Unmanned container truck intersection identification method and system based on map data

Abstract

The invention provides an unmanned container truck intersection identification method based on map data, and the method comprises the steps: 1, adding the longitudes and latitudes of four vertexes of an electronic fence on an unmanned platform, and constructing an intersection electronic fence; step 2, performing one-to-one correspondence on the intersection electronic fences and intersection areas in the map data of the unmanned container truck operation area; 3, the unmanned container truck sends the vehicle message to the unmanned driving platform in real time; step 4, selecting lane lines associated with the intersection area in the vehicle path planning lane line set; step 5, judging whether the current lane line of the vehicle is a lane line associated with the intersection area or not, and if so, judging whether the distance between the unmanned container truck and the intersection electronic fence corresponding to the intersection area is smaller than a threshold value or not according to the real-time latitude and longitude information of the vehicle; and step 6, if yes, determining that the current driving vehicle arrives at the intersection. The technical problems that the prior art cannot be suitable for scene migration and the implementation cost is high are effectively solved.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and in particular to a method and system for identifying unmanned truck intersections based on map data. Background Art

[0002] In the field of autonomous driving, map navigation is the premise of end-to-end automatic driving. Navigation technology actually solves two problems. One is path planning, that is, calculating the road or lane numbers covered by the end-to-end path; the other is to formulate corresponding actions according to the intersections passed in the path, guide the vehicle to achieve correct actions at the intersection, and ensure that the vehicle remains on the planned path. In the prior art, the identification of intersections in the driving area includes: obtaining the distance between the traffic light at the intersection and the current driving vehicle; judging whether the vehicle is close to the intersection based on the distance. However, based on the unmanned container port environment, many intersections do not have traffic lights. At this time, the intersection cannot be determined based on the distance between the traffic light and the current driving vehicle, resulting in low accuracy of intersection identification. Therefore, accurate definition and identification of intersection patterns are necessary prerequisites for the safe and smooth progress of autonomous driving. At the same time, for existing autonomous driving vehicles, the method of driving at intersections is generally to set several action points in a fixed area, and pre-store the trajectory lines that need to be followed by the relevant actions, and then specify the actions that may occur at each action point. This method obviously does not have the universality of application in a wider area. Summary of the invention

[0003] In view of this, the present invention provides a method and system for unmanned truck intersection recognition based on map data to solve technical problems such as the prior art being unable to be applied to scene migration and having high implementation costs.

[0004] The present invention provides a method for identifying an unmanned truck intersection based on map data, the method comprising: step 1, adding the longitude and latitude of four vertices of an electronic fence to an unmanned driving platform to construct an intersection electronic fence; step 2, according to the longitude and latitude of the four vertices of the electronic fence, one-to-one correspondence between the intersection electronic fence and an intersection area in map data of an unmanned truck operation area; step 3, the unmanned truck sends a vehicle message to the unmanned driving platform in real time, the vehicle message comprising real-time longitude and latitude information of the vehicle, a lane line set of a vehicle path planning, and information on a lane line where the vehicle is currently located; step 4, selecting a lane line associated with the intersection area in the lane line set of the vehicle path planning; step 5, judging whether the lane line where the vehicle is currently located is a lane line associated with the intersection area, and if so, judging whether the distance between the unmanned truck and the intersection electronic fence corresponding to the intersection area is less than a threshold value according to the real-time longitude and latitude information of the vehicle; and step 6, if less than the threshold value, determining that the currently traveling vehicle has arrived at the intersection.

[0005] Furthermore, the method also includes: step 7, generating a vehicle rectangle based on the real-time latitude and longitude information of the vehicle; step 8, judging whether the vehicle rectangle intersects with the intersection electronic fence based on the latitude and longitude of the four corners of the electronic fence; step 9, if intersecting, determining that the current driving vehicle has entered the intersection.

[0006] Furthermore, the method also includes: step 10, if the vehicle rectangle does not intersect with the intersection electronic fence, then determine whether the distance between the vehicle and the entrance edge of the intersection area is less than the distance from the exit edge of the intersection area. If so, determine that the current driving vehicle has left the intersection.

[0007] Furthermore, the step 6 also includes: if it is not less than the threshold, go to step 4.

[0008] Furthermore, the method for determining the distance between the unmanned truck and the electronic fence at the intersection corresponding to the intersection area includes: step 51, selecting the two vertices closest to the unmanned truck for connection according to the longitude and latitude of the four vertices of the electronic fence to form a line segment AB; step 52, calculating the distance from the current longitude and latitude coordinate point X of the unmanned truck to the line segment AB according to the real-time longitude and latitude information of the vehicle.

