A road guardrail state detection method based on artificial intelligence

Abstract

The present invention belongs to the technical field of road guardrail status detection, and specifically relates to a road guardrail status detection method based on artificial intelligence. It includes using machine vision, ultrasonic sensors and laser radar to collect guardrail status information; using improved U-Net network architecture to perform intelligent enhancement processing on image data, improving adaptive filtering to process distance data, and improving cluster analysis to separate guardrail point cloud from surrounding environment; introducing knowledge graph enhancement fusion to construct a multimodal fusion neural network, and outputting guardrail status feature vectors; finally, using advanced status evaluation algorithms, combined with fuzzy logic and expert systems, to determine the final state of the guardrail and issue an alarm message. The present invention can accurately and quickly detect the state of the guardrail and improve the level of road traffic safety.

Description

Technical Field

[0001] The present invention belongs to the technical field of road guardrail status detection, and particularly relates to a method for detecting the status of road guardrails based on artificial intelligence. Background Art

[0002] Road guardrails are important facilities for ensuring road traffic safety, and the quality of their status directly affects driving safety. However, during long-term use, guardrails may have problems such as damage and deformation, which need to be detected and maintained in a timely manner. The existing methods for detecting the status of guardrails mainly rely on manual inspections. This method is not only inefficient but also prone to missed detections and false detections. With the rapid development of artificial intelligence technology, applying it to the detection of road guardrail status has become a trend. At present, some detection methods based on artificial intelligence have achieved certain results, but there are still some deficiencies. For example, in terms of data collection, a single sensor may not be able to comprehensively and accurately obtain the status information of guardrails; in terms of data processing, traditional image processing and analysis algorithms may be difficult to effectively handle complex environments and changing guardrail statuses; in terms of status evaluation, there is a lack of effective methods for integrating multiple features, resulting in inaccurate and unreliable evaluation results. Therefore, a method for detecting the status of road guardrails based on artificial intelligence is needed, which can comprehensively utilize the data collected by multiple sensors, adopt advanced image processing and analysis technologies, achieve accurate and rapid detection of the status of guardrails, timely discover potential problems of guardrails, and improve the level of road traffic safety. Summary of the Invention

[0003] The present invention aims at the above-mentioned existing technical problems and proposes a method for detecting the status of road guardrails based on artificial intelligence, which is simple, easy to operate and can effectively solve the problems.

[0004] To achieve the above object, the technical solution adopted by the present invention is as follows, including the following steps:

[0005] S1. First, use machine vision, ultrasonic sensors, and lidar to collect comprehensive and comprehensive road guardrail status information;

[0006] S2. Secondly, use the improved U-Net network architecture to perform intelligent enhancement processing on the collected machine vision image data; use the improved adaptive filtering to process the distance data detected by the ultrasonic sensors; perform improved clustering analysis on the three-dimensional point cloud data collected by the lidar to quickly separate the point cloud of the guardrail from the surrounding environment and improve the subsequent processing efficiency;

[0007] S3. Then, introduce knowledge graph enhancement fusion to construct a multi-modal fusion neural network, input the preprocessed image data, distance data, and three-dimensional point cloud data into this network, and output a comprehensive guardrail status feature vector;

[0008] S4. Finally, use an advanced state evaluation algorithm to analyze the fused feature vectors, judge the state of the guardrail, introduce an evaluation method that combines fuzzy logic and an expert system, and perform fuzzy reasoning and decision-making based on various indicators in the fused feature vectors to determine the final state of the guardrail and send corresponding alarm information in a timely manner;

[0009] The specific operation of introducing knowledge graph enhancement fusion in step S3 to construct a multi-modal fusion neural network and achieve the comprehensive output of the comprehensive guardrail state feature vector is as follows:

[0010] S31. First, construct a guardrail knowledge graph G=(V, E), where V is the set of nodes and E is the set of edges. For image data, use a convolutional neural network for feature extraction. For distance data, use a fully connected network for feature extraction. For 3D point cloud data, use a point cloud processing network for feature extraction;

[0011] S32. Secondly, define a fusion function. The fused comprehensive feature vector is where F merge is the fusion operation function, G represents the knowledge graph, are the feature extractions of image, distance, and point cloud data enhancement respectively;

[0012] S33. Then, perform a fusion operation. Using the weighted summation method, the fused feature vector is where ω I , ω D , ω P are the weight coefficients, and their weights are set through the information in the knowledge graph;

[0013] S34. Finally, after the knowledge graph-guided feature fusion, a comprehensive guardrail state feature vector is obtained for subsequent guardrail state judgment and analysis.

