Lane positioning method and device, electronic equipment and vehicle

By acquiring and fusing the vehicle perceived semantic lane lines, calculating lane index and observation probability, and combining lane line distance judgment, the problem of inaccurate existing lane positioning algorithms in the occlusion environment is solved, and the positioning accuracy and safety of the autonomous driving system are improved sex.

CN119085673APending Publication Date: 2024-12-06HOZON NEW ENERGY AUTOMOBILE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411070776.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-06
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The existing lane positioning algorithms are inaccurate in the occlusion environment, and the lane where the vehicle is located is unreasonable only by observing the probability, which affects the robustness and safety of the autonomous driving system.

Method used

By obtaining the perceived semantic lane lines of the target vehicle, the local map is obtained through the fusion processing, the observation probability of lane index and lane change candidate lane change, the lane change trend is judged based on the lane line distance at the front and rear moments, and the weighted average calculation comprehensive probability is used to determine the lane where the vehicle is located. .

Benefits of technology

It improves the accuracy and safety of autonomous driving positioning, especially in shading environments or perceived missed detection scenarios, which can accurately identify the lane where the vehicle is located.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119085673A_ABST
    Figure CN119085673A_ABST
Patent Text Reader

Abstract

The invention provides a lane positioning method and device, electronic equipment and a vehicle, and the method comprises the steps: determining a lane index of a lane where a target vehicle is located based on the number of lanes of a road where the target vehicle is located and a local map; determining an observation probability of a lane change candidate lane of the target vehicle according to the lane index and the local map, and calculating a prediction probability of the target vehicle changing to each lane of the road; and according to a weighted average value of the prediction probability and the observation probability, determining and obtaining a comprehensive probability of changing the target vehicle to each lane, and determining the lane with the maximum comprehensive probability as the lane where the target vehicle is located. The accuracy of automatic driving positioning can be improved, so that the driving safety is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle positioning, and in particular to a lane positioning method, device, electronic equipment and vehicle. Background Art

[0002] Lane positioning technology is a technology that determines the lane in which the vehicle is located by sensing the vehicle's surrounding environment, especially road signs and road boundaries. Lane positioning is of great significance in autonomous driving without map technology. First of all, lane positioning technology does not require high-precision maps, nor does it need to achieve centimeter-level positioning, thus greatly reducing the cost of mass-producing autonomous driving without map technology. At the same time, lane positioning determines the lane index of the vehicle in the SD map, which can help the vehicle make accurate planning, control and decisions, such as lane changes or obstacle avoidance in complex road conditions. In addition, it is also crucial for environmental perception and map updates. By continuously and accurately updating the vehicle's position information, the robustness and safety of the autonomous driving system can be improved. However, the existing lane positioning algorithm has inaccurate estimates of each lane based on the observation probability in an occluded environment, and it is also unreasonable to determine the lane in which the vehicle is located only by the observation probability. Summary of the invention

[0003] The purpose of this application is to provide a lane positioning method, device, electronic device and computer storage medium, which can improve the accuracy of autonomous driving positioning and thus improve the safety of autonomous driving.

[0004] To achieve the above objectives:

[0005] In a first aspect, an embodiment of the present invention provides a lane positioning method, the method comprising:

[0006] Determine a lane index of a lane where the target vehicle is located based on the number of lanes of the road where the target vehicle is located and the local map;

[0007] Determining an observed probability of a lane change candidate lane for the target vehicle according to the lane index and the local map, and calculating a predicted probability of the target vehicle changing lanes to each lane of the road;

[0008] The comprehensive probability of the target vehicle changing lanes to each lane is determined based on the weighted average of the predicted probability and the observed probability, and the lane with the largest comprehensive probability is determined as the lane where the target vehicle is located.

[0009] In one embodiment, before determining the lane index of the lane where the target vehicle is located based on the number of lanes of the road where the target vehicle is located and the local map, the method further includes:

[0010] Acquire the perceived semantic lane lines of the road where the target vehicle is located, fuse multiple frames of the perceived semantic lane lines to obtain a fused image, and obtain a local map of the road where the target vehicle is located through the fused image.