[0009] Furthermore, the step 7 includes: step 71, initializing the real-time latitude and longitude information of the vehicle; step 72, performing coordinate conversion on the initialized longitude and latitude of the vehicle; step 73, calculating corresponding rectangle vertices according to the converted longitude and latitude of the vehicle and the heading angle of the vehicle to form a vehicle rectangle.

[0010] Furthermore, the step 8 includes: step 81, forming an electronic fence rectangle according to the longitude and latitude of the four vertices of the electronic fence; step 82, judging whether any side of the electronic fence rectangle intersects with any side of the vehicle rectangle; step 83, if so, judging whether the vehicle rectangle intersects with the intersection electronic fence; if not, judging whether the vehicle rectangle intersects with the intersection electronic fence.

[0011] The present invention also provides an unmanned truck intersection recognition system based on map data, the system comprising: an unmanned driving platform, used for adding the longitude and latitude of four vertices of an electronic fence to construct an intersection electronic fence, according to the longitude and latitude of the four vertices of the electronic fence, one-to-one correspondence between the intersection electronic fence and the intersection area in the map data of the unmanned truck operation area, selecting a lane line associated with the intersection area in a vehicle path planning lane line set, judging whether the lane line currently located by the vehicle is a lane line associated with the intersection area, if so, judging whether the distance between the unmanned truck and the intersection electronic fence corresponding to the intersection area is less than a threshold value according to the real-time longitude and latitude information of the vehicle, if less than the threshold value, determining that the currently traveling vehicle has arrived at the intersection; an unmanned truck, connected to the unmanned driving platform via a network, used for sending vehicle messages to the unmanned driving platform in real time, the vehicle message comprising the real-time longitude and latitude information of the vehicle, the vehicle path planning lane line set and the lane line information of the vehicle currently located.

[0012] Furthermore, the unmanned driving platform is also used to generate a vehicle rectangle based on the real-time latitude and longitude information of the vehicle, and determine whether the vehicle rectangle intersects with the electronic fence of the intersection based on the latitude and longitude of the four corners of the electronic fence. If so, it is determined that the currently traveling vehicle has entered the intersection. If the vehicle rectangle does not intersect with the electronic fence of the intersection, it is determined whether the distance between the vehicle and the entrance edge of the intersection area is less than the distance from the exit edge of the intersection area. If so, it is determined that the currently traveling vehicle has left the intersection.

[0013] Furthermore, the unmanned driving platform is also used to reselect the lane line associated with the intersection area in the vehicle path planning lane line set if it is not less than the threshold, and repeat the subsequent steps.

[0014] The present invention provides a method and system for identifying unmanned truck intersections based on map data. The technical solution corresponds the intersection area shown in the map to an electronic fence, and according to the vehicle driving lane line planning and the lane where the vehicle is currently located, part of the lane line is intercepted as a prerequisite for judging whether the vehicle is about to enter, has entered, or has left. If the intercepted lane line is related to the intersection area of ​​the map, the relationship between the lane line and the electronic fence corresponding to the intersection area of ​​the map is determined, so as to determine whether the vehicle is about to enter the intersection, has entered the intersection, or has left the intersection. This technical solution is suitable for the application scenarios of unmanned trucks, does not need to rely on intersection recognition devices, and can effectively reduce costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a flow chart of a method for identifying an unmanned truck intersection based on map data provided by the present invention;

[0016] Figure 2 It is a schematic flow chart of a method for generating a vehicle rectangle provided by the present invention;

[0017] Figure 3 It is an explanatory diagram of the judgment structure of the unmanned container truck arriving at the intersection provided by the present invention;

[0018] Figure 4 It is an explanatory diagram of the judgment structure of the unmanned container truck having entered the intersection provided by the present invention;

[0019] Figure 5 It is an explanatory diagram of the judgment structure of the unmanned container truck leaving the intersection provided by the present invention. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] Method Item Embodiment

[0022] The present invention provides a method for identifying unmanned truck intersections based on map data, such as Figure 1 As shown, the method steps are as follows.