[0014] Preferably, the implementation of using the improved U-Net network architecture to perform intelligent enhancement processing on the collected machine vision image data in step S2 is as follows:

[0015] S211. First, the input image is I(x, y), and the output feature map of the l-th layer of the encoder is The output feature map of the l-th layer of the decoder is The calculation formula of the encoder is where represents the l-th layer convolution operation of the encoder. The calculation formula of the decoder is where represents the l-th layer convolution operation of the decoder, represents the splicing operation of the feature maps;

[0016] S212. Then, perform multi-scale feature fusion and introduce the attention mechanism. In the loss function, the loss function consists of the mean square loss function The structural similarity index loss function L ssim = 1 - SSIM(I enhanced , I ideal ), the perceptual loss function The adversarial loss function L adv = log(1 - D(I enhanced )) + log(D(I ideal ))), where I, I enhanced I ideal are the original image, the enhanced image, and the ideal enhanced image respectively, Φ represents the features extracted by the pre-trained neural network, D is the discriminator network used to distinguish the enhanced image from the ideal enhanced image, and the total loss function L = αL mse + βL ssim + γL perceptual + δL adv ), where α, β, γ, δ are weight coefficients;

[0017] S213. Finally, after improvement, the collected images are input into the network and the enhanced images are output.

[0018] Preferably, the implementation of improving the adaptive filter to process the distance data detected by the ultrasonic sensor in step S2 is as follows:

[0019] S221. First, design the structure of the filter. The finite impulse response filter structure is adopted, and its output is where d(t - i) is the distance data sequence, t represents time, N is the length of the filter, and ω i (t) is the coefficient of the filter at time t;

[0020] S222. Secondly, based on the gradient descent method, combine the momentum term and the learning rate adjustment strategy to update the filter coefficients where μ is the learning rate, is the gradient of the loss function L with respect to the coefficient ω i (t), and α is the momentum coefficient; the learning rate is adjusted according to the exponentially decaying learning rate adjustment strategy μ(t) = μ 0 γ t , where μ 0 is the initial learning rate and γ is the decay coefficient;

[0021] S223. Finally, after improvement, input the new data into the filter and calculate the output of the filter.

[0022] Preferably, the implementation of the improved clustering analysis in step S2 to quickly separate the point cloud of the guardrail from the surrounding environment is as follows:

[0023] S231. First, perform feature extraction. For each point p i , calculate its normal vector n i and curvature c i . Define the similarity measure between points considering the spatial distance and the similarity of local features. Use the comprehensive similarity measure formula s(p i , p j ) = αd(p i , p j ) + β|cosθ i,j | + γ|c i - c j |, where is the spatial distance, θ i,j is the angle between the normal vectors of points p i and p j , and α, β, γ are weight coefficients;

[0024] S232. Then, adopt the density-based clustering algorithm. Select an unmarked point as the seed point, determine its neighborhood points according to the similarity measure. If the number of neighborhood points is greater than the threshold, mark these points as the same cluster and continue to expand the cluster. Repeat the steps until all points are marked;

[0025] S233. Finally, according to the prior features of the shape and height range of the guardrail, screen the clustering results to separate the point cloud belonging to the guardrail.

[0026] Preferably, the implementation of the evaluation method combining fuzzy logic and expert system in step S4 is as follows:

[0027] S41. First, establish a fuzzy logic system. Determine the input variables for evaluating the guardrail status and define fuzzy sets for each input variable. Determine the output variable as the guardrail status and define its fuzzy set. Determine the fuzzy rule R i : IFC img isA i ANDD dist isB i ANDI pc isC i THENS fence isD i , where A i , B i , C i , D i are the fuzzy sets of the corresponding variables respectively, and C img , D dist , Ipc , S fence are the image feature index, the distance feature index, and the point cloud feature index respectively;

[0028] S42. Then, fuzzify the input variables, convert the specific input values into the membership degrees of the corresponding fuzzy sets, perform fuzzy inference, and then defuzzify to convert into the specific road guardrail status values;

[0029] S43. Next, establish an expert knowledge database to store various rules, experiences, and cases for judging the guardrail status, which are used to supplement and correct the results of the fuzzy logic system;

[0030] S44. Finally, determine the final status of the guardrail according to the comprehensive results of the fuzzy logic and the expert system.

