Road adhesion coefficient identification method and system based on road side unit and network connection train set

Abstract

The invention discloses a road adhesion coefficient identification method based on a road side unit and a network-connected train set. The method comprises the following steps: shooting a road surface condition through a plurality of road side unit nodes arranged on the side of a road and obtaining a corresponding road surface type; comparing with a pre-stored road surface type and road surface adhesion coefficient comparison table to obtain a first road surface adhesion coefficient value of the corresponding road surface; the real-time information of the vehicle is synchronized back to the central cloud, and at the moment, the central cloud calculates a second road adhesion coefficient value of the current road according to the tire slip rate of the intelligent networked vehicle; and the center cloud carries out numerical value fusion based on the first road adhesion coefficient value and the second road adhesion coefficient value to obtain an accurate road adhesion coefficient and issues the accurate road adhesion coefficient to a vehicle. According to the invention, the cost of large-scale deployment can be effectively reduced, resource waste is saved, the accuracy and efficiency of vehicle-road-cloud fusion perception are improved, and the driving safety of the intelligent networked vehicle is ensured at the same time.

Description

Technical Field

[0001] The present invention relates to the field of vehicle technology, and in particular to a method and system for identifying a road adhesion coefficient based on a roadside unit and a networked vehicle train. Background Art

[0002] Road risk warning research has always been an important part of research. The diversity of road changes and changes in traffic flow, changes in road conditions, and high vehicle speeds can all lead to accidents. Therefore, the roadside perception system is the key to smart road construction. At present, the roadside perception facilities for highways include millimeter wave radars, high-definition cameras, micro-weather stations, etc., but the above systems have high laying costs and are difficult to meet the requirements of large-area coverage. Summary of the invention

[0003] The main purpose of the present invention is to provide a method and system for identifying the road adhesion coefficient based on a roadside unit and a connected vehicle train, aiming to solve the above-mentioned technical problems.

[0004] To achieve the above-mentioned purpose, the present invention provides a road adhesion coefficient identification method based on a roadside unit and a networked vehicle train.

[0005] The road adhesion coefficient identification method based on the roadside unit and the networked vehicle train comprises the following steps:

[0006] Photographing road conditions by a plurality of road test unit nodes arranged on the side of the road and determining corresponding road types after distinguishing the road conditions photographed by the plurality of road test units;

[0007] Compare with a pre-stored comparison table of road surface types and road surface adhesion coefficients to obtain a range of first road surface adhesion coefficient values ​​corresponding to the road surface;

[0008] The road test edge gateway collects the road adhesion coefficient values ​​of each road section, integrates and packages them, and sends them to the central cloud;

[0009] When the intelligent network-connected vehicle is started, the real-time information of the vehicle is synchronized back to the central cloud, and the central cloud calculates the second road adhesion coefficient value of the current road surface according to the tire slip rate of the intelligent network-connected vehicle itself;

[0010] The central cloud obtains an accurate road adhesion coefficient based on the numerical fusion of the first road adhesion coefficient value and the second road adhesion coefficient value and distributes it to the vehicle.

[0011] In one embodiment, the step of photographing road conditions by a plurality of road test unit nodes disposed beside the road and determining the road conditions photographed by the plurality of road test units to obtain the corresponding road type specifically includes:

[0012] The road test unit collects road conditions and performs image analysis on the road conditions to establish a SVM model;

[0013] The taken road surface pictures are classified according to the SVM model to obtain the results, and after statistics, the category with the largest number of occurrences is selected as the final classification result.

[0014] In one embodiment, the step of collecting road conditions by a road test unit and performing image analysis on the road conditions to establish an SVM model includes:

[0015] Obtain the color feature value and image texture value of the road image extracted based on the HSV color space and gray-level co-occurrence matrix;

[0016] Construct the feature vector of the road surface image and realize the road surface type recognition through the support vector machine algorithm;

[0017] A one-to-one classification method is used to read the training road surface images and corresponding labels to obtain the trained SVM model.

[0018] In one embodiment, the relationship between the road surface image feature, the road surface color feature and the road surface texture feature is: F = [F C ,F T ] T ;

[0019] Among them, F C is the road surface color feature, F T It is the road surface texture feature.

[0020] In one embodiment, the road surface color feature F C Calculated using the following formula:

[0021] F C =[μ H ,μ S ,μ V ,δ H ,δ S ,δ V ,] T ;

[0022]

[0023]

[0024] Where H is hue, S is saturation, and V is brightness.