[0011] In one embodiment, determining the lane index of the lane where the target vehicle is located based on the number of lanes of the road where the target vehicle is located and the local map includes:

[0012] The lanes are numbered according to a preset direction to obtain lane numbers, so that the lane index is represented by the lane numbers, and the number of curbs on the left and right sides of the road is obtained;

[0013] The lane index is calculated according to the following formula:

[0014]

[0015] Among them, lane_index represents the lane index where the target vehicle is located, l_laneline represents the number of lane lines on the left side of the target vehicle in the road where the target vehicle is located, n is the number of lanes on the road where the target vehicle is located, r_laneline represents the number of lane lines on the right side of the target vehicle in the road where the target vehicle is located, l_roadedge is the number of curbs on the left side of the target vehicle in the road where the target vehicle is located, and r_roadedge is the number of curbs on the right side of the target vehicle in the road where the target vehicle is located.

[0016] In one implementation, determining the observation probability of the lane change candidate lane of the target vehicle includes:

[0017] The observation probability is calculated according to the following formula:

[0018] P measure =e -||d|| ;

[0019] Among them, p measure is the observation probability of the lane change candidate lane of the target vehicle, e is a natural number e, and d is the difference between the lane index of the lane change candidate lane and the lane where the target vehicle is located.

[0020] In one embodiment, before calculating the predicted probability of the target vehicle changing lanes to each lane of the road, the method further includes:

[0021] Detecting a first lateral distance between the target vehicle and the lane line closest to the left side of the target vehicle at the current moment and a second lateral distance between the target vehicle and the lane line closest to the left side of the target vehicle at the previous moment;

[0022] It is determined whether the target vehicle meets a preset lane changing condition according to the first lateral distance and the second lateral distance.

[0023] In one implementation, the preset lane changing condition includes:

[0024]

[0025] in, is the first lateral distance, is the second lateral distance, s1 represents the difference threshold between the first lateral distance and the second lateral distance, s2 represents the maximum value of the first lateral distance, s3 represents the maximum value of the second lateral distance, s1 and s2 are less than the lane width, s2 is greater than s1, and s3 is greater than 0.

[0026] In one embodiment, the determining whether the target vehicle satisfies a preset lane changing condition further includes:

[0027] Acquire a first relative position of the target vehicle in the local map at a previous moment and a second relative position of the target vehicle in the local map at a current moment;

[0028] The lane-changing tendency of the target vehicle is determined according to the positional relationship between the first relative position and the second relative position to determine whether the target vehicle is going to change lanes.

[0029] In a second aspect, an embodiment of the present invention provides a lane positioning device, including:

[0030] A determination module, configured to determine a lane index of a lane where the target vehicle is located based on the number of lanes of the road where the target vehicle is located and a local map;

[0031] a calculation module, configured to determine an observed probability of a lane change candidate lane of the target vehicle according to the lane index and the local map, and calculate a predicted probability of the target vehicle changing lanes to each lane of the road;

[0032] The positioning module is used to determine the comprehensive probability of the target vehicle changing lanes to each lane based on the weighted average of the predicted probability and the observed probability, and determine the lane with the largest comprehensive probability as the lane where the target vehicle is located.

[0033] In one embodiment, the device further comprises:

[0034] A fusion module is used to obtain the perceived semantic lane lines of the road where the target vehicle is located, fuse multiple frames of the perceived semantic lane lines to obtain a fused image, and obtain a local map of the road where the target vehicle is located through the fused image.

[0035] In one implementation, the computing module is specifically configured to:

[0036] The lanes are numbered according to a preset direction to obtain lane numbers, so that the lane index is represented by the lane numbers, and the number of curbs on the left and right sides of the road is obtained;

[0037] The lane index is calculated according to the following formula:

[0038]

[0039] Among them, lane_index represents the lane index where the target vehicle is located, l_laneline represents the number of lane lines on the left side of the target vehicle in the road where the target vehicle is located, n is the number of lanes on the road where the target vehicle is located, r_laneline represents the number of lane lines on the right side of the target vehicle in the road where the target vehicle is located, l_roadedge is the number of curbs on the left side of the target vehicle in the road where the target vehicle is located, and r_roadedge is the number of curbs on the right side of the target vehicle in the road where the target vehicle is located.

