Lane line existence judgment method, system and storage medium

Abstract

The present invention discloses a method for judging the existence of lane lines, which is based on environmental perception data and includes: obtaining environmental perception data; obtaining the feature probability distributions #imgabs0# and #imgabs1##imgabs2# under different classifications through a machine learning algorithm in the data with features and labels. #imgabs0# and #imgabs1##imgabs2# are the probabilities of the existence of the lane to be judged under the observation feature F L and #imgabs3# is the probability of the existence of the inner lane under the observation feature F L ; matching the perceived lane lines with the existing lane boundaries; judging the lane boundary types one by one according to whether there is a boundary type feature probability distribution under the condition of the existence of the outer lane; calculating the probabilities of the existence of each lane according to the lane lines and the lane line types, and if the probability is greater than a specified threshold, judging that the lane lines exist. The present invention converts the existence of the lane into a probability description through the naive Bayes algorithm. Tests with a large amount of road data show that the present invention can solve the problem of judging the accuracy of the existence of the lane under the condition of the uncertainty of the perception data, and avoid misjudgment and missed judgment.

Description

Technical Field

[0001] The present invention relates to the field of automobiles, and in particular to a method and system for determining the existence of a lane line for intelligent driving or advanced driving assistance. Background Art

[0002] In intelligent driving or advanced driver assistance systems, intelligent vehicles need to identify the road environment in which they are located. One of the important tasks is to determine whether the adjacent lane of the vehicle exists. Existing lane information recognition methods are divided into two categories. The first category is based on on-board sensors to identify lane lines, road edges, etc. within the road range and then process them to obtain lane information. The second category is based on pre-collected high-precision map data and matches the real-time positioning position of the vehicle with the map to obtain lane information. Due to the limited coverage of high-precision maps at present, the first category of methods based on on-board sensors is more commonly used. Lane information acquisition methods based on on-board sensors usually directly use sensor perception results to determine the existence of adjacent lanes of the vehicle, such as whether the left and right side lines of the lane are identified and whether the lane width is reasonable to determine whether the lane exists. Such methods do not consider the uncertainty of sensor perception results and are prone to wrong judgments when the sensor recognition error is large. On this basis, there are methods that use the confidence of lane line recognition to describe the uncertainty of sensor perception, but such methods still need to use a certain confidence threshold to determine whether the sensor perception result is valid, and the judgment process is still deterministic. The present invention proposes a lane existence judgment method based on the Naive Bayes algorithm, which replaces deterministic judgment with probabilistic description and can well cope with the uncertainty of sensor perception. Summary of the invention

[0003] A series of simplified concepts are introduced in the Summary of the Invention, which are all simplifications of the prior art in the field, and will be further described in detail in the Detailed Description of the Invention. The Summary of the Invention of the present invention does not mean to attempt to define the key features and essential technical features of the claimed technical solution, nor does it mean to attempt to determine the scope of protection of the claimed technical solution.

[0004] The technical problem to be solved by the present invention is to provide a lane line existence judgment method that can avoid the influence of environmental perception uncertainty on the accuracy of lane line existence, and a computer-readable storage medium for storing a program, which implements the steps in the lane line existence judgment method when executed.

[0005] The present invention also provides a lane line existence judgment system that can avoid the influence of environmental perception uncertainty on the accuracy of lane line existence.

[0006] In order to solve the above technical problems, the present invention provides a lane line existence judgment method, which is based on environmental perception data and includes the following steps:

[0007] S1, obtain environmental perception data;

[0008] S2, using machine learning algorithms to obtain feature probability distributions under different classifications in data with lane boundary line type features and lane presence or absence labels and

[0009] is the observed feature F L The probability of the lane to be judged under the condition is: is the observed feature F L The probability of the inner lane existing under the condition;

[0010] S3, matching the sensed lane line with the existing lane boundary;

[0011] S4, judging the lane boundary type one by one according to the probability distribution of the boundary type characteristics under the condition of whether there is an outer lane;

[0012] S5, calculating the probability of each lane existing according to the lane line and the lane line type, and determining that the lane line exists if the probability is greater than a specified threshold.