[0025] In one embodiment, the conversion formulas of hue, saturation and brightness are as follows:

[0026] C = rR + gG + bB;

[0027] v = max(R,G,B);

[0028]

[0029]

[0030] In the formula, if H<0, then let H=H+360, H∈[0,360], S, V∈[0,1].

[0031] In one embodiment, the road surface texture feature F T Calculated using the following formula:

[0032] F T =[ASM,CON,ENT,IDM] T ;

[0033] Where ASM is energy, CON is contrast, ENT is entropy, and IDM is inverse difference moment.

[0034] In one embodiment, the step of obtaining an accurate road adhesion coefficient by numerically fusing the first road adhesion coefficient value and the second road adhesion coefficient value and issuing the accurate road adhesion coefficient to the vehicle specifically includes:

[0035] comparing the first road surface adhesion coefficient value and the second road surface adhesion coefficient value;

[0036] If the difference between the first road adhesion coefficient value and the second road adhesion coefficient value exceeds a set threshold, it is determined that the current road surface has a sudden change, and the central cloud performs interpolation calculation based on the first road adhesion coefficient value and sends the calculation result to the vehicle;

[0037] If the difference between the first road adhesion coefficient value and the second road adhesion coefficient value does not exceed the set threshold, it is determined that the current road surface has not undergone a sudden change. The central cloud will perform interpolation calculations based on the second road adhesion coefficient value and send the calculation results to the vehicle.

[0038] In one embodiment, the step of obtaining an accurate road adhesion coefficient by numerically fusing the first road adhesion coefficient value and the second road adhesion coefficient value and issuing the accurate road adhesion coefficient to the vehicle is specifically performed by calculating the accurate road adhesion coefficient by the following formula:

[0039] u=u b +c(u a -u b );

[0040] Where: u is the final estimated result of the road adhesion coefficient, u a The estimated value of the road adhesion coefficient is obtained by the image method for the roadside unit, u bis the road adhesion coefficient value obtained based on the tire slip rate, c is the variable gain, which changes with (u a -u b ) changes with changes, (u a -u b ) is larger, the larger c is.

[0041] In addition, to achieve the above-mentioned purpose, the present invention also provides a road adhesion coefficient identification system based on a roadside unit and a connected vehicle train, wherein the road adhesion coefficient identification system based on a roadside unit and a connected vehicle train includes a road test edge gateway, several road test units, a central cloud and several intelligent connected vehicles, data is exchanged between several of the road test units and the road test edge gateway, and data is exchanged between the central cloud and the road test edge gateway and several of the intelligent connected vehicles.

[0042] Beneficial effects that can be achieved by the present invention: A method for identifying the road adhesion coefficient based on a roadside unit and a networked vehicle train proposed in an embodiment of the present invention is based on LoRa wireless communication and IoT data transmission of a roadside edge gateway, and has the characteristics of low cost, low power consumption, and wide coverage. The impact of the road adhesion coefficient on vehicle stability is considered, making the driving of smart networked vehicles more stable. The invention can effectively reduce the cost of large-scale deployment, save resource waste, improve the accuracy and efficiency of vehicle-road-cloud fusion perception, and ensure the safety of smart networked vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a schematic diagram of the structure of a device in a hardware operating environment involved in an embodiment of the present invention;

[0044] Figure 2 It is a flow chart of a method for identifying a road adhesion coefficient based on a roadside unit and a networked vehicle train of the present invention;

[0045] Figure 3 It is a schematic diagram of a road adhesion coefficient identification system based on a roadside unit and a networked vehicle train according to the present invention;

[0046] Figure 4 A comparison table of road surface types and road surface adhesion coefficients of the present invention;

[0047] Figure 5 For different adhesion coefficients on the road surface Relationship diagram.

[0048] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0049] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0050] In the present application, the road adhesion coefficient identification method based on the roadside unit and the connected vehicle train can be used by a terminal with a hardware operating environment.

[0051] The terminal of the embodiment of the present invention can be a PC, or it can be a smart phone, a tablet computer, an e-book reader, an MP3 (Moving Picture Experts Group Audio Layer III, dynamic image experts compression standard audio layer 3) player, an MP4 (Moving Picture Experts Group Audio Layer IV, dynamic image experts compression standard audio layer 4) player, a portable computer and other portable terminal devices with display function.