[0031] Compared with the prior art, the advantages and positive effects of the present invention are as follows: comprehensively use a variety of sensors to collect data, comprehensively and accurately obtain the guardrail status information; adopt improved algorithms and technologies to improve the data processing efficiency and accuracy; introduce knowledge graphs and fuzzy logic to more accurately judge the guardrail status; high degree of automation, reduce manual intervention, reduce costs, and improve the level of road traffic safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0033] Figure 1 is the implementation flow schematic diagram of the present invention;

[0034] Figure 2 is the structural schematic diagram of the multi-modal neural network; DETAILED DESCRIPTION OF THE EMBODIMENTS

[0035] In order to be able to more clearly understand the above objects, features, and advantages of the present invention, the following further describes the present invention in conjunction with embodiments. It should be noted that, without conflict, the embodiments of the present application and the features in the embodiments can be combined with each other.

[0036] In the following description, many specific details are set forth in order to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Therefore, the present invention is not limited by the specific embodiments disclosed in the following specification.

[0037] Embodiment. To improve the accuracy and efficiency of road guardrail status detection, promptly discover potential problems with guardrails, and ensure road traffic safety, this embodiment adopts a road guardrail status detection method based on artificial intelligence. The implementation process of the present invention is as follows Figure 1 as shown.

[0038] First, data collection is carried out. A machine vision system equipped with a high-resolution camera is installed on a patrol vehicle to collect images of the road guardrail at a certain frequency to ensure clear and comprehensive guardrail image information is obtained. Multiple ultrasonic sensors are installed on the patrol vehicle, evenly distributed around the vehicle, for real-time detection of the distance data between the guardrail and the sensors, covering all parts of the guardrail comprehensively. At the same time, a high-precision lidar is installed to collect the three-dimensional point cloud data of the guardrail to obtain the three-dimensional shape and spatial position information of the guardrail.

[0039] To improve the image quality and highlight the guardrail details and potential problem areas. The present invention uses an improved U-Net network architecture to perform intelligent enhancement processing on the machine vision image data. First, the input image is I(x,y), and the output feature map of the l-th layer of the encoder is The output feature map of the l-th layer of the decoder is The calculation formula of the encoder is where represents the l-th layer convolution operation of the encoder. The calculation formula of the decoder is where represents the l-th layer convolution operation of the decoder, represents the splicing operation of the feature maps; then multi-scale feature fusion is carried out and an attention mechanism is introduced. In the loss function, the loss function consists of the mean square loss function the structural similarity index loss function L ssim =1 - SSIM(I enhanced , I ideal ), the perceptual loss function and the adversarial loss function L adv =log(1 - D(I enhanced )) + log(D(I ideal ))), where II enhanced I ideal are the original image, the enhanced image, and the ideal enhanced image respectively. Φ represents the features extracted by the pre-trained neural network, and D is the discriminator network used to distinguish the enhanced image from the ideal enhanced image. Then the total loss function L = αL mse + βL ssim + γL perceptual + δL adv, where α, β, γ, and δ are weight coefficients; finally, after improvement, the collected images are input into the network and the enhanced images are output. This processing method can fuse multi-scale features and attention mechanisms, automatically learn the optimal enhancement method, improve the clarity, contrast, and accuracy of the images, and provide more reliable information for subsequent guardrail status detection.

[0040] Then, for the purpose of improving the accuracy of the distance data of the ultrasonic sensor by improving the adaptive filtering to remove noise interference, first is the filter structure design. The finite impulse response filter structure is adopted, and its output is where d(t - i) is the distance data sequence, t represents time, N is the length of the filter, and ω i (t) is the coefficient of the filter at time t; secondly, based on the gradient descent method, combined with the momentum term and the learning rate adjustment strategy to update the filter coefficients where μ is the learning rate, is the gradient of the loss function L with respect to the coefficient ω i (t), and α is the momentum coefficient; the learning rate is adjusted according to the exponentially decaying learning rate adjustment strategy μ(t) = μ 0 γ t , where μ 0 is the initial learning rate, and γ is the decay coefficient; finally, after improvement, the new data is input into the filter, and the output of the filter is calculated.