[0013] Among them, the lane line existence judgment method is that the machine learning algorithm is a naive Bayes algorithm.

[0014] Optionally, the lane line existence determination method implements step S3 in the following manner:

[0015] The lane boundary is set according to the lane width, and the nearest lane boundary is selected as a match based on the distance between the perceived lane line and the lane boundary.

[0016] Optionally, the lane line existence judgment method is used, and the lane boundary types are divided into traversable lane lines, non-traversable lines, road edges and unknown types.

[0017] The probability of lane existence is calculated using the following formula;

[0018]

[0019] F L = {B inner , B outer}|

[0020]

[0021]

[0022] is the observed feature F under the condition that the lane to be judged and the inner lane existL The distribution probability of is the observed feature F under the condition that the inner lane exists but the lane to be judged does not exist L The distribution probability, B inner is the inner boundary of the lane, B outer is the outer boundary of the lane, is the probability distribution of the inner boundary type characteristics under the condition that the inner lane exists and the lane to be determined exists, is the probability distribution of the inner boundary type feature when the inner lane exists and the lane to be determined does not exist, is the probability distribution of the outer boundary type characteristics under the condition that the inner lane exists and the lane to be determined exists, is the probability distribution of the outer boundary type feature when the inner lane exists and the lane to be determined does not exist, is the probability distribution of the outer boundary type characteristics under the condition that the lane to be judged exists and the outer lane exists, is the probability distribution of the outer boundary type characteristics under the condition that the lane to be judged exists and the outer lane does not exist, It is the probability distribution of the outer boundary type feature under the condition that the lane to be judged does not exist and the outer lane does not exist.

[0023] Optionally, in the lane line existence determination method, the specified threshold range is 0.6 to 0.8, preferably 0.7.

[0024] In order to solve the above technical problems, the present invention provides a computer-readable storage medium, which stores a program internally. When the program is executed, the steps in any one of the above-mentioned lane line existence judgment methods are implemented.

[0025] In order to solve the above technical problems, the present invention provides a lane line existence judgment system, which is based on environmental perception data and includes:

[0026] A receiving module, which is used to receive environmental perception data;

[0027] The learning module uses a machine learning algorithm to obtain the feature probability distribution under different classifications from the data with lane boundary line type features and lane presence or absence labels. and

[0028] is the observed feature F L The probability of the lane to be judged under the condition is: is the observed feature F L The probability of the inner lane existing under the condition;

[0029] A matching module, which is used to match the perceived lane lines with the existing lane boundaries;

[0030] A first judgment module, which judges the lane boundary type one by one according to the probability distribution of the boundary type feature under the condition of whether there is an outer lane;

[0031] The second judgment module calculates the probability of each lane existing according to the lane line and the lane line type, and judges that the lane line exists if the probability is greater than a specified threshold.

[0032] Wherein, the machine learning algorithm is a Naive Bayes algorithm.

[0033] Optionally, the lane line existence judgment system, the matching module matches the sensed lane line with the existing lane boundary in the following manner;

[0034] The lane boundary is set according to the lane width, and the nearest lane boundary is selected as a match based on the distance between the perceived lane line and the lane boundary.

[0035] Optionally, the lane line existence judgment system divides lane boundary types into traversable lane lines, non-traversable lines, road edges and unknown types.