[0040] In one implementation, the computing module is specifically configured to:

[0041] The observation probability is calculated according to the following formula:

[0042] p measure =e -||d|| ;

[0043] Among them, p measure is the observation probability of the lane change candidate lane of the target vehicle, e is a natural number e, and d is the difference between the lane index of the lane change candidate lane and the lane where the target vehicle is located.

[0044] In one embodiment, the device further includes a detection module for detecting a first lateral distance between the target vehicle and the lane line closest to the left side of the target vehicle at the current moment and a second lateral distance between the target vehicle and the lane line closest to the left side of the target vehicle at the previous moment;

[0045] It is determined whether the target vehicle meets a preset lane changing condition according to the first lateral distance and the second lateral distance.

[0046] In one implementation, the preset lane changing condition includes:

[0047]

[0048] in, is the first lateral distance, is the second lateral distance, s1 represents the difference threshold between the first lateral distance and the second lateral distance, s2 represents the maximum value of the first lateral distance, s3 represents the maximum value of the second lateral distance, s1 and s2 are less than the lane width, s2 is greater than s1, and s3 is greater than 0.

[0049] In one embodiment, the detection module is further used to obtain a first relative position of the target vehicle in the local map at a previous moment and a second relative position of the target vehicle in the local map at a current moment;

[0050] The lane-changing tendency of the target vehicle is determined according to the positional relationship between the first relative position and the second relative position to determine whether the target vehicle is going to change lanes.

[0051] In a third aspect, an embodiment of the present invention provides an electronic device, including:

[0052] processor;

[0053] a memory for storing instructions executable by the processor;

[0054] Wherein, the processor is configured to execute the instructions to implement the lane positioning method as described in any one of the first aspects.

[0055] In a fourth aspect, an embodiment of the present invention provides a vehicle, comprising the electronic device as described in the third aspect.

[0056] The lane positioning method, device, electronic device and vehicle provided by the embodiment of the present invention obtain the perceived semantic lane line through the SD map, thereby obtaining the local map of the road where the target vehicle is located, combining the local map with the relative position relationship of the vehicle to obtain the lane index of the lane where the vehicle is located, and calculating the observation probability of the lane to be changed by the target vehicle through the lane index, and calculating the predicted probability of the target vehicle changing lanes to any lane through the lateral distance between the target vehicle and the lane line at two moments before and after and the total probability formula, and performing weighted averaging on the observation probability and the predicted probability, and the lane with the largest weighted average value is the lane where the target vehicle is located. In this way, the accuracy of automatic driving positioning can be improved, thereby improving driving safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0058] Figure 1A schematic diagram of a process flow of a lane positioning method provided by an embodiment of the present invention;

[0059] Figure 2 A schematic diagram of a lane index when the number of left curbs on the road where the target vehicle is located is 1 in the present invention;

[0060] Figure 3 A lane index diagram when the number of left curbs on the road where the target vehicle is located is 0 and the number of right curbs is 1 in the present invention;

[0061] Figure 4 A schematic diagram of lane indexing when the number of curbs on both the left and right sides of the road where the target vehicle is located is 0 in the present invention;

[0062] Figure 5 A schematic diagram of a lane-changing scenario in which a missed detection is detected and there is no curb in the present invention;

[0063] Figure 6 A schematic diagram of a module of a lane positioning device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0064] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0065] It should be noted that, in this article, the terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element. In addition, components, features, and elements with the same name in different embodiments of the present application may have the same meaning or different meanings, and their specific meanings need to be determined by their explanation in the specific embodiment or further combined with the context of the specific embodiment.