[0023] Step 1: Add the longitude and latitude of the four vertices of the electronic fence on the unmanned driving platform to build the intersection electronic fence;

[0024] Step 2: According to the longitude and latitude of the four vertices of the electronic fence, the electronic fence at the intersection is matched one by one with the intersection area in the map data of the unmanned truck operation area;

[0025] Step 3: The unmanned container truck sends a vehicle message to the unmanned driving platform. The vehicle message includes the real-time latitude and longitude information of the vehicle, the set of lane lines planned for the vehicle path, and the lane line information where the vehicle is currently located;

[0026] Step 4, determining whether each lane line in the vehicle path planning lane line set is associated with the intersection area;

[0027] Step 5: If the lane line that the vehicle is currently in is a lane line associated with the intersection area, determine whether the distance between the unmanned truck and the intersection electronic fence corresponding to the intersection area is less than a threshold value based on the real-time latitude and longitude information of the vehicle;

[0028] The method for determining the distance between an unmanned container truck and an intersection electronic fence corresponding to the intersection area comprises the following steps.

[0029] Step 51, according to the longitude and latitude of the four vertices of the electronic fence, select the two vertices closest to the unmanned truck and connect them to form a line segment AB;

[0030] Step 52, based on the real-time latitude and longitude information of the vehicle, calculate the distance from the current longitude and latitude coordinate point X of the unmanned container truck to the line segment AB.

[0031] like Figure 3 As shown in the figure, the rectangular box is the intersection electronic fence, where A and B are the two vertices closest to the unmanned truck, and X is the current coordinate point of the unmanned truck. The determination of whether the distance between the unmanned truck and the intersection electronic fence corresponding to the intersection area is less than the threshold is to determine whether the distance from point X to line segment AB is less than the threshold. Of course, the threshold here can be adjusted according to the actual situation. Let the projection point of X to line segment AB be C, then is the inner product of two vectors, and Where θ is the angle between vector AP and AB. is the length of the vector. Then This is the length of line segment AC, which has no directionality. This value is consistent with the above-mentioned direction representation. The whole constitutes a new vector with size and direction That is exist The projection vector in the direction of From the directionality of the vector we can know that:

[0032] Then d is the distance from point X to line segment AB, that is, the distance from the unmanned truck to the electronic fence at the intersection.

[0033] Step 6: If it is less than the threshold, it is determined that the current vehicle has arrived at the intersection; if it is not less than the threshold, go to step 4.

[0034] It can be seen from the above steps that the unmanned truck will send vehicle messages to the unmanned driving platform in real time. When the distance between the unmanned truck and the intersection electronic fence corresponding to the intersection area is not less than the threshold, the unmanned driving platform will re-judge whether each lane line in the vehicle path planning lane line set is associated with the intersection area based on the latest received vehicle message, and then repeat the subsequent steps.

[0035] Step 7, generating a vehicle rectangle according to the real-time latitude and longitude information of the vehicle;

[0036] like Figure 2 As shown, the specific process of step 7 is as follows.

[0037] Step 71, initializing the real-time latitude and longitude information of the vehicle;

[0038] Step 72, coordinate conversion of the initialized vehicle longitude and latitude;

[0039] Step 73, calculating corresponding rectangle vertices according to the converted vehicle longitude and latitude and vehicle heading angle to form a vehicle rectangle.

[0040] Step 8, judging whether the vehicle rectangle intersects with the intersection electronic fence according to the latitude and longitude of the four corners of the electronic fence;

[0041] like Figure 4 As shown, the rectangular frame in the figure is an electronic fence at the intersection, and the unmanned container truck has entered the electronic fence. The specific process of step 8 is as follows.

[0042] Step 81, forming an electronic fence rectangle according to the longitude and latitude of the four vertices of the electronic fence;

[0043] Step 82, determining whether any side of the electronic fence rectangle intersects with any side of the vehicle rectangle;

[0044] Step 83, if yes, then determine whether the vehicle rectangle intersects with the intersection electronic fence; if no, then determine whether the vehicle rectangle intersects with the intersection electronic fence.

[0045] Step 9: If they intersect, it is determined that the current vehicle has entered the intersection.

[0046] Step 10: If the vehicle rectangle does not intersect the intersection electronic fence, determine whether the distance between the vehicle and the entrance edge of the intersection area is less than the distance from the exit edge of the intersection area. If so, determine that the current driving vehicle has left the intersection.

[0047] like Figure 4 and Figure 5 As shown in Figure 1, if the vehicle rectangle does not intersect the electronic fence at the intersection, there may be two situations: one is that the unmanned truck has entered the electronic fence, such as Figure 4 As shown; the second is that the unmanned container truck has left, such as Figure 5 Therefore, the next step is to determine whether the distance between the vehicle and the entrance edge of the intersection area is less than the distance from the exit edge of the intersection area, so as to further confirm whether the vehicle has left the intersection.