[0052] like Figure 1 As shown, the terminal may include: a processor 1001, such as a CPU, a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface). The memory 1005 may be a high-speed RAM memory, or a stable memory (non-volatile memory), such as a disk memory. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0053] Optionally, the terminal may also include a camera, an RF (Radio Frequency) circuit, a sensor, an audio circuit, a WiFi module, and the like. Among them, sensors include light sensors, motion sensors, and other sensors. Specifically, the light sensor may include an ambient light sensor and a proximity sensor, wherein the ambient light sensor may adjust the brightness of the display screen according to the brightness of the ambient light, and the proximity sensor may turn off the display screen and / or backlight when the mobile terminal is moved to the ear. As a type of motion sensor, the gravity acceleration sensor can detect the magnitude of acceleration in each direction (generally three axes), and can detect the magnitude and direction of gravity when stationary. It can be used for applications that identify the posture of the mobile terminal (such as horizontal and vertical screen switching, related games, magnetometer posture calibration), vibration recognition related functions (such as pedometer, tapping), etc.; of course, the mobile terminal may also be equipped with other sensors such as gyroscopes, barometers, hygrometers, thermometers, infrared sensors, etc., which will not be repeated here.

[0054] Those skilled in the art will understand that Figure 1 The terminal structure shown in the figure does not constitute a limitation on the terminal, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0055] Please refer to Figure 3 The road adhesion coefficient identification system based on roadside units and connected vehicle trains includes a road test edge gateway, a plurality of road test units, a central cloud and a plurality of intelligent connected vehicles. Data is exchanged between the plurality of road test units and the road test edge gateway, and data is exchanged between the central cloud and the road test edge gateway and the plurality of intelligent connected vehicles.

[0056] Please refer to Figure 2 The road adhesion coefficient identification method based on the roadside unit and the connected vehicle train includes the following steps:

[0057] Photographing road conditions by a plurality of road test unit nodes arranged on the side of the road and determining corresponding road types after distinguishing the road conditions photographed by the plurality of road test units;

[0058] Compare with a pre-stored comparison table of road surface types and road surface adhesion coefficients to obtain a range of first road surface adhesion coefficient values ​​corresponding to the road surface;

[0059] The road test edge gateway collects the road adhesion coefficient values ​​of each road section, integrates and packages them, and sends them to the central cloud;

[0060] When the intelligent network-connected vehicle is started, the real-time information of the vehicle is synchronized back to the central cloud, and the central cloud calculates the second road adhesion coefficient value of the current road surface according to the tire slip rate of the intelligent network-connected vehicle itself;

[0061] The central cloud obtains an accurate road adhesion coefficient based on the numerical fusion of the first road adhesion coefficient value and the second road adhesion coefficient value and distributes it to the vehicle.

[0062] In this embodiment, the application is based on LoRa wireless communication and IoT data transmission of roadside edge gateway, which has the characteristics of low cost, low power consumption and wide coverage. The influence of road adhesion coefficient on vehicle stability is considered to make the driving of smart networked cars more stable. The invention can effectively reduce the cost of large-scale deployment, save resource waste, improve the accuracy and efficiency of vehicle-road-cloud fusion perception, and ensure the safety of smart networked cars.

[0063] The road test unit includes but is not limited to humidity sensors, temperature sensors, cameras, ultrasonic radars and other detection devices. The road test unit node is responsible for collecting data from different sensors, using cameras to shoot road conditions (dry roads, rainy and snowy roads, etc.), and temperature and humidity sensors can sense the weather conditions (rain, snow, fog, etc.) within the current range (to determine the amount of rainfall and the possibility of road icing). After the camera shoots, the road recognition algorithm of the support vector machine is used to identify and calculate the road type of the captured image, and the road adhesion coefficient value within a certain range is predicted by looking up the table.

[0064] Ultrasonic radar identifies obstacles and measures distance, etc. The roadside unit calculates the collected information. After multiple roadside units complete the calculation, the information is exchanged between the roadside units and sent to the roadside edge gateway at the same time. Therefore, the environmental information of the entire road section can be detected to ensure the continuity of the road. The calculated road adhesion coefficient value, traffic flow, emergency situation, etc. can be sent to the vehicle that is about to arrive at this section of the road, and the vehicle can be warned of the ground conditions of the road ahead, the environmental conditions around the road section, and the weather conditions.