[0066] See also Figure 1, is a flow chart of a lane positioning method provided in an embodiment of the present application. The lane positioning method can be executed by a lane positioning device provided in an embodiment of the present application. The device can be implemented in software and / or hardware, such as a server, a computer, a processor and other electronic devices. In this embodiment, the execution subject of the lane positioning method is a processor of a vehicle terminal as an example. The lane positioning method provided in this embodiment includes:

[0067] Step S1, determining a lane index of a lane where the target vehicle is located based on the number of lanes of the road where the target vehicle is located and a local map.

[0068] Here, the number of lanes can be obtained through SD maps, GPS, etc., and the local map contains road landmark features such as lane lines and curbs. These features can be combined to infer the lane index of the vehicle's lane.

[0069] It should be noted that the lane index refers to the sequence number of the lane where the vehicle is located after the lanes are numbered in a certain order (for example, from left to right or from left to right), that is, the lane index.

[0070] In one embodiment, before determining the lane index of the lane where the target vehicle is located based on the number of lanes of the road where the target vehicle is located and the local map, the method further includes:

[0071] Acquire the perceived semantic lane lines of the road where the target vehicle is located, fuse multiple frames of the perceived semantic lane lines to obtain a fused image, and obtain a local map of the road where the target vehicle is located through the fused image.

[0072] As mentioned above, the number of lanes can be obtained through SD maps or GPS. It can be understood that SD maps are a form of geographic information data used to represent the simulation of geographic real scenes. SD maps usually exist in digital form and can provide rich geographic information, including roads, buildings, water systems, etc. Therefore, the perceived semantic lane lines of the road where the target vehicle is located can be obtained through the SD map, that is, a large number of points are used to form the lane line contour at a specific moment. Then, a local map can be obtained by fusing the perceived semantic lane lines of multiple frames. In this way, by fusing multiple frames of perception data to obtain a local map, it is possible to provide higher resolution and accuracy in the target area, update geographic information in real time and make the details of the local map richer, which can improve the accuracy of autonomous driving lane positioning.

[0073] Preferably, each time lane positioning is performed in the embodiment of the present invention, 10 frames of perceived semantic lane lines are fused to obtain a local map.

[0074] In one embodiment, determining the lane index of the lane where the target vehicle is located based on the number of lanes of the road where the target vehicle is located and the local map includes:

[0075] The lanes are numbered according to a preset direction to obtain lane numbers, so that the lane index is represented by the lane numbers, and the number of curbs on the left and right sides of the road is obtained;

[0076] The lane index is calculated according to the following formula:

[0077]

[0078] Among them, lane_index represents the lane index where the target vehicle is located, l_laneline represents the number of lane lines on the left side of the target vehicle in the road where the target vehicle is located, n is the number of lanes on the road where the target vehicle is located, r_laneline represents the number of lane lines on the right side of the target vehicle in the road where the target vehicle is located, l_roadedge is the number of curbs on the left side of the target vehicle in the road where the target vehicle is located, and r_roadedge is the number of curbs on the right side of the target vehicle in the road where the target vehicle is located.

[0079] like Figure 2 As shown in , the number of lane lines on the left is 2, and the number of curbs is 1. From the geometric relative relationship, we can know that the lane index of the vehicle is 3. When the number of curbs on the left is 0 and the number of curbs on the right is 1, as shown in Figure 3 As shown in , the number of lane lines on the right is 2. According to the geometric relative relationship, the lane index of the vehicle is 3. When the number of curbs on both sides is 0, such as Figure 4 As shown, the number of lane lines on the left is 2. From the geometric relative relationship, we can know that the lane index where the vehicle is located is 2.

[0080] Step S2, determining the observed probability of the lane change candidate lane of the target vehicle according to the lane index and the local map, and calculating the predicted probability of the target vehicle changing lanes to each lane of the road.