[0041] In addition, to accurately extract the point cloud data of the guardrail for more precise analysis of the guardrail status, the present invention uses improved clustering analysis to quickly separate the point cloud of the guardrail from the surrounding environment. First, feature extraction is performed. For each point p i , its normal vector n i and curvature c i are calculated. The similarity measure between points is defined considering the similarity of spatial distance and local features. The comprehensive similarity measure formula s(p i , p j ) = αd(p i , p j ) + β|cosθ i,j | + γ|c i - c j | is used, where is the spatial distance, θ i,j is the angle between point p i and p jThe angles between the normal vectors are α, β, and γ, which are weight coefficients. Then, a density-based clustering algorithm is used. An unlabeled point is selected as the seed point, and its neighborhood points are determined according to the similarity metric. If the number of neighborhood points is greater than the threshold, these points are labeled as the same cluster, and the cluster is continuously expanded. Repeat the steps until all points are labeled. Finally, based on the prior features of the guardrail's shape and height range, the clustering results are screened to separate the point cloud belonging to the guardrail. This measure can combine the prior features of the guardrail, such as shape and height range, to quickly and accurately separate the guardrail point cloud, improve the efficiency and accuracy of subsequent processing, and reduce the cases of misjudgment and missed judgment.

[0042] Considering that the multi-modal fusion neural network may have defects such as insufficient fusion of different modal data and difficulty in effectively using prior knowledge. The advantage of introducing the knowledge graph for enhanced fusion is that it can better guide feature extraction and fusion, enhance the fusion of key information, and enable the network to more accurately output the comprehensive guardrail status feature vector. The structure of the multi-modal fusion neural network is as Figure 2 shown. First, a guardrail knowledge graph G=(V, E) is constructed, where V is the set of nodes and E is the set of edges. For image data, a convolutional neural network is used for feature extraction. For distance data, a fully connected network is used for feature extraction. For 3D point cloud data, a point cloud processing network is used for feature extraction. In the state detection of road guardrails in this embodiment, for the image data collected by machine vision, the convolutional neural network slides the convolutional kernel on the image, multiplies and sums with the elements in the local area of the image, and extracts more abstract features as the network deepens. Features such as texture and shape are extracted from the guardrail image to provide a basis for subsequent judgment of the guardrail status. For the distance data of the ultrasonic sensor, the fully connected network takes it as the input, and the neurons are fully connected and processed through linear transformation by the weight matrix and the activation function to extract features such as the change trend of the distance between the guardrail and the sensor to assist in judging whether there is a displacement of the guardrail. For the 3D point cloud data collected by the lidar, the point cloud processing network first calculates features such as the normal vector and curvature of each point, and then integrates the spatial information. By analyzing the spatial distribution features of the guardrail point cloud, the integrity and shape change of the guardrail are judged.

[0043] Secondly, a fusion function is defined. The fused comprehensive feature vector is where F merge is the fusion operation function, G represents the knowledge graph, are the feature extractions enhanced by image, distance, and point cloud data respectively. Then, the fusion operation is carried out. In the weighted summation method, the fused feature vector is where ω I , ω D , ω Pis the weight coefficient, and its weight is set according to the information in the knowledge graph, and is set according to the connection strength between nodes and the importance factors of edges. There are nodes A and B in the knowledge graph, and the edge weight between them is ω AB In the detection of the guardrail state, if the data related to A has a strong impact on the final result, then a larger weight is given to the data corresponding to it during fusion; finally, after the feature fusion guided by the knowledge graph, a comprehensive guardrail state feature vector is obtained for subsequent guardrail state judgment and analysis.

