

In some embodiments, the image capturing unit CM1 of the lane positioning system 100 may obtain the first image MS1 of the road on the right of the vehicle 10 and a part of the front lane image MS3. The image capturing unit CM2 can obtain the second image MS2 of the road on the left of the vehicle 10 and another part of the front lane image MS3. In other words, the lane positioning system 100 can obtain the first image MS1, the second image MS2, and the front lane image MS3 through the image capturing units CM1 and CM2.

In this embodiment, the vehicle positioning module 110 is installed in the vehicle 10. In step S110, the vehicle positioning module 110 obtains the geographic location where the vehicle 10 is located. In this embodiment, the vehicle positioning module 110 can obtain the geographic location of the vehicle 10 through a Global Positioning System (GPS).

In this embodiment, the processor 120 is coupled to the image capturing units CM1 to CM3 and the vehicle positioning module 110. In step S120, the processor 120 determines the number of multiple lanes of the positioning road where the vehicle 10 is located according to the lane data LD1 corresponding to the geographic location. In this embodiment, the aforementioned lane data LD1 may come from the geographic location provided by the vehicle positioning module 110. The processor 120 may obtain the lane data LD1 of the positioning road from the map data of the geographic location (the present invention is not limited to this). In step S130, the processor 120 marks at least one pair of lanes of the first image MS1, the second image MS2, and the front lane image MS3 to generate a lane marking value LV corresponding to the lane and a positioning marking value PV of the vehicle 10. After learning the lane marking value LV and the positioning marking value PV, the processor 120 will perform lane positioning on the vehicle 10 according to the positioning marking value PV and the lane marking value LV in step S140. In this embodiment, the aforementioned location road is at least one of an expressway and an expressway. In other words, the lane positioning system 100 may be suitable for lane positioning of expressways and expressways. The processor 120 is, for example, a central processing unit (Central Processing Unit, CPU), or other programmable general-purpose or special-purpose microprocessor (Microprocessor), digital signal processor (Digital Signal Processor, DSP), programmable Controller, Application Specific Integrated Circuits (ASIC), Programmable Logic Device (PLD) or other similar devices or a combination of these devices, which can load and execute computer programs.

It is worth mentioning here that the lane positioning system 10 and the lane positioning method of this embodiment determine the number of lanes based on geographic locations, and generate them based on at least one of the first image MS1, the second image MS2, and the front lane image MS3. A plurality of lane marking values LV and a positioning mark value PV of the vehicle 10 are used, and the vehicle 10 is positioned according to the lane marking value LV and the positioning mark value PV. In this way, the present invention can realize accurate lane positioning.

In addition, the lane positioning system 10 and the lane positioning method of this embodiment obtain the number of lanes and the lane information LD1 through the image capturing units CM1 to CM3 and the vehicle positioning module 110, thereby generating the lane marking value LV and positioning Marked value PV. Therefore, the lane positioning system and the lane positioning method of the present invention can be implemented at a lower cost.

To further describe the implementation details of steps S120 and S130, please refer to FIG. 1, FIG. 2, FIG. 3, and FIG. 4 at the same time. FIG. 3 is a schematic diagram of generating lane marking values according to an embodiment of the present invention. FIG. 4 is a flowchart of the method shown in step S130. In this embodiment, the processor 120 may determine that the location road where the vehicle 10 is located has 5 lanes L1 to L5 in step S120 based on the lane data LD1 corresponding to the geographic location.

In this embodiment, step S130 includes steps S131 to S135. In step S131, the lane positioning system 100 uses the processor 120 to initialize the lane marking value LV and the positioning marking value PV. In step S131a in step S131, the processor 120 sets the lane marking value LV corresponding to the outside lane (ie, lane L1) to "1", and sets the lane corresponding to the inside lane (ie, lane L5) The value LV is set to "N". N is equal to the number of lanes L1 to L5 of the positioning road and is a positive integer greater than 1. Therefore, the lane marking value LV corresponding to the inside lane is set to "5". In addition, the lane marking values LV of the remaining lanes L2 to L4 are also set. From right to left, the lane marking value LV of lane L2 is set to "2", the lane marking value LV of lane L3 is set to "3", and the lane marking value LV of lane L4 is set to "4". In step S131a, the processor 120 recognizes according to the first image MS1 that the lane L1 whose right side line is a solid line (for example, a white solid line) is an outside lane, and sets the lane marking value corresponding to the outside lane to "1" . The right and left side lines of lanes L2~L4 are dotted lines. Therefore, the lane marking values LV of lanes L2~L4 will be set in order from right to left, and will be incremented starting from "2". Therefore, the lane marking value LV of the lane L2 is set to "2", the lane marking value LV of the lane L3 is set to "3", and the lane marking value LV of the lane L4 is set to "4".