[0081] Here, the observation probability is used to help the system determine the possibility of the target vehicle's current lane-changing intention, which is achieved by analyzing the target vehicle's behavior, position, speed, and surrounding environment. Accurate observation probability can reveal whether the target vehicle really intends to change lanes, or the strength of its lane-changing intention. The purpose of calculating the observation probability is to accurately evaluate and predict the lane-changing behavior of the target vehicle, and to calculate the predicted probability of the target vehicle changing lanes to any lane through the full probability formula. By calculating the observation probability of the lane to be changed and the predicted probability of changing lanes to any lane, the accuracy of autonomous driving positioning can be improved.

[0082] In one embodiment, the determining of the observed probability of the lane change candidate lane of the target vehicle in step S2 includes:

[0083] The observation probability is calculated according to the following formula: measure =e -||d|| ;

[0084] Among them, p measure is the observation probability of the lane to be changed, e is a natural number e, and d is the difference between the lane index of the lane to be changed and the lane where the target vehicle is located.

[0085] It can be understood that according to the above formula, the greater the index difference between the lane to be changed and the lane where the target vehicle is located, that is, the farther the distance, the smaller the probability of observing the target vehicle changing lanes to the lane, which is an exponential decay relationship.

[0086] In one embodiment, before calculating the predicted probability of the target vehicle changing lanes to each lane of the road, the method further includes: detecting a first lateral distance between the target vehicle and the lane line closest to the left side of the target vehicle at a current moment and a second lateral distance between the target vehicle and the lane line closest to the left side of the target vehicle at a previous moment; and determining whether the target vehicle meets a preset lane changing condition based on the first lateral distance and the second lateral distance.

[0087] It can be understood that by comparing the lateral distances between the vehicle and the nearest lane line on the left at different times, the movement trend of the vehicle can be determined more accurately, so as to judge whether the vehicle intends to change lanes through the movement trend of the vehicle, and when the first lateral distance and the second lateral distance meet the vehicle lane change condition, it is judged that the vehicle is about to change lanes. In this way, the accuracy of the judgment can be improved.

[0088] Here, see Figure 2 , 3 As shown in , 4 , the lane line closest to the left side of the vehicle is the first lane line on the left side of the vehicle, and the same is true for the lane line closest to the right side of the vehicle.

[0089] In one implementation, the preset lane changing condition includes:

[0090]

[0091] in, is the first horizontal distance, is the second lateral distance, s1 represents the difference threshold between the first lateral distance and the second lateral distance, s2 represents the maximum value of the first lateral distance, s3 represents the maximum value of the second lateral distance, s1 and s2 are less than the lane width, s2 is greater than s1, and s3 is greater than 0.

[0092] It is understandable that in actual applications, the vehicle lane changing time is about 4-6 seconds. In order to make lane positioning have practical significance, in the embodiment of the present invention, the lane positioning frequency is set to 10 seconds each time.

[0093] It can be understood that when the condition for the vehicle to change lanes to the left is met, the probability of not changing lanes for vehicles in the leftmost lane is 1, and the probability of changing lanes to the left and the probability of changing lanes to the right are 0, and the probability of changing lanes to the left for vehicles in other lanes is 1, and the probability of changing lanes to the right and the probability of not changing lanes are 0; when the condition for the vehicle to change lanes to the right is met, the probability of not changing lanes for vehicles in the rightmost lane is 1, and the probability of changing lanes to the left and the probability of changing lanes to the right are 0, and the probability of changing lanes to the right for vehicles in other lanes is 1, and the probability of not changing lanes is 0; when the condition for the vehicle not to change lanes is met, the probability of not changing lanes in all lanes is 1, and the probability of changing lanes to the left and the probability of changing lanes to the right are 0.

[0094] After determining the vehicle's intention to change lanes according to the preset lane-changing conditions, the total probability formula is used recursively to obtain the predicted probability of each lane at the current moment from the comprehensive probability of each lane at the previous moment.

[0095] In one embodiment, determining whether the vehicle satisfies a preset lane changing condition further includes: obtaining a first relative position of the target vehicle in the local map at a previous moment and a second relative position of the target vehicle in the local map at a current moment; and determining the lane changing trend of the vehicle based on the positional relationship between the first relative position and the second relative position to determine whether the vehicle is to change lanes.