[0044] Finally, the present invention considers introducing an evaluation method that combines fuzzy logic and an expert system. According to various indicators in the fusion feature vector, fuzzy reasoning and decision-making are carried out to determine the final state of the guardrail. First, a fuzzy logic system is established, the input variables for evaluating the guardrail state are determined and fuzzy sets are defined for each input variable, the output variable is determined as the guardrail state and its fuzzy set is defined, and the fuzzy rule R is determined according to expert experience and knowledge i :IFC img isA i ANDD dist isB i ANDI pc isC i THENS fence isD i where A i ,B i ,C i ,D i are the fuzzy sets of the corresponding variables respectively, C img ,D dist ,I pc ,S fenceThey are respectively the image feature index, the distance feature index, and the point cloud feature index. Then, the input variables are fuzzified, converting the specific input values into the membership degrees of the corresponding fuzzy sets, followed by fuzzy inference, and then defuzzified to convert into the specific road guardrail status value. The implementation method of the fuzzy rules is to first determine the fuzzy rules based on expert experience and knowledge. These rules are based on the fuzzy sets of the input variables including the image feature index, the distance feature index, and the point cloud feature index, as well as the fuzzy set of the output variable including the guardrail status. Then, the input variables are fuzzified, converting the specific numerical values into the membership degrees of the corresponding fuzzy sets. Then, fuzzy inference is carried out according to the fuzzy rules, and the inference result is obtained by comprehensively considering the membership degrees of each input variable. Finally, defuzzification is performed to convert the inference result into the specific road guardrail status value, thereby determining the final status of the guardrail. Next, an expert knowledge database is established to store various rules, experiences, and cases regarding the judgment of the guardrail status, which are used to supplement and correct the results of the fuzzy logic system. Finally, based on the comprehensive results of the fuzzy logic and the expert system, the final status of the guardrail is determined. This way of status judgment and analysis can fully consider the uncertainty and fuzziness of various indicators in the fusion feature vector, and perform more flexible reasoning and decision-making. Combining the knowledge and experience of the expert system can improve the accuracy and reliability of the evaluation and more accurately determine the final status of the guardrail.

[0045] The above are only the preferred embodiments of the present invention, and are not intended to limit the present invention in any other form. Any person skilled in the art may use the technical content disclosed above to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change, and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A road guardrail status detection method based on artificial intelligence, characterized in that: The following steps are involved: S1. First, use machine vision, ultrasonic sensors and lidar to collect comprehensive road guardrail status information; S2. Secondly, the improved U-Net network architecture is used to perform intelligent enhancement processing on the collected machine vision image data; Improved adaptive filtering is used to process the distance data detected by the ultrasonic sensor; Improved cluster analysis of the 3D point cloud data collected by the LiDAR can quickly separate the point cloud of the guardrail from the surrounding environment, improving the efficiency of subsequent processing; S3, then introduce knowledge graph enhanced fusion to construct a multimodal fusion neural network, input the preprocessed image data, distance data and three-dimensional point cloud data into the network, and output a comprehensive guardrail state feature vector; S4. Finally, the fused feature vector is analyzed using an advanced state assessment algorithm to determine the state of the guardrail. An assessment method combining fuzzy logic and expert system is introduced to perform fuzzy reasoning and decision-making based on various indicators in the fused feature vector to determine the final state of the guardrail and issue corresponding alarm information in a timely manner. In step S3, the knowledge graph enhancement fusion is introduced to construct a multimodal fusion neural network, and the specific operation of realizing the comprehensive output of the comprehensive guardrail state feature vector is as follows: S31. First, construct a guardrail knowledge graph G = (V, E), where V is a node set and E is an edge set. For image data, a convolutional neural network is used for feature extraction. For distance data, a fully connected network is used for feature extraction. For three-dimensional point cloud data, a point cloud processing network is used for feature extraction. S32, then define the fusion function, the fused comprehensive feature vector is Among them, F merge is the fusion operation function, G represents the knowledge graph, They are feature extraction for image, distance and point cloud data enhancement; S33, then perform a fusion operation, using a weighted summation method, and the fused feature vector is where ω I ,ω D ,ω P is the weight coefficient, whose weight is set by the information in the knowledge graph; S34. Finally, after feature fusion guided by the knowledge graph, a comprehensive guardrail status feature vector is obtained for subsequent guardrail status judgment and analysis.