[0048] The present invention provides a method and system for identifying unmanned truck intersections based on map data. The technical solution corresponds the intersection area shown in the map to an electronic fence, and according to the vehicle driving lane line planning and the lane where the vehicle is currently located, part of the lane line is intercepted as a prerequisite for judging whether the vehicle is about to enter, has entered, or has left. If the intercepted lane line is related to the intersection area of ​​the map, the relationship between the lane line and the electronic fence corresponding to the intersection area of ​​the map is determined, so as to determine whether the vehicle is about to enter the intersection, has entered the intersection, or has left the intersection. This technical solution is suitable for the application scenarios of unmanned trucks, does not need to rely on intersection recognition devices, and can effectively reduce costs.

[0049] Device Item Embodiment

[0050] The present invention provides a system for identifying unmanned container truck intersections based on map data. The system comprises an unmanned driving platform and an unmanned container truck, as well as a processor and a memory.

[0051] The unmanned driving platform is used to add the longitude and latitude of the four vertices of the electronic fence to construct the intersection electronic fence. According to the longitude and latitude of the four vertices of the electronic fence, the intersection electronic fence is matched one by one with the intersection area in the map data of the unmanned truck operation area, and it is judged whether each lane line in the vehicle path planning lane line set is associated with the intersection area. If the lane line where the vehicle is currently located is a lane line associated with the intersection area, then according to the real-time longitude and latitude information of the vehicle, it is judged whether the distance between the unmanned truck and the intersection electronic fence corresponding to the intersection area is less than the threshold value. If it is less than the threshold value, the current When a moving vehicle arrives at an intersection, if it is not less than a threshold, the lane line associated with the intersection area in the vehicle path planning lane line set is reselected and the subsequent steps are repeated; a vehicle rectangle is generated based on the real-time latitude and longitude information of the vehicle, and the latitude and longitude of the four corners of the electronic fence are used to determine whether the vehicle rectangle intersects with the intersection electronic fence. If so, it is determined that the current moving vehicle has entered the intersection. If the vehicle rectangle does not intersect with the intersection electronic fence, it is determined whether the distance between the vehicle and the entrance edge of the intersection area is less than the distance from the exit edge of the intersection area. If so, it is determined that the current moving vehicle has left the intersection.

[0052] The unmanned container truck is connected to the unmanned driving platform through the network and is used to send vehicle messages to the unmanned driving platform. The vehicle message includes the real-time latitude and longitude information of the vehicle, the lane line set of the vehicle path planning and the lane line information of the vehicle's current location.

[0053] The present invention provides a method and system for identifying unmanned truck intersections based on map data. The technical solution corresponds the intersection area shown in the map to an electronic fence, and according to the vehicle driving lane line planning and the lane where the vehicle is currently located, part of the lane line is intercepted as a prerequisite for judging whether the vehicle is about to enter, has entered, or has left. If the intercepted lane line is related to the intersection area of ​​the map, the relationship between the lane line and the electronic fence corresponding to the intersection area of ​​the map is determined, so as to determine whether the vehicle is about to enter the intersection, has entered the intersection, or has left the intersection. This technical solution is suitable for the application scenarios of unmanned trucks, does not need to rely on intersection recognition devices, and can effectively reduce costs.

[0054] In summary, the embodiments of the present invention provide a method and system for identifying unmanned truck intersections based on map data. This technical solution is applied to unmanned trucks, and there is no need to add an intersection identification device. It only needs to be judged based on the vehicle message containing the longitude and latitude of the vehicle and the map data information. In addition, this technical solution can accurately identify the relationship between the vehicle and the intersection for the definition of about to enter, already entered, and already left, which is of great help for further coordinated processing of multiple vehicle intersections. The present invention has obvious advantages in cost control for unmanned driving scenarios, procurement, installation, and debugging of intersection identification devices. The relationship between vehicles and intersections can be identified directly through messages and program algorithms, and there is no need to purchase additional intersection identification devices, which is conducive to reducing costs and increasing efficiency.

[0055] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for identifying unmanned truck intersections based on map data, characterized in that: The method comprises: Step 1: Add the longitude and latitude of the four vertices of the electronic fence on the unmanned driving platform to build the intersection electronic fence; Step 2: According to the longitude and latitude of the four vertices of the electronic fence, the electronic fence at the intersection is matched one by one with the intersection area in the map data of the unmanned truck operation area; Step 3: The unmanned container truck sends the vehicle message to the unmanned driving platform in real time. The vehicle message includes the real-time latitude and longitude information of the vehicle, the set of lane lines planned for the vehicle path, and the lane line information where the vehicle is currently located; Step 4, selecting a lane line associated with the intersection area from the vehicle path planning lane line set; Step 5, determine whether the lane line the vehicle is currently in is a lane line associated with the intersection area. If so, determine whether the distance between the unmanned truck and the intersection electronic fence corresponding to the intersection area is less than a threshold value based on the real-time latitude and longitude information of the vehicle; Step 6: If it is less than the threshold, it is determined that the current vehicle has arrived at the intersection.