[0036] The second judgment module uses the following formula to calculate the probability of lane existence:

[0037]

[0038] F L = {B inner , B outer}|

[0039]

[0040] is the observed feature F under the condition that the lane to be judged and the inner lane exist L The distribution probability of is the observed feature F under the condition that the inner lane exists but the lane to be judged does not exist L The distribution probability, B inner is the inner boundary of the lane, B outer is the outer boundary of the lane, is the probability distribution of the inner boundary type characteristics under the condition that the inner lane exists and the lane to be determined exists, is the probability distribution of the inner boundary type feature when the inner lane exists and the lane to be determined does not exist, is the probability distribution of the outer boundary type characteristics under the condition that the inner lane exists and the lane to be determined exists, is the probability distribution of the outer boundary type feature when the inner lane exists and the lane to be determined does not exist, is the probability distribution of the outer boundary type characteristics under the condition that the lane to be judged exists and the outer lane exists, is the probability distribution of the outer boundary type characteristics under the condition that the lane to be judged exists and the outer lane does not exist, It is the probability distribution of the outer boundary type feature under the condition that the lane to be judged does not exist and the outer lane does not exist.

[0041] The design principle of the present invention is further described as follows:

[0042] The present invention expresses the lane existence judgment problem as a naive Bayes classification problem. For a lane to be judged, the classification includes two categories: lane existence and lane non-existence. The observation feature is the type of the left and right side lines of the lane. The side line types are divided into crossable lane lines, non-crossable lines, road edges and unknown types.

[0043] The Naive Bayes algorithm is a machine learning algorithm that is often used to solve classification problems. The general form of the Naive Bayes algorithm is as follows:

[0044]

[0045] Where P(C|F1, F2...F n ) indicates that in the observed features F1, F2...F n The probability distribution of the sample category C under the condition, P(C) is the prior probability distribution of the sample classification, which is usually obtained from empirical knowledge, ( F i |C), i=1, 2...n is the feature F under the known sample classification condition i The probability distribution of is usually obtained through machine learning in a large amount of sample data, P( F1 , F2...F n ) is the feature occurrence probability.

[0046] The present invention expresses the lane existence judgment problem as a naive Bayes classification problem. For a lane to be judged, the classification includes two categories: lane existence and lane non-existence. The observation feature is the type of the left and right side lines of the lane. The side line types are divided into crossable lane lines, non-crossable lines, road edges and unknown types.

[0047] Assume that the lane where the vehicle is located must exist and the lanes on the road are continuous. Therefore, for other lanes, the premise for the existence of the lane to be judged is the existence of the inner adjacent lane. For the five-lane scenario, the correspondence between the inner lane and the outer lane of the lane to be judged is shown in the following table.

[0048] Table 1 Correspondence between inner lane and outer lane

[0049] Lane to be judged Left left lane Left Lane Right Lane Right lane Inside lane Left Lane Self-driving car Self-driving car Right Lane Outside lane none Left left lane Right lane none

[0050] According to the total probability formula, the problem of the existence of the lane to be determined can be converted into the problem of the probability of the existence of the lane to be determined under the conditions of the existence and non-existence of the inner lane, that is:

[0051]

[0052] In the formula is the observed feature F L The probability of the lane to be judged under the condition is: is the observed feature F L The probability of the inner lane existing under the condition, is the probability of the existence of the lane to be determined under the condition that the inner lane exists, is the probability of the existence of the lane to be judged when the inner lane does not exist, is the observed feature F L The probability that the inner lane does not exist under the condition.

[0053] According to the assumption, the premise of the lane to be judged is the existence of the inner adjacent lane, so Will Substituting into (2) we obtain:

[0054]

[0055] Since the lane where the vehicle is located is the inner lane of the adjacent lanes on the left and right sides, and the inner lanes of the left-left and right-right lanes are the left and right lanes respectively, there is a probability that the inner lane It can be calculated recursively. According to Table 1, for the left lane and the right lane, the inner lane is the self-lane, which must exist, so we have:

[0056]

[0057] Further calculation can obtain the probability of the left lane and the right lane existing and For the left-left lane and the right-right lane, their inner lanes are the left lane and the right lane respectively, so:

[0058]

[0059] The probability of lane existence under the condition of inner lane existence Transformed by Bayesian formula:

[0060]

[0061] in is the observed feature F under the condition that the lane to be judged and the inner lane exist L The distribution probability of is the probability of the existence of the lane to be judged when there is no observed feature and the inner lane exists, is the observed feature F L and the joint probability distribution of the existence of the inner lane.