In some embodiments, the processor 120 recognizes from the first image MS1 that the right side line is a solid line and the left side line is a dashed lane L1 is an outside lane, and sets the lane marking value LV corresponding to the outside lane to "1" . Based on the second image MS2, the processor 120 recognizes that the left side line is a solid line (eg, a yellow solid line) and the right side line is a dashed line is an inside lane, and sets the lane marking value LV corresponding to the inside lane to "5" .

In step S131b in step S131, when the processor 120 determines that the vehicle 10 is traveling in an outside lane according to at least one of the first image MS1 and the second image MS2, the positioning flag value PV is set to "1". When the processor 120 determines that the vehicle 10 is traveling in the interior lane according to at least one of the first image MS1 and the second image MS2, the positioning flag value PV is set to "N", that is, "5".

In addition, when the processor 120 determines that the vehicle 10 is not driving in an inner lane or an outer lane, it will determine that the vehicle 10 is driving in one of the lanes L2 to L4 according to the front lane image MS3, the first image MS1, and the second image MS2. For example, when the processor 120 determines that the lane on which the vehicle 10 is traveling is the left lane adjacent to the outer lane based on the first image MS1 and the front lane image MS3, it knows that the vehicle 10 is currently traveling in the lane L2. Therefore, the processor 120 sets the positioning flag value PV to "2". For another example, when the processor 120 determines that the lane on which the vehicle 10 is traveling is the right lane adjacent to the inner lane based on the second image MS2 and the front lane image MS3, it knows that the vehicle 10 is currently traveling in the lane L4. Therefore, the processor 120 sets the positioning flag value PV to "4". For another example, when the processor 120 determines that the adjacent right lane of the lane in which the vehicle 10 is traveling is not an outside lane, and the adjacent left lane is not an inside lane, the positioning flag value PV is set to "3".

In this embodiment, when the positioning flag value PV is set to "1", it means that the vehicle 10 is traveling on the lane L1 in which the lane flag value LV is set to "1". When the positioning mark value PV is set to "2", it means that the vehicle 10 is traveling on the lane L2 where the lane mark value LV is set to "2", and so on. Therefore, the positioning mark value PV may indicate that the vehicle 10 is traveling on a lane with the same lane marking value LV as the positioning mark value PV.

In this embodiment, the lane positioning system 100 may execute step S131 when entering the positioning road. The lane positioning system 100 may also execute step S131 when entering the outer lane or the inner lane of the positioning road.

In step S132, the lane positioning system 100 uses the processor 120 to determine whether the vehicle 10 switches lanes. If the processor 120 determines from the front lane image MS3 that the vehicle 10 has completely crossed the left dotted line/right dotted line and successfully entered the next lane, the processor 120 knows that the vehicle 10 has switched lanes. The lane positioning system 100 will execute step S133. On the other hand, if the processor 120 determines that the vehicle 10 has not completely crossed the left/right dashed line according to the front lane image MS3, the processor 120 knows that the vehicle 10 has not successfully switched lanes. The lane positioning method returns to step S132.

In step S133, the lane positioning system 100 determines through the processor 120 that the vehicle 10 has successfully switched a lane to the left or has successfully switched a lane to the right. If the processor 120 determines from the front lane image MS3 that the vehicle 10 has completely crossed the left dashed line and successfully entered the next lane, the lane positioning method proceeds to step S134. In step S134, the processor 120 adds 1 to the positioning flag value PV. In this embodiment, after step S134 ends, the lane positioning method returns to step S132.

On the other hand, if the processor 120 determines from the front lane image MS3 that the vehicle 10 has completely crossed the left dashed line and successfully entered the next lane, the lane positioning method proceeds to step S135. In step S135, the processor 120 subtracts one from the value of the positioning flag. After step S135 is completed, the lane positioning method will return to step S132.

For example, if the vehicle 10 successfully switches to the left from the lane L1 to the lane L2, the positioning flag value PV will be set from "1" to "2". If the vehicle 10 successfully switches to the left from the lane L2 to the lane L3, the positioning flag value PV will be set from "2" to "3". If the vehicle 10 successfully switches to the left from the lane L3 to the lane L2, the positioning flag value PV will be set from "3" to "2". It can be seen from this that the positioning mark value PV can indicate that the vehicle 10 is traveling on a lane with the same lane marking value LV as the positioning mark value PV. In this way, the processor 120 can locate the lane of the vehicle 10 according to the positioning mark value PV and the lane mark value LV in step S140.