[0065] The roadside environment, road conditions, local weather and other warning information sent by the roadside edge gateway in the cloud, as well as the large-scale weather warning information combined with the meteorological platform, the perception information of the formation vehicles and the intelligent connected vehicles, and the vehicle surrounding environment information perceived are integrated, and the integrated information is exchanged with the intelligent connected vehicles, which can realize the optimal route for vehicle driving, coordinate traffic flow, and dispatch, manage and optimize traffic.

[0066] The self-organizing network formed between the roadside units can use Bluetooth for wireless communication, and the communication between the roadside units ensures the continuity of the road section. The roadside edge gateway is responsible for packaging the collected data into MQTT protocol data packets and communicating with the central cloud platform through the 4G / 5G mobile communication network. The central cloud integrates the roadside information, meteorological platform information, and the real-time surrounding environment information and vehicle status information perceived by the intelligent networked car. The central cloud makes vehicle decisions based on the integrated information, and sends the processed environmental information and route information to the intelligent networked cars to provide emergency warning functions, precise positioning information, real-time remote alarm and other functions.

[0067] In this application, the roadside unit is used as a fixed and reliable routing node, independently deployed on both sides of the road for large-area sensing and communication. The roadside information sensed by the roadside unit is exchanged with the roadside edge gateway, and the roadside units also interact with each other to ensure the continuity of the road section.

[0068] Based on the above embodiment, N cameras on the road side are used to shoot the road conditions. The road conditions shot by the N cameras are calculated by the road side unit to obtain the road type, and then a pre-set table is queried to obtain the range of the corresponding road adhesion coefficient value. The road side edge gateway is responsible for fusing and packaging the collected road adhesion coefficient values ​​of each road section and sending them to the central cloud. After the intelligent connected vehicle is started, the vehicle information will be interacted with the central cloud in real time. The central cloud determines whether the vehicle is the lead vehicle. If it is the lead vehicle, the lead vehicle calculates the adhesion coefficient value of the current road surface according to its own tire slip rate, and uploads the adhesion coefficient value to the central cloud. The central cloud subtracts the road adhesion coefficient value obtained by the road side camera through the image method and the adhesion coefficient value calculated by the vehicle itself, and determines whether the difference result exceeds the set threshold. If it exceeds the set threshold, it proves that the road surface has a sudden change. The central cloud uses the road adhesion coefficient value obtained by the road side camera through the image method as a reference to perform interpolation calculations, and sends the calculation results to the lead vehicle, and the lead vehicle sends them to the followers.

[0069] If the difference result does not exceed the set threshold and there is no sudden change in the road surface, the central cloud will use the road adhesion coefficient of the lead vehicle as a benchmark for interpolation calculations and send the calculation results to the lead vehicle, which then sends them to the followers. If it is not the lead vehicle, the follower can directly obtain the road adhesion coefficient of the lead vehicle. Finally, the roadside camera obtains the estimated road adhesion coefficient value through the image method and the road adhesion coefficient value obtained based on the tire slip rate to obtain an accurate road adhesion coefficient estimation strategy.

[0070] Furthermore, the step of photographing the road surface conditions by a plurality of road test unit nodes arranged beside the road and determining the road surface conditions photographed by the plurality of road test units to obtain the corresponding road surface type specifically includes:

[0071] The road test unit collects road conditions and performs image analysis on the road conditions to establish a SVM model;

[0072] The taken road surface pictures are classified according to the SVM model to obtain the results, and after statistics, the category with the largest number of occurrences is selected as the final classification result.

[0073] The roadside camera can capture the types of road surfaces within a certain range, and select the texture and color features of the road surface image to describe the road surface image information in order to realize road surface type recognition. These two types of information directly reflect the road surface type and road surface dryness and wetness, and are directly related to the road surface adhesion coefficient. The specific flow chart is as follows Figure 4 As shown, the final road surface image features are as follows:

[0074] F=[F C ,F T ] T ;

[0075] Where FC Pavement color characteristics, F T It is the road surface texture feature.