[0062] According to the total probability formula, the observed feature F L The joint probability distribution of the existence of the inner lane line is expressed as:

[0063]

[0064] in F is the observed feature under the condition that the inner lane exists but the lane to be judged does not exist L The distribution probability of is the probability that the inner lane exists but the lane to be judged does not exist when there is no observed feature.

[0065] Assume that when there is no observed feature, the probability of the lane to be judged existing is equal under the condition that the inner lane exists, that is:

[0066]

[0067] Substituting into equation (6) and (7) we get

[0068]

[0069] For the lane to be judged, its features include the type of the inner boundary and the type of the outer boundary: F L = {B inner , B outerr} That is, the inner boundary B inner and the outer boundary B outer , Naive Bayes assumes that features are independent of each other, so:

[0070]

[0071] in and They are the probability distributions of the inner boundary type features under the conditions of the existence of the inner lane and the existence and non-existence of the lane to be judged, respectively, which can be obtained through statistics in the data.

[0072] and are the probability distributions of the outer boundary type features under the conditions of the presence of the inner lane and the presence and absence of the lane to be determined, respectively. Because the lanes on both sides of the outer boundary are the lane to be determined and the outer lane, they can be calculated using the total probability formula:

[0073]

[0074] in Indicates the existence of the outer lane and the absence of the outer lane respectively. The outer boundary B is the condition that the inner lane and the outer lane of the lane to be judged exist. outer The probability distribution of The outer boundary B is the condition that the inner lane of the lane to be judged exists but the outer lane does not exist. outer The probability distribution of and are the probabilities of the existence and non-existence of the outer lane under the condition that the lane to be judged and the inner lane exist, respectively. Assuming that the probability of the existence and non-existence of the outer lane is equal when there is no observation feature:

[0075]

[0076] Substituting into formula (12) we can obtain:

[0077]

[0078] Because the distribution of boundary lines is only related to the existence of lanes on both sides, therefore:

[0079]

[0080] Substituting in:

[0081]

[0082] Similarly, for It can be calculated that:

[0083]

[0084] in and They are respectively the probability distribution of the outer boundary type feature under the condition that the outer lane exists in the lane to be judged, the probability distribution of the outer boundary type feature under the condition that the outer lane does not exist in the lane to be judged, and the probability distribution of the outer boundary type feature under the condition that the outer lane does not exist in the lane to be judged, which can be obtained through data statistics.

[0085] As shown in the following table, the probability distribution of the boundary line under different conditions of the existence of lanes on both sides is obtained from the data of known features and labels. From the data in the table, it can be seen that when the inner lane and the lane to be judged exist at the same time, the boundary line type between the two lanes is more likely to be a crossable lane line, such as a dotted line, a dotted and solid line, etc., while the probability of the edge of the road is extremely low. When the inner lane exists and the lane to be judged does not exist, the boundary line type between the two lanes is most likely to be a non-crossable lane line, such as a solid line, a double solid line, etc., and it may also be a road edge, with a very small probability of being a crossable lane line. The feature probability distribution obtained from the data is consistent with human observation experience of actual roads.

[0086] Table 2 Lane line feature probability distribution

[0087]

[0088] After sensing the lane lines and lane line types, the probability of each lane existing can be recursively calculated according to formulas (3)-(5), (9)-(11), (17)-(18). If the probability is greater than the threshold, the lane is considered to exist.