Please return to the embodiment of FIG. 1 and FIG. 2. In this embodiment, at least one of the image capturing units CM1 to CM3 can also obtain at least one landmark image in front of the vehicle 10. For example, the image capturing unit CM3 can obtain a road sign image MK1 in addition to the front lane image MS3. The processor 120 receives the road sign image MK1 and the front lane image MS3, and determines the actual number of lanes of the positioning road based on the road sign image MK1 and the front lane image MS3, so as to generate lane data LD2 associated with the road sign image MK1. The landmark image MK1 in this embodiment may be an image of a region of interest (ROI) in the front image. It should be noted that the lane data LD2 associated with the road sign image MK1 is the current actual number of lanes on which the road is located. Therefore, the lane data LD2 associated with the road sign image MK1 is more accurate information.

In this embodiment, the lane positioning system 100 further includes a lane database 130. The lane database 130 is coupled to the vehicle positioning module 110 and the processor 120. The lane database 130 is used to store lane data, such as lane data LD1 corresponding to the geographic location and/or lane data LD2 associated with the road sign image MK1. In this embodiment, the lane database 130 also provides the stored lane data to the processor 120. Therefore, once the processor 120 receives the road sign image MK1, the lane database 130 will provide the lane data LD2 associated with the road sign image MK1 to the processor 120. Once the processor 120 receives the geographic location obtained by the vehicle positioning module 110, the lane database 130 will provide the lane data LD1 corresponding to the geographic location to the processor 120.

In this embodiment, the processor 120 can also compare the front lane image MS3 with at least one of the lane data LD1 and LD2. To further explain, the processor 120 will determine whether the front lane image MS3 is consistent with the current lane data LD1 and LD2. When the processor 120 determines that the front lane image MS3 is inconsistent with the current lane data LD1 and LD2, it indicates that the current lane data LD1 and LD2 need to be updated. Therefore, the processor 120 updates the current lane data LD1 and LD2, and stores the updated lane data in the lane database 130. On the other hand, when the processor 120 determines that the front lane image MS3 is consistent with the current lane data LD1 and LD2, the processor 120 will not update the current lane data.

In some embodiments, the lane database 130 may be set in the cloud. Therefore, the lane data LD1 and LD2 can be shared with the lane positioning system of other vehicles or updated by the lane positioning system of other vehicles.

In summary, the lane positioning system and lane positioning method of the present invention determine the number of lanes based on geographic location, and generate multiple lane marking values and vehicle status based on at least one of the first image, the second image, and the front lane image. Locate the mark value, and perform lane positioning on the vehicle and the lane mark value according to the positioning mark value. In this way, the present invention can realize accurate lane positioning. In addition, the positioning system and lane positioning method of the present invention obtain the number of lanes, lane marking values, and positioning marking values through multiple image capturing units and vehicle positioning modules. Therefore, the lane positioning system and the lane positioning method of the present invention can be implemented at a lower cost.

Although the present invention has been disclosed in the above embodiments, it is not intended to limit the present invention. Anyone with ordinary knowledge in the relevant technical field can make some changes and modifications without departing from the spirit and scope of the present invention. The scope of protection of the present invention shall be determined by the scope of the attached patent application.

10: Vehicle 100: Lane positioning system 110: Vehicle positioning module 120: processor 130: Lane database CM1, CM2, CM3: image capture unit L1, L2, L3, L4, L5: Lane LD: Lane information LV: Lane marking value MK1: Road sign image MS1: First image MS2: Second image MS3: Image of the front lane PV: Positioning mark value S110, S120, S130, S140: steps S131, S131a, S131b, S132, S133, S134, S135: steps

Fig. 1 is a system schematic diagram of a lane positioning system according to an embodiment of the present invention. Fig. 2 is a method flowchart of a lane positioning method according to an embodiment of the present invention. FIG. 3 is a schematic diagram of generating lane marking values according to an embodiment of the present invention. FIG. 4 is a flowchart of the method shown in step S130.