[0076] Color features are closely related to the scene and can reflect the road surface such as dryness or wetness of snow. When there is water or snow on the road surface, the reflection of light is enhanced, so the image brightness will be enhanced. The extraction of color features needs to be combined with a specific color space. The RGB model is a commonly used color representation method in computers. R, G, and B represent red, green, and blue, respectively. RGB represents a unit color vector, and r, g, and b represent the specific values ​​of color classification. According to the principle of physical three primary colors, any color Cρ can be represented by superposition of these three colors, as shown in the following formula:

[0077] C = rR + gG + bB;

[0078] The RGB color space conforms to the color display principle of display devices, but it is different from human visual perception. The human eye cannot directly obtain the red, green, and blue components of color. In order to adapt to human vision, the HSV color space was created. H, S, V represent hue, saturation, and brightness respectively. HSV and RGB are different representations of color, so they can be converted to each other. The formula is as follows

[0079] v = max(R,G,B);

[0080]

[0081]

[0082] Where if H<0, H=H+360, H∈[0,360], S,V∈[0,1]

[0083] When extracting the final features, the HSV space that is more in line with human vision was selected, and the first-order moment and second-order moment of hue, saturation, and brightness were selected to form the color feature vector, as shown in the following formula.

[0084] F C =[μ H ,μ S ,μ V, δ H ,δ S ,δ V ,] T ;

[0085]

[0086]

[0087] Image texture in a computer is the distribution law of pixel grayscale value changes. It is a specific reflection of the roughness, smoothness, granularity, etc. of the object surface. The grayscale co-occurrence matrix based on statistical methods is used to describe the image texture characteristics. A series of texture features can be extracted from the grayscale co-occurrence matrix, and energy, contrast, entropy, and inverse moment are used to form the texture features of the image.

[0088] Energy reflects the statistical result of the gray value distribution of pixel matching pairs in the image. The more concentrated the distribution, the larger the ASM value, indicating that the image texture is uniform and changes regularly. The specific calculation expression is shown in the following formula:

[0089]

[0090] Contrast reflects the statistical result of the grayscale value contrast in the image. The greater the grayscale change, the more cases of large changes in the pixel grayscale value in the image, the greater the deviation of the grayscale co-occurrence matrix from the diagonal element, the greater the contrast, indicating that the texture is more obvious. The specific calculation expression is shown in the following formula, where k is the maximum value of the grayscale range.

[0091]

[0092] Entropy reflects the amount of information an image has. The more random the image is and the more uniform the gray-level co-occurrence matrix is, the greater the entropy value is and the higher the degree of disorder of the pixels in the image is. The specific calculation expression is shown in the following formula.

[0093]

[0094] The above four quantities are selected to represent the texture features of the image, and the final result is as shown in the following formula.

[0095] F T =[ASM,CON,ENT,IDM] T

[0096] In summary, the roadside camera can collect road conditions such as road type (dry, rainy, snowy, etc.), select the texture and color features of the road image to describe the road image information, and extract the eigenvalues ​​of the road image color and texture based on the HSV color space and grayscale co-occurrence matrix through the analysis of the road image information. The road image feature vector is constructed, and then the road type recognition is realized through the support vector machine algorithm. The training road surface pictures and corresponding labels are read, and a one-to-one classification method is used for multi-classification. For different combinations of two categories, SVM feature training is performed in turn to obtain the trained SVM model. For different combinations of two categories, SVM feature training is performed in turn to obtain the trained SVM model.

[0097] Use multiple SVM models to classify the experimental road surface pictures, obtain the results and count them, and select the category with the most occurrences as the final classification result. Then, look up the table by the measured road surface type to get the corresponding road surface adhesion coefficient value. Image recognition is predictive of roads and can perceive road changes in advance. The image is based on the first road surface adhesion coefficient value and the second road surface adhesion coefficient value. The central cloud performs numerical fusion to obtain the accurate road surface adhesion coefficient and distributes it to the vehicle. The accurate road surface adhesion coefficient is calculated specifically by the following formula:

[0098] u=u b +c(u a -u b );

[0099] Where: u is the final estimated result of the road adhesion coefficient, u a The estimated value of the road adhesion coefficient is obtained by the image method for the roadside unit, u b is the road adhesion coefficient value obtained based on the tire slip rate, c is the variable gain, which changes with (u a -u b ) changes with changes, (u a -u b ) is larger, the larger c is.

[0100] Based on the vehicle dynamics model, the road adhesion coefficient is estimated by using the recursive least squares method using the relationship between tire longitudinal force and slip rate and between road adhesion coefficient and slip rate.