[0089] The present invention converts lane existence into a probabilistic description through a naive Bayes algorithm. Tests on a large amount of road data show that the present invention can solve the problem of judging the accuracy of lane existence under the condition of uncertainty in perception data and avoid misjudgment and missed judgment. BRIEF DESCRIPTION OF THE DRAWINGS

[0090] The drawings of the present invention are intended to illustrate the general characteristics of the methods, structures and / or materials used in specific exemplary embodiments of the present invention, and to supplement the description in the specification. However, the drawings of the present invention are schematic diagrams not drawn to scale, and thus may not accurately reflect the precise structure or performance characteristics of any given embodiment, and the drawings of the present invention should not be interpreted as limiting or restricting the range of values ​​or properties covered by the exemplary embodiments of the present invention. The present invention is further described in detail below in conjunction with the drawings and specific embodiments:

[0091] Figure 1 It is a schematic diagram of the process of the present invention.

[0092] Figure 2 This is a schematic diagram of lane line and lane matching.

[0093] Description of Reference Numerals

[0094] w e , w l , w ll , w r , w rr Indicates lane width. DETAILED DESCRIPTION

[0095] The following describes the implementation methods of the present invention through specific specific embodiments, and those skilled in the art can fully understand other advantages and technical effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through different specific implementation methods, and the details in this specification can also be applied based on different viewpoints, and various modifications or changes can be made without deviating from the overall design concept of the invention. It should be noted that the following embodiments and the features in the embodiments can be combined with each other in the absence of conflict. The following exemplary embodiments of the present invention can be implemented in a variety of different forms and should not be interpreted as being limited to the specific embodiments described herein. It should be understood that these embodiments are provided to make the disclosure of the present invention thorough and complete, and to fully convey the technical solutions of these exemplary specific embodiments to those skilled in the art.

[0096] First embodiment;

[0097] The present invention provides a method for determining the existence of a lane line, which is based on environmental perception data and includes the following steps:

[0098] S1, obtain environmental perception data;

[0099] S2, using the naive Bayes algorithm to obtain the feature probability distribution under different classifications in the data with lane boundary line type features and lane presence or absence labels and Refer to Table 2 above;

[0100] is the observed feature F L The probability of the lane to be judged under the condition is: is the observed feature F L The probability of the inner lane existing under the condition;

[0101] S3, based on the lane boundary set by the lane width, selects the nearest lane boundary as the match according to the distance between the perceived lane line and the lane boundary, refer to Figure 2 As shown;

[0102] S4, according to the probability distribution of boundary type characteristics under the condition of whether there is an outer lane, the lane boundary type is judged one by one, including the crossable lane line, the non-crossable line, the road edge and the unknown type; reference Figure 2 In the middle, the dashed line is the lane that can be crossed, and the solid line is the lane that cannot be crossed;

[0103] S5, calculating the probability of each lane existing according to the lane line and the lane line type, and determining that the lane line exists if the probability is greater than a specified threshold.

[0104] According to the above formulas (3)-(5), (9)-(11), (17)-(18), the probability of each lane existing is recursively calculated. It is known that the self-driving lane must exist, that is, the probability of the self-driving lane existing is For the left lane, it and the inner lane boundary line are crossable lane lines, according to the above Table 2:

[0105]

[0106] The outer lane boundary is a non-crossable line. According to Table 2 and formulas (17) and (18), we get:

[0107]

[0108] Substituting into equation (10) (11) we get:

[0109]

[0110] Substituting into equation (9)(3) we get the probability of the left lane existing:

[0111] The threshold is preferably specified to be 0.7, and the lane existence probability is greater than 0.7 to determine that the left lane exists.

[0112] Second embodiment;

[0113] The present invention provides a computer-readable storage medium (including semiconductors, magnetic cores, magnetic drums, magnetic tapes, laser disks, etc.), which stores a program internally. When the program is executed, the steps in the lane line existence judgment method described in the first embodiment are implemented.