[0096] As mentioned above, by comparing the positions at different times, the movement trend of the vehicle can be determined more accurately. When a certain relationship is satisfied between the first relative position and the second relative position, it is determined that the vehicle is about to change lanes.

[0097] Step S3, determining the comprehensive probability of the target vehicle changing lanes to each lane based on the weighted average of the predicted probability and the observed probability, and determining the lane with the largest comprehensive probability as the lane where the target vehicle is located.

[0098] Here, by assigning certain weights to the observation probability and the prediction probability and calculating the weighted average, the information of both the prediction probability and the observation is integrated. In this way, the obtained comprehensive probability is more comprehensive and accurate. Finally, the lane with the largest comprehensive probability is identified as the lane where the target vehicle is located. This method can improve the accuracy and reliability of vehicle positioning in the scenario of missed detection, with high observation accuracy and improved robustness.

[0099] For example, assuming there is a 4-lane scenario, the comprehensive probabilities of the vehicle in each lane at T0 are 0.1, 0.75, 0.1, and 0.05 respectively. The vehicle changes lanes to the right at T0. According to the full probability formula, as shown in Table 1, the predicted probability of the vehicle at T1 can be obtained as shown in the following table.

[0100] Table 1 Recursive results of the total probability formula

[0101]

[0102] The predicted probability of each lane calculated in Table 1 and the observed probability in the previous text are weighted averaged to obtain the comprehensive probability of each lane at the current moment. The lane with the largest comprehensive probability is the lane where the vehicle is located.

[0103] The solution proposed in the embodiment of the present invention can still obtain the correct lane when the vehicle blocks the environment or the weather conditions are bad and the perception is missed and the curb is not sensed. Figure 5 As shown, the vehicle is in a road scene without a curb, and there are different degrees of missed detections at T1 and T2, which leads to errors in the lane index obtained from the geometric relationship between the local map and the vehicle.

[0104] However, by taking the weighted average of the predicted probability and observed probability of each lane obtained by recursion through the total probability formula, the comprehensive probability of each lane is obtained, and the lane with the largest comprehensive probability is the lane where the target vehicle is located. Assuming that the predicted probability is 2 / 3 and the observed probability is 1 / 3 in this example, as shown in Table 2, the lane where the vehicle is located can be correctly located.

[0105] Table 2 Comprehensive probability results of each lane

[0106]

[0107]

[0108] In Table 1, 2 / 3 is the predicted probability, 1 / 3 is the observed probability, the decimal multiplied by the predicted probability and the observed probability is the weight value, and the result after the equation is the comprehensive probability. In the calculation results shown in Table 2, at time T0, the lane with the largest comprehensive probability is lane 3, so at time T0, the target vehicle is in the lane with lane index 3. Similarly, at time T1, the target vehicle is in the lane with lane index 3; at time T2, the target vehicle is in the lane with lane index 4; at time T3, the target vehicle is in the lane with lane index 3.

[0109] It should be noted that the weight value will be adjusted according to the situation of the local map. When the local map shows that there are curbs on both sides of the vehicle, the weight of the predicted probability will be lowered and the weight of the observed probability will be increased. The opposite is true when there is no curb.

[0110] The lane positioning method provided by the embodiment of the present invention obtains the perceived semantic lane line through the SD map, thereby obtaining the local map of the road where the target vehicle is located, and obtains the lane index of the lane where the vehicle is located by combining the local map and the relative position relationship of the vehicle, and calculates the observation probability of the lane to be changed by the target vehicle through the lane index, and calculates the predicted probability of the target vehicle changing lanes to any lane through the lateral distance between the target vehicle and the lane line at two moments before and after and the total probability formula, and performs weighted averaging on the observation probability and the predicted probability, and the lane with the largest weighted average value is the lane where the target vehicle is located. In this way, the accuracy of automatic driving positioning can be improved, thereby improving driving safety.