10: Vehicle

100: Lane positioning system

110: Vehicle positioning module

120: processor

130: Lane database

CM1, CM2, CM3: image capture unit

LD: Lane information

LV: Lane marking value

MK1: Road sign image

MS1: First image

MS2: Second image

MS3: Image of the front lane

PV: Positioning mark value

Claims (10)

A lane positioning system is used to locate a lane of a vehicle. The lane positioning system includes: a plurality of image capturing units respectively arranged on the vehicle to obtain a first image of the road on the right side of the vehicle, the vehicle A second image of the road on the left and a front lane image of the vehicle; a vehicle positioning module arranged on the vehicle to obtain a geographic location where the vehicle is located; and a processor coupled to the images The capturing unit and the vehicle positioning module are used to determine the number of lanes of a positioning road where the vehicle is located according to a first lane data corresponding to the geographic location, according to the first image, the second image, and At least one pair of the lanes of the front lane image is marked to generate a plurality of lane marking values corresponding to the lanes and a positioning mark value of the vehicle, and the positioning mark value and the lane marking values The vehicle performs the lane positioning, wherein when the processor determines that the vehicle has switched a lane to the left according to the front lane image, the positioning mark value is increased by 1, wherein when the processor determines that the vehicle is based on the front lane image When a lane has been switched to the right, the location mark value is reduced by 1, where the location road is at least one of a highway and an expressway. For example, in the lane positioning system described in item 1 of the scope of patent application, the processor recognizes according to the first image that the lane whose right side line is a solid line is an outside lane among the lanes, and will correspond to the outside lane The lane marking value is set to 1, and According to the second image, it is recognized that the lane whose left side line is a solid line is an inside lane among the lanes, and the lane marking value corresponding to the inside lane is set to N, where N is equal to the number of the lanes of the positioning road The number is a positive integer greater than 1. For example, the lane positioning system described in item 2 of the scope of patent application, wherein: when the processor determines that the vehicle is driving in the outer lane according to at least one of the first image and the second image, the positioning mark value is Set to 1, when the processor determines that the vehicle is driving in the inner lane according to at least one of the first image and the second image, set the positioning mark value to N. For example, the lane positioning system according to the first item of the patent application, wherein: at least one of the image capturing units is also used to obtain at least one road mark image in front of the vehicle, and the processor is also used to receive the at least one road mark image Based on the image and the front lane image, determine the actual number of multiple lanes of the positioning road based on the at least one road landmark image and the front lane image, so as to generate a second lane data associated with the at least one road landmark image. The lane positioning system described in item 4 of the scope of patent application further includes: a lane database coupled to the processor for storing at least one of the first lane data and the second lane data. For the lane positioning system described in item 4 of the scope of patent application, the processor is also used to compare the image of the front lane with the data of the first lane and the second lane. At least one of the lane data is used to determine whether at least one of the first lane data and the second lane data needs to be updated. A lane positioning method is suitable for a lane positioning system, the lane positioning system is used to perform a lane positioning for a vehicle, the lane positioning system includes a plurality of image capturing units and a vehicle positioning module, wherein the lane positioning method includes : Obtain a first image of the road on the right of the vehicle, a second image of the road on the left of the vehicle, and an image of the front lane of the vehicle by the image capturing units, and obtain the vehicle by the vehicle positioning module A geographic location where the vehicle is located; determine the number of lanes on a location road where the vehicle is located based on a first lane data corresponding to the geographic location; based on the first image, the second image, and the front lane image At least one pair of the lanes is marked to generate a plurality of lane marking values corresponding to the lanes and a positioning mark value of the vehicle, wherein the positioning road is at least one of an expressway and an expressway; and according to the The positioning mark value and the lane mark values perform the lane positioning on the vehicle, wherein the step of obtaining the lane mark values corresponding to the lanes and the positioning mark value of the vehicle includes: when it is determined according to the image of the front lane When the vehicle has switched a lane to the left, the positioning mark value is increased by 1; and when it is determined that the vehicle has switched a lane to the right according to the image of the front lane, the positioning mark value is decreased by 1. For the lane positioning method described in item 7 of the scope of patent application, the step of initializing the lane marking values and the positioning marking value includes: identifying, according to the first image, the lanes whose right side line is a solid line are the lanes And set the lane marking value corresponding to the outer lane to 1; and according to the second image, it is recognized that the lane whose left side line is a solid line is an inner lane among the lanes, and will correspond to The lane marking value of the inner lane is set to N, where N is equal to the number of lanes of the positioning road and is a positive integer greater than 1. For the lane positioning method described in item 8 of the scope of patent application, the step of initializing the lane marking values and the positioning marking value further includes: determining according to at least one of the first image and the second image When the vehicle is driving in the outer lane, the positioning mark value is set to 1; and when it is determined that the vehicle is driving in the inner lane according to at least one of the first image and the second image, the positioning mark value is Set to N. For example, the lane positioning method according to claim 7, wherein the lane positioning method further includes: obtaining at least one landmark image in front of the vehicle by at least one of the image capturing units; and receiving the at least one landmark image Based on the image and the front lane image, determine the actual number of multiple lanes of the positioning road based on the at least one road landmark image and the front lane image, so as to generate the second lane data associated with the at least one road landmark image.

Images