[0114] Second embodiment;

[0115] The present invention provides a lane line existence judgment system, which is based on environmental perception data and can be implemented using existing hardware through computer programming technology, including:

[0116] A receiving module, which is used to receive environmental perception data;

[0117] A learning module that uses the naive Bayes algorithm to obtain the feature probability distribution under different classifications from the data with lane boundary line type features and lane presence or absence labels and

[0118] is the observed feature F L The probability of the lane to be judged under the condition is: is the observed feature F L The probability of the inner lane existing under the condition;

[0119] The matching module selects the nearest lane boundary as a match based on the lane boundary set according to the lane width and the distance between the perceived lane line and the lane boundary;

[0120] The first judgment module judges the lane boundary type one by one according to the probability distribution of the boundary type characteristics under the condition of whether there is an outer lane. The lane boundary types are divided into traversable lane lines, non-traversable lines, road edges and unknown types;

[0121] The second judgment module calculates the probability of each lane existing according to the lane line and the lane line type, and judges that the lane line exists if the probability is greater than a specified threshold;

[0122] The second judgment module uses the following formula to calculate the probability of lane existence;

[0123]

[0124] F L = {B inner , B outer}|

[0125]

[0126] is the observed feature F under the condition that the lane to be judged and the inner lane exist L The distribution probability of is the observed feature F under the condition that the inner lane exists but the lane to be judged does not exist L The distribution probability, B inner is the inner boundary of the lane, B outer is the outer boundary of the lane, is the probability distribution of the inner boundary type characteristics under the condition that the inner lane exists and the lane to be determined exists, is the probability distribution of the inner boundary type feature when the inner lane exists and the lane to be determined does not exist, is the probability distribution of the outer boundary type characteristics under the condition that the inner lane exists and the lane to be determined exists, is the probability distribution of the outer boundary type feature when the inner lane exists and the lane to be determined does not exist, is the probability distribution of the outer boundary type characteristics under the condition that the lane to be judged exists and the outer lane exists, is the probability distribution of the outer boundary type characteristics under the condition that the lane to be judged exists and the outer lane does not exist, It is the probability distribution of the outer boundary type feature under the condition that the lane to be judged does not exist and the outer lane does not exist.

[0127] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those commonly understood by those skilled in the art to which the present invention belongs. It will also be understood that, unless expressly defined herein, terms such as those defined in general dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art, and not interpreted in an ideal or overly formal sense.

[0128] The present invention has been described in detail above through specific implementation modes and embodiments, but these do not constitute limitations of the present invention. Without departing from the principles of the present invention, those skilled in the art may also make many variations and improvements, which should also be regarded as the protection scope of the present invention.

Claims

1. A method for determining the existence of a lane line, based on environmental perception data, characterized in that: The following steps are involved: S1, obtain environmental perception data; S2, using machine learning algorithms to obtain feature probability distributions under different classifications in data with lane boundary line type features and lane presence or absence labels and is the observed feature F L The probability of the lane to be judged under the condition is: is the observed feature F L The probability of the inner lane existing under the condition; S3, matching the sensed lane line with the existing lane boundary; S4, judging the lane boundary type one by one according to the probability distribution of the boundary type characteristics under the condition of whether there is an outer lane; S5, calculating the probability of each lane existing according to the lane line and the lane line type, and determining that the lane line exists if the probability is greater than a specified threshold.

2. The method for determining the presence of a lane line according to claim 1, wherein: The machine learning algorithm is the Naive Bayes algorithm.

3. The method for determining the presence of a lane line according to claim 1, wherein: Step S3 is implemented in the following manner: The lane boundary is set according to the lane width, and the nearest lane boundary is selected as a match based on the distance between the perceived lane line and the lane boundary.

4. The method for determining the presence of a lane line according to claim 1, wherein: Lane boundary types are divided into traversable lane lines, non-traversable lines, road edges, and unknown types.