[0111] Based on the same concept of the above embodiments, the embodiment of the present invention provides a lane positioning device, such as Figure 6 As shown, the lane positioning device includes:

[0112] A determination module, configured to determine a lane index of a lane where the target vehicle is located based on the number of lanes of the road where the target vehicle is located and a local map;

[0113] a calculation module, configured to determine an observed probability of a lane change candidate lane of the target vehicle according to the lane index and the local map, and calculate a predicted probability of the target vehicle changing lanes to each lane of the road;

[0114] The positioning module is used to determine the comprehensive probability of the target vehicle changing lanes to each lane based on the weighted average of the predicted probability and the observed probability, and determine the lane with the largest comprehensive probability as the lane where the target vehicle is located.

[0115] In one embodiment, the device further comprises:

[0116] A fusion module is used to obtain the perceived semantic lane lines of the road where the target vehicle is located, fuse multiple frames of the perceived semantic lane lines to obtain a fused image, and obtain a local map of the road where the target vehicle is located through the fused image.

[0117] In one implementation, the computing module is specifically configured to:

[0118] The lanes are numbered according to a preset direction to obtain lane numbers, so that the lane index is represented by the lane numbers, and the number of curbs on the left and right sides of the road is obtained;

[0119] The lane index is calculated according to the following formula:

[0120]

[0121] Among them, lane_index represents the lane index where the target vehicle is located, l_laneline represents the number of lane lines on the left side of the target vehicle in the road where the target vehicle is located, n is the number of lanes on the road where the target vehicle is located, r_laneline represents the number of lane lines on the right side of the target vehicle in the road where the target vehicle is located, l_roadedge is the number of curbs on the left side of the target vehicle in the road where the target vehicle is located, and r_roadedge is the number of curbs on the right side of the target vehicle in the road where the target vehicle is located.

[0122] In one implementation, the computing module is specifically configured to:

[0123] The observation probability is calculated according to the following formula:

[0124] P measure =e -||d|| ;

[0125] Among them, p measure is the observation probability of the lane change candidate lane of the target vehicle, e is a natural number e, and d is the difference between the lane index of the lane change candidate lane and the lane where the target vehicle is located.

[0126] In one embodiment, the device further includes a detection module for detecting a first lateral distance between the target vehicle and the lane line closest to the left side of the target vehicle at the current moment and a second lateral distance between the target vehicle and the lane line closest to the left side of the target vehicle at the previous moment;

[0127] It is determined whether the target vehicle meets a preset lane changing condition according to the first lateral distance and the second lateral distance.

[0128] In one implementation, the preset lane changing condition includes:

[0129]

[0130] in, is the first lateral distance, is the second lateral distance, s1 represents the difference threshold between the first lateral distance and the second lateral distance, s2 represents the maximum value of the first lateral distance, s3 represents the maximum value of the second lateral distance, s1 and s2 are less than the lane width, s2 is greater than s1, and s3 is greater than 0.

[0131] In one embodiment, the detection module is further used to obtain a first relative position of the target vehicle in the local map at a previous moment and a second relative position of the target vehicle in the local map at a current moment;

[0132] The lane-changing tendency of the target vehicle is determined according to the positional relationship between the first relative position and the second relative position to determine whether the target vehicle is going to change lanes.

[0133] For the specific definition of the lane positioning device, please refer to the definition of the lane positioning method above, which will not be repeated here. Each module in the above lane positioning device can be implemented in whole or in part by software, hardware and a combination thereof. The above modules can be embedded in or independent of the processor of the electronic device in the form of hardware, or can be stored in the memory of the electronic device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0134] Based on the same inventive concept as the aforementioned embodiments, an embodiment of the present invention further provides a computer-readable storage medium. When the instructions in the computer-readable storage medium are executed by a processor, a lane positioning method as described in any one of the aforementioned embodiments is implemented.

[0135] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0136] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the attached claims.