[0101] When a vehicle is driving on a road with different adhesion coefficients, the slip rate and the adhesion coefficient show different relationships. This characteristic can be used to identify the road adhesion coefficient.

[0102] Wheel slip rate S and utilization adhesion coefficient The definitions are as follows:

[0103]

[0104]

[0105] Using the adhesion coefficient The relationship with the tire longitudinal force is as follows:

[0106]

[0107] Different adhesion coefficients on the road Different relationships Figure 5 As shown in the figure. is the corresponding road adhesion coefficient, that is, the maximum utilization adhesion coefficient. The road identification method based on the relationship curve between road adhesion coefficient and slip rate is improved. The relationship between road adhesion coefficient and slip rate is used to realize the real-time estimation of road adhesion coefficient through the recursive least squares algorithm.

[0108] The vehicle longitudinal dynamics model is as follows:

[0109] ma x =F x -R x -mgsinβ-D A ;

[0110] Among them, F x is the vehicle traction, R x is the rolling resistance of the vehicle, D A For rolling resistance.

[0111] Among them, F x =F xf +F xr , R x =R xf +R xr ,

[0112] Estimation of road adhesion coefficient based on least squares method

[0113] According to the relationship curve between slip rate and road adhesion coefficient, the vehicle driving state can be divided into three stages according to the size of slip rate, namely linear region (slip rate meets 0≤s<0.05), transient region (slip rate meets 0.05≤s<0.3), saturation region (slip rate meets 0.3≤s<1). During normal driving of the vehicle, the wheel speed is basically equal to the vehicle speed, and the slip rate is very small, usually less than 0.05. During acceleration or braking, the relationship curve between tire slip rate and longitudinal force quickly transitions from the linear region to the saturation region, and the time spent in the transient region is very short and can be ignored. Therefore, only the road adhesion coefficient estimation problem in the linear region and saturation region can be considered.

[0114] In the linear region, the adhesion coefficient is proportional to the slip rate. In the small slip rate region, the adhesion coefficient can be expressed as follows:

[0115] when

[0116] Substituting the above formula into the vehicle longitudinal dynamics equation, the following formula is obtained:

[0117]

[0118] The basic form of recursive least squares is as follows:

[0119]

[0120] The system outputs observation information as follows:

[0121] y(t)=ma x +mgsinβ+R xf +R xr +D A ;

[0122] The system input observation information is as follows:

[0123]

[0124] Where θ(t) = K(u), e(t) is the system estimation error, and the road adhesion coefficient value is estimated by identifying the slope K(u).

[0125] Estimated road adhesion coefficient in saturated areas

[0126] The dynamic traction force equation is expressed as follows:

[0127] F x =uF z =umg cosβ;

[0128] Substituting the formula into the vehicle dynamics equation, we get the following formula:

[0129]

[0130] The basic form of recursive least squares is as follows:

[0131]

[0132] The observation information output by the system is as follows:

[0133]

[0134] System input observation information: Parameters to be changed By identifying Get the road adhesion coefficient value.

[0135] In order to track time-varying parameters and improve estimation accuracy, it is necessary to use the recursive least squares method for identification. Each time new observation data is obtained, the previous estimation result is corrected based on the newly introduced observation data according to the recursive algorithm to obtain a new parameter estimate. Compared with the batch least squares method, it has the advantages of less calculation and less computer storage.

[0136] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0137] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0138] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0139] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for identifying road adhesion coefficient based on roadside units and connected vehicle trains, characterized in that: The road adhesion coefficient identification method based on the roadside unit and the networked vehicle train comprises the following steps: Photographing road conditions by a plurality of road test unit nodes arranged on the side of the road and determining corresponding road types after distinguishing the road conditions photographed by the plurality of road test units; Compare with a pre-stored comparison table of road surface types and road surface adhesion coefficients to obtain a range of first road surface adhesion coefficient values ​​corresponding to the road surface; The road test edge gateway collects the road adhesion coefficient values ​​of each road section, integrates and packages them, and sends them to the central cloud; When the intelligent network-connected vehicle is started, the real-time information of the vehicle is synchronized back to the central cloud, and the central cloud calculates the second road adhesion coefficient value of the current road surface according to the tire slip rate of the intelligent network-connected vehicle itself; The central cloud obtains an accurate road adhesion coefficient based on the numerical fusion of the first road adhesion coefficient value and the second road adhesion coefficient value and distributes it to the vehicle.