2. According to the method of claim 1, the method is characterized by: The method further comprises: Step 7, generating a vehicle rectangle according to the real-time latitude and longitude information of the vehicle; Step 8, judging whether the vehicle rectangle intersects with the intersection electronic fence according to the latitude and longitude of the four corners of the electronic fence; Step 9: If they intersect, it is determined that the current vehicle has entered the intersection.

3. According to claim 2, a method for identifying unmanned truck intersections based on map data is characterized in that: The method further comprises: Step 10: If the vehicle rectangle does not intersect the intersection electronic fence, determine whether the distance between the vehicle and the entrance edge of the intersection area is less than the distance from the exit edge of the intersection area. If so, determine that the current driving vehicle has left the intersection.

4. The method for identifying unmanned truck intersections based on map data according to claim 1, characterized in that: The step 6 also includes: if it is not less than the threshold, go to step 4.

5. The method for identifying unmanned truck intersections based on map data according to claim 1, characterized in that: The method for determining the distance between the unmanned truck and the intersection electronic fence corresponding to the intersection area includes: Step 51, according to the longitude and latitude of the four vertices of the electronic fence, select the two vertices closest to the unmanned truck and connect them to form a line segment AB; Step 52, based on the real-time latitude and longitude information of the vehicle, calculate the distance from the current longitude and latitude coordinate point X of the unmanned container truck to the line segment AB.

6. The method for identifying unmanned truck intersections based on map data according to claim 2, characterized in that: The step 7 comprises: Step 71, initializing the real-time latitude and longitude information of the vehicle; Step 72, coordinate conversion of the initialized vehicle longitude and latitude; Step 73, calculating corresponding rectangle vertices according to the converted vehicle longitude and latitude and vehicle heading angle to form a vehicle rectangle.

7. The method for identifying unmanned truck intersections based on map data according to claim 2, characterized in that: The step 8 comprises: Step 81, forming an electronic fence rectangle according to the longitude and latitude of the four vertices of the electronic fence; Step 82, determining whether any side of the electronic fence rectangle intersects with any side of the vehicle rectangle; Step 83, if yes, then determine whether the vehicle rectangle intersects with the intersection electronic fence; if no, then determine whether the vehicle rectangle intersects with the intersection electronic fence.

8. A system for implementing the unmanned truck intersection identification method based on map data as described in claims 1-7, characterized in that: The system comprises: The unmanned driving platform is used to add the longitude and latitude of the four vertices of the electronic fence to construct the intersection electronic fence. According to the longitude and latitude of the four vertices of the electronic fence, the intersection electronic fence is matched one by one with the intersection area in the map data of the unmanned truck operation area, and the lane line associated with the intersection area in the vehicle path planning lane line set is selected to determine whether the lane line currently located by the vehicle is the lane line associated with the intersection area. If so, the distance between the unmanned truck and the intersection electronic fence corresponding to the intersection area is determined to be less than a threshold value according to the real-time longitude and latitude information of the vehicle. If it is less than the threshold value, it is determined that the current driving vehicle has arrived at the intersection; The unmanned container truck is connected to the unmanned driving platform through the network and is used to send vehicle messages to the unmanned driving platform in real time. The vehicle message includes the real-time latitude and longitude information of the vehicle, the lane line set of the vehicle path planning and the lane line information of the vehicle's current location.

9. The system for identifying unmanned truck intersections based on map data according to claim 7, characterized in that: The unmanned driving platform is also used to generate a vehicle rectangle based on the real-time latitude and longitude information of the vehicle, and determine whether the vehicle rectangle intersects with the electronic fence of the intersection based on the latitude and longitude of the four corners of the electronic fence. If so, it is determined that the current driving vehicle has entered the intersection. If the vehicle rectangle does not intersect with the electronic fence of the intersection, it is determined whether the distance between the vehicle and the entrance edge of the intersection area is less than the distance from the exit edge of the intersection area. If so, it is determined that the current driving vehicle has left the intersection.

10. The system for identifying unmanned truck intersections based on map data according to claim 7, characterized in that: The unmanned driving platform is also used to reselect the lane line associated with the intersection area in the vehicle path planning lane line set if it is not less than the threshold, and repeat the subsequent steps.

Images