Claims

1. A lane positioning method, characterized in that: include: Determine a lane index of a lane where the target vehicle is located based on the number of lanes of the road where the target vehicle is located and the local map; Determining an observed probability of a lane change candidate lane for the target vehicle according to the lane index and the local map, and calculating a predicted probability of the target vehicle changing lanes to each lane of the road; The comprehensive probability of the target vehicle changing lanes to each lane is determined based on the weighted average of the predicted probability and the observed probability, and the lane with the largest comprehensive probability is determined as the lane where the target vehicle is located.

2. The method according to claim 1, characterized in that Before determining the lane index of the lane where the target vehicle is located based on the number of lanes of the road where the target vehicle is located and the local map, the method further includes: Acquire the perceived semantic lane lines of the road where the target vehicle is located, fuse multiple frames of the perceived semantic lane lines to obtain a fused image, and obtain a local map of the road where the target vehicle is located through the fused image.

3. The method according to claim 1 or 2, characterized in that: The determining, based on the number of lanes of the road where the target vehicle is located and the local map, a lane index of the lane where the target vehicle is located, comprises: The lanes are numbered according to a preset direction to obtain lane numbers, so that the lane index is represented by the lane numbers, and the number of curbs on the left and right sides of the road is obtained; The lane index is calculated according to the following formula: Among them, lane_index represents the lane index where the target vehicle is located, l_laneline represents the number of lane lines on the left side of the target vehicle in the road where the target vehicle is located, n is the number of lanes on the road where the target vehicle is located, r_laneline represents the number of lane lines on the right side of the target vehicle in the road where the target vehicle is located, l_roadedge is the number of curbs on the left side of the target vehicle in the road where the target vehicle is located, and r_roadedge is the number of curbs on the right side of the target vehicle in the road where the target vehicle is located.

4. The method according to claim 3, characterized in that The determining the observation probability of the lane change candidate lane of the target vehicle includes: The observation probability is calculated according to the following formula: p measure =and -‖d‖ ; Among them, p measure is the observation probability of the lane change candidate lane of the target vehicle, e is a natural number e, and d is the difference between the lane index of the lane change candidate lane and the lane where the target vehicle is located.

5. The method according to claim 1, characterized in that Before calculating the predicted probability of the target vehicle changing lanes to each lane of the road, the method further includes: Detecting a first lateral distance between the target vehicle and the lane line closest to the left side of the target vehicle at the current moment and a second lateral distance between the target vehicle and the lane line closest to the left side of the target vehicle at the previous moment; It is determined whether the target vehicle meets a preset lane changing condition according to the first lateral distance and the second lateral distance.

6. The method according to claim 5, characterized in that The preset lane changing conditions include: Left lane change conditions Right lane change conditions in, is the first lateral distance, is the second lateral distance, s1 represents the difference threshold between the first lateral distance and the second lateral distance, s2 represents the maximum value of the first lateral distance, s3 represents the maximum value of the second lateral distance, s1 and s2 are less than the lane width, s2 is greater than s1, and s3 is greater than 0.

7. The method according to claim 5, characterized in that The determining whether the target vehicle meets the preset lane changing condition also includes: Acquire a first relative position of the target vehicle in the local map at a previous moment and a second relative position of the target vehicle in the local map at a current moment; The lane-changing tendency of the target vehicle is determined according to the positional relationship between the first relative position and the second relative position to determine whether the target vehicle is going to change lanes.

8. A lane positioning device, characterized in that: include: A determination module, configured to determine a lane index of a lane where the target vehicle is located based on the number of lanes of the road where the target vehicle is located and a local map; a calculation module, configured to determine an observed probability of a lane change candidate lane of the target vehicle according to the lane index and the local map, and calculate a predicted probability of the target vehicle changing lanes to each lane of the road; The positioning module is used to determine the comprehensive probability of the target vehicle changing lanes to each lane based on the weighted average of the predicted probability and the observed probability, and determine the lane with the largest comprehensive probability as the lane where the target vehicle is located.

9. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; Wherein, the processor is configured to execute the instructions to implement the lane positioning method as described in any one of claims 1-7.

10. A vehicle, characterized in that: An electronic device comprising the electronic device described in claim 9.