2. The method for identifying road adhesion coefficient based on roadside unit and connected vehicle train according to claim 1, characterized in that: The step of photographing the road surface conditions by a plurality of road test unit nodes arranged on the road side and determining the road surface conditions photographed by the plurality of road test units to obtain the corresponding road surface type specifically includes: The road test unit collects road conditions and performs image analysis on the road conditions to establish a SVM model; The taken road surface pictures are classified according to the SVM model to obtain the results, and after statistics, the category with the largest number of occurrences is selected as the final classification result.

3. The method for identifying road adhesion coefficient based on roadside unit and connected vehicle train according to claim 2, characterized in that: The step of collecting road conditions by the road test unit and performing image analysis on the road conditions to establish the SVM model includes: Obtain the color feature value and image texture value of the road image extracted based on the HSV color space and gray-level co-occurrence matrix; Construct the feature vector of the road surface image and realize the road surface type recognition through the support vector machine algorithm; A one-to-one classification method is used to read the training road surface images and corresponding labels to obtain the trained SVM model.

4. The method for identifying road adhesion coefficient based on roadside unit and connected vehicle train according to claim 3, characterized in that: The relationship between the road surface image feature, the road surface color feature and the road surface texture feature is: F = [F C ,F T ] T ; Among them, F C is the road surface color feature, F T It is the road surface texture feature.

5. The method for identifying road adhesion coefficient based on roadside unit and connected vehicle train according to claim 4, characterized in that: The road surface color feature F C Calculated using the following formula: F C =[μ H ,m S ,m V ,d H ,d S ,d V ,] T ; X=H,S,V; Where H is hue, S is saturation, and V is brightness.

6. The method for identifying road adhesion coefficient based on roadside unit and connected vehicle train according to claim 5, characterized in that: The conversion formulas of hue, saturation and brightness are as follows: C = rR + gG + bB; v = max(R,G,B); In the formula, if H<0, then let H=H+360, H∈[0,360], S, V∈[0,1].

7. The method for identifying road adhesion coefficient based on roadside unit and connected vehicle train according to claim 3, characterized in that: The road surface texture feature F T Calculated using the following formula: F T =[ASM,CON,ENT,IDM] T ; Where ASM is energy, CON is contrast, ENT is entropy, and IDM is inverse difference moment.

8. The method for identifying road adhesion coefficient based on roadside unit and connected vehicle train according to claim 1, characterized in that: The step of obtaining an accurate road adhesion coefficient by numerically fusing the first road adhesion coefficient value and the second road adhesion coefficient value and issuing the accurate road adhesion coefficient to the vehicle specifically includes: comparing the first road surface adhesion coefficient value and the second road surface adhesion coefficient value; If the difference between the first road adhesion coefficient value and the second road adhesion coefficient value exceeds a set threshold, it is determined that the current road surface has a sudden change, and the central cloud performs interpolation calculation based on the first road adhesion coefficient value and sends the calculation result to the vehicle; If the difference between the first road adhesion coefficient value and the second road adhesion coefficient value does not exceed the set threshold, it is determined that the current road surface has not undergone a sudden change. The central cloud will perform interpolation calculations based on the second road adhesion coefficient value and send the calculation results to the vehicle.

9. The method for identifying road adhesion coefficient based on roadside unit and connected vehicle train according to claim 1, characterized in that: The step of obtaining an accurate road adhesion coefficient by numerically fusing the first road adhesion coefficient value and the second road adhesion coefficient value and issuing the accurate road adhesion coefficient to the vehicle is specifically calculated by the following formula: in=in b +c(in a -in b ); Where: u is the final estimated result of the road adhesion coefficient, u a The estimated value of the road adhesion coefficient is obtained by the image method for the roadside unit, u b is the road adhesion coefficient value obtained based on the tire slip rate, c is the variable gain, which changes with (u a -u b ) changes with changes, (u a -u b ) is larger, the larger c is.

10. A road adhesion coefficient identification system based on roadside units and connected vehicle trains, characterized in that: The road adhesion coefficient identification system based on roadside units and connected vehicle trains includes a road test edge gateway, several road test units, a central cloud and several intelligent connected vehicles. Data is exchanged between the several road test units and the road test edge gateway, and data is exchanged between the central cloud and the road test edge gateway and the several intelligent connected vehicles.

Images