2. The method for detecting road guardrail status based on artificial intelligence according to claim 1, characterized in that: The implementation of intelligent enhancement processing of the collected machine vision image data using the improved U-Net network architecture in step S2 is: S211, first input image is I(x,y), the output feature map of the encoder layer l is The lth output feature map of the decoder is The calculation formula of the encoder is in represents the lth convolution operation of the encoder, and the calculation formula of the decoder is in represents the l-th convolution operation of the decoder, Represents the concatenation operation of feature maps; S212, then multi-scale feature fusion is performed and the attention mechanism is introduced. In the loss function, the loss function is composed of the mean square loss function Structural similarity index loss function L ssim =1-SSIM(I enhanced , I ideal ), perceptual loss function Adversarial loss function L adv =log(1-D(I enhanced ))+log(D(I ideal )), where II enhanced I ideal are the original image, the enhanced image, and the ideal enhanced image, respectively. Φ represents the features extracted by the pre-trained neural network. D is the discriminator network used to distinguish the enhanced image from the ideal enhanced image. The total loss function L = αL mse +βL ssim +γL perceptual +δL adv , where α, β, γ, δ are weight coefficients; S213. Finally, after improvement, the collected image is input into the network and the enhanced image is output.

3. The road guardrail state detection method based on artificial intelligence according to claim 1 is characterized in that: The improved adaptive filtering in step S2 is implemented by processing the distance data detected by the ultrasonic sensor as follows: S221, the first is the structure design of the filter, using a finite impulse response filter structure, its output is Where d(ti) is the distance data sequence, t represents time, N is the length of the filter, ω i (t) is the coefficient of the filter at time t; S222, secondly, based on the gradient descent method, the momentum term and the learning rate adjustment strategy are combined to update the filter coefficients where μ is the learning rate, is the loss function L with respect to coefficient ω i The gradient of (t), α is the momentum coefficient; The learning rate is adjusted according to the exponentially decaying learning rate strategy μ(t)=μ0γ t , where μ0 is the initial learning rate and γ is the decay coefficient; S223. Finally, after the improvement, the new data is input into the filter and the output of the filter is calculated.

4. The road guardrail state detection method based on artificial intelligence according to claim 1 is characterized in that: The improved cluster analysis in step S2 is implemented to quickly separate the point cloud of the guardrail from the surrounding environment as follows: S231, first perform feature extraction, for each point p i , calculate its normal vector n i and curvature c i , the similarity measure between the defined points takes into account the similarity of spatial distance and local features, and the comprehensive similarity measure formula is s(p i ,p j )=αd(p i ,p j )+β|cosθ i,j |+γ|c i -c j |, where is the spatial distance, θ i,j It's point p i and p j The angle between the normal vectors, α, β, γ are weight coefficients; S232, then using a density-based clustering algorithm, select an unmarked point as a seed point, determine its neighborhood points based on a similarity metric, and if the number of neighborhood points is greater than a threshold, mark these points as the same cluster, and continue to expand the cluster, repeating the steps until all points are marked; S233. Finally, according to the prior features of the shape and height range of the guardrail, the clustering results are screened to separate the point cloud belonging to the guardrail.

5. The road guardrail state detection method based on artificial intelligence according to claim 1 is characterized in that: The evaluation method combining fuzzy logic and expert system introduced in step S4 is implemented as follows: S41. First, establish a fuzzy logic system, determine the input variables used to evaluate the guardrail status and define fuzzy sets for each input variable, determine the output variable as the guardrail status and define its fuzzy set, and determine the fuzzy rule R based on expert experience and knowledge. i :IFC img iA i ANDD dist iB i ANDI pc iC i THENS fence iD i , where A i , B i , C i , D i are the fuzzy sets of the corresponding variables, C img , D dist , I pc , S fence They are image feature index, distance feature index and point cloud feature index; S42, then fuzzify the input variables, convert the specific input values ​​into corresponding fuzzy set membership, perform fuzzy reasoning, and then defuzzify and convert them into specific road guardrail state values; S43, next, establish an expert knowledge database to store various rules, experiences and cases about the guardrail status judgment, which is used to supplement and correct the results of the fuzzy logic system; S44. Finally, the final state of the guardrail is determined based on the comprehensive results of fuzzy logic and expert system.

Images