5. The method for determining the presence of a lane line according to claim 1, wherein: The probability of lane existence is calculated using the following formula; F L ={B inner ,B outer }| is the observed feature F under the condition that the lane to be judged and the inner lane exist L The distribution probability of is the observed feature F under the condition that the inner lane exists but the lane to be judged does not exist L The distribution probability, B inner is the inner boundary of the lane, B outer is the outer boundary of the lane, is the probability distribution of the inner boundary type characteristics under the condition that the inner lane exists and the lane to be determined exists, is the probability distribution of the inner boundary type feature when the inner lane exists and the lane to be determined does not exist, is the probability distribution of the outer boundary type characteristics under the condition that the inner lane exists and the lane to be determined exists, is the probability distribution of the outer boundary type feature when the inner lane exists and the lane to be determined does not exist, is the probability distribution of the outer boundary type characteristics under the condition that the lane to be judged exists and the outer lane exists, is the probability distribution of the outer boundary type characteristics under the condition that the lane to be judged exists and the outer lane does not exist, is the probability distribution of the outer boundary type characteristics under the condition that the lane to be judged does not exist and the outer lane does not exist, is the observed feature F L and the joint probability distribution of the existence of the inner lane, is the probability of the existence of the lane to be determined when the inner lane exists.

6. The method for determining the presence of a lane line according to claim 1, wherein: The specified threshold range is 0.6 to 0.

8.

7. A computer-readable storage medium, characterized in that: A program is stored inside the device, and when the program is executed, the steps in the lane line existence judgment method described in any one of claims 1-6 are implemented.

8. A lane line existence judgment system based on environmental perception data, characterized in that: include: A receiving module, which is used to receive environmental perception data; A learning module that uses a machine learning algorithm to obtain the feature probability distribution under different classifications in data with lane boundary line type features and lane presence or absence labels and is the observed feature F L The probability of the lane to be judged under the condition is: is the observed feature F L The probability of the inner lane existing under the condition; A matching module, which is used to match the perceived lane lines with the existing lane boundaries; A first judgment module, which judges the lane boundary type one by one according to the probability distribution of the boundary type feature under the condition of whether there is an outer lane; The second judgment module calculates the probability of each lane existing according to the lane line and the lane line type, and judges that the lane line exists if the probability is greater than a specified threshold.

9. The lane line existence determination system according to claim 8, characterized in that: The machine learning algorithm is the Naive Bayes algorithm.

10. The lane line existence determination system according to claim 8, characterized in that: The matching module matches the sensed lane lines with the existing lane boundaries in the following way; The lane boundary is set according to the lane width, and the nearest lane boundary is selected as a match based on the distance between the perceived lane line and the lane boundary.

11. The lane line existence determination system according to claim 8, characterized in that: Lane boundary types are divided into traversable lane lines, non-traversable lines, road edges, and unknown types.

12. The lane line existence determination system according to claim 8, characterized in that: The second judgment module uses the following formula to calculate the probability of lane existence; F L ={B inner ,B outer } is the observed feature F under the condition that the lane to be judged and the inner lane exist L The distribution probability of is the observed feature F under the condition that the inner lane exists but the lane to be judged does not exist L The distribution probability, B inner is the inner boundary of the lane, B outer is the outer boundary of the lane, is the probability distribution of the inner boundary type characteristics under the condition that the inner lane exists and the lane to be determined exists, is the probability distribution of the inner boundary type feature when the inner lane exists and the lane to be determined does not exist, is the probability distribution of the outer boundary type characteristics under the condition that the inner lane exists and the lane to be determined exists, is the probability distribution of the outer boundary type feature when the inner lane exists and the lane to be determined does not exist, is the probability distribution of the outer boundary type characteristics under the condition that the lane to be judged exists and the outer lane exists, is the probability distribution of the outer boundary type characteristics under the condition that the lane to be judged exists and the outer lane does not exist, is the probability distribution of the outer boundary type characteristics under the condition that the lane to be judged does not exist and the outer lane does not exist, is the observed feature F L and the joint probability distribution of the existence of the inner lane, is the probability of the existence of the lane to be determined when the inner lane exists.

Images