Mining area pavement maintenance decision-making method and system based on pavement state estimation and storage medium

Abstract

The invention discloses a mining area road surface maintenance decision method and system based on road surface state estimation and a storage medium, and belongs to the technical field of mining area road surface detection.The method comprises the steps that road surface state real-time relevant parameters collected when a vehicle runs in a current area of a mining area road surface are obtained, and the real-time relevant parameters are preprocessed; performing feature extraction on the preprocessed parameters to obtain a related feature set; feature learning is carried out on the related feature set according to historical samples, a road surface state estimation value of the current area is obtained, the historical samples comprise historical related features and road surface state values corresponding to the historical related features, and the road surface state values represent the road surface bumping degree; optimizing the road surface state estimation value according to the historical state of the current area; and according to the optimized road surface state estimation value, a corresponding vehicle passing and road surface maintenance scheduling instruction is made, so that the unmanned driving potential safety hazard of the mining area is reduced, and the unmanned driving working efficiency and stability are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of mining area road surface detection, and in particular relates to a mining area road surface maintenance decision method, system and storage medium based on road surface state estimation. Background Art

[0002] The definition of bumpy road surface in mining scenarios refers to the unevenness, potholes, cracks and gravel on the road surface to varying degrees due to long-term mining operations, crushing by mining cars, falling slag, rain erosion and other factors. As a result, the center of gravity of the vehicle fluctuates continuously during driving, and the direction and size of the resistance it encounters constantly changes, thus affecting the vehicle's controllability and system stability.

[0003] Although the unmanned driving roads in the mining area are designed as structured roads, the long-term heavy load and the inherently unfavorable natural conditions often make the conditions complex and changeable. Potholes, falling rocks during transportation, and rutting caused by vehicles pose a severe challenge to the vehicle's power performance, system stability, and even overall safety. Therefore, accurate judgment and immediate maintenance of bumpy roads have become the key to ensuring the stable operation of the unmanned driving system. Summary of the invention

[0004] The present invention provides a mining area road maintenance decision-making method, system and storage medium based on road surface state estimation. The estimated value of the bumpiness of a road section is obtained through feature extraction and feature learning, and the estimated value is optimized using the historical bumpiness information of the road section, thereby achieving an accurate estimation of the state of the road section, and then taking corresponding measures according to the road surface state, thereby reducing safety hazards, keeping the unmanned operation road surface at an ideal level, and reducing operation interruptions.

[0005] To achieve the above object, the present invention is implemented by adopting the following technical solutions:

[0006] In a first aspect, the present invention provides a mining area road maintenance decision-making method based on road condition estimation, comprising:

[0007] Acquire the real-time relevant parameters of the road surface state collected by the vehicle driving on the current area of ​​the mining area road surface, and pre-process the real-time relevant parameters;

[0008] Perform feature extraction on the preprocessed parameters to obtain relevant feature sets;

[0009] Performing feature learning on the relevant feature set according to historical samples to obtain a road surface condition estimation value of the current area, wherein the historical samples include historical relevant features and their corresponding road surface condition values, and the road surface condition values ​​represent the degree of road surface bumps;

[0010] Optimizing the estimated value of the road surface condition according to the historical state of the current area;

[0011] Make corresponding dispatch instructions for vehicle traffic and road maintenance based on the optimized road condition estimation value.

[0012] Optionally, the real-time relevant parameters of the road surface condition collected by the vehicle in the current area of ​​the mining road surface are obtained, including: collecting point cloud data, IMU data and wheel speed data respectively during the vehicle driving process through a laser radar, an IMU inertial measurement unit and a wheel speed sensor installed on the vehicle.

[0013] Optionally, the preprocessing of the real-time related parameters includes:

[0014] Perform spatial processing on the point cloud data of the current area collected by the laser radar to obtain a ground point cloud set of the current area;

[0015] The inertial data and wheel speed data of the vehicle traveling in the current area are respectively processed in time series to obtain time series IMU data and time series wheel speed data.

[0016] Optionally, extracting features from the preprocessed parameters to obtain a relevant feature set includes:

[0017] Use pointpillars to encode the ground point cloud set to obtain the point cloud grid map of the current area, and calculate the features of the point cloud inside each grid to obtain the discrete degree of road height, the convexity of the local road surface and the local geometric information of the point cloud, and generate the road surface features according to the calculated point cloud features ,in is the range of the point cloud, and C is the feature of the point cloud;

[0018] Perform PCA principal component analysis on the IMU data at each moment in the time series IMU data, use the eigenvector of the principal component as the IMU data feature, and encode the IMU data feature according to the time dimension to generate an IMU feature matrix;

[0019] Calculate the standard deviation of the wheel speed and the first-order derivative or difference of the wheel speed of each wheel during the driving time in the current area according to the time-series wheel speed data, obtain the wheel rotation stability and the wheel speed change rate, and encode the wheel rotation stability and the wheel speed change rate according to the time dimension to generate a wheel speed feature matrix;

[0020] The road surface characteristics, IMU feature matrix and wheel speed feature matrix of the current area are combined into a relevant feature set of the road surface state of the current area.

[0021] Optionally, performing feature learning on the relevant feature set according to historical samples to obtain a road surface state estimation value of the current area includes:

[0022] Inputting the relevant feature set into a plurality of pre-trained road surface condition prediction models to obtain a plurality of road surface condition prediction values;

[0023] averaging the plurality of road surface condition prediction values ​​to obtain a road surface condition estimation value;

[0024] The method for acquiring the plurality of road surface condition prediction models includes:

[0025] The pre-constructed decision trees are trained respectively through the historical samples in the pre-constructed historical sample set to obtain a plurality of road surface condition prediction models, wherein the road surface condition values ​​in the historical samples are in percentage form.

[0026] Optionally, during the construction of the historical sample set, the historical data related to the road surface status of each area of ​​the mining area road surface are sampled with replacement;

[0027] During the construction of each decision tree, some relevant features are randomly selected to perform optimal division on the nodes.

[0028] Optionally, optimizing the estimated value of the road surface state according to the historical state of the current area includes:

[0029] Get the influence weight of each historical state of the current region on the current state. The calculation formula of the influence weight is:

[0030] ,

[0031] in, is the weight factor corresponding to time t, is the attenuation rate, , is the current moment, n is the number of historical moments, ;

[0032] The influence weight is normalized and the calculation formula is as follows:

[0033] ,

[0034] in, is the normalized weight factor;

[0035] The road surface condition estimation value is adjusted by the normalized weight factor to obtain an optimized road surface condition estimation value, and the calculation formula is as follows:

[0036] ,

[0037] Among them, IP represents the initial road surface state estimation value, AP represents the optimized road surface state estimation value, To adjust the parameters used to control the impact of historical data on the initial road condition estimate, Represents the estimated value of the road surface condition corresponding to time t.

[0038] Optionally, making corresponding vehicle traffic and road maintenance scheduling instructions according to the optimized road condition estimation value includes:

[0039] The optimized road surface condition estimation value is compared with a preset bumpiness threshold value. and For comparison, if the degree of turbulence , indicating that the road surface is flat and does not require maintenance for the time being. , indicating that the road surface is not bumpy, a decision signal is sent to the unmanned mine truck to slow down the current area, and the idle grader on site is dispatched to perform ground maintenance work in the current area after the unmanned mine truck passes. , indicating that the road surface is bumpy and undulating, a decision signal is sent to the unmanned mining truck to bypass the current area, and the grader is dispatched to repair the road condition in time.

[0040] In a second aspect, the present invention provides a mining area road maintenance decision system based on road condition estimation, comprising:

[0041] Parameter preprocessing module: used to obtain the real-time relevant parameters of the road surface state collected by the vehicle when driving on the current area of ​​the mining area road surface, and preprocess the real-time relevant parameters;

[0042] Feature extraction module: used to extract features from preprocessed parameters to obtain relevant feature sets;

[0043] A road surface state estimation module is used to perform feature learning on the relevant feature set according to historical samples to obtain a road surface state estimation value of the current area, wherein the historical samples include historical relevant features and their corresponding road surface state values, and the road surface state value represents the degree of road bumps;

[0044] An estimation value optimization module: used for optimizing the road surface state estimation value according to the historical state of the current area;

[0045] Pavement maintenance decision module: used to make corresponding vehicle traffic and pavement maintenance scheduling instructions based on the optimized pavement condition estimation value.

[0046] In a third aspect, the present invention provides a computer storage medium having a computer program stored thereon, which, when executed by a processor, implements the mining area road maintenance decision-making method based on road surface condition estimation as described in any step of the first aspect.

[0047] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention reads the bumpiness-related data collected by the vehicle driving on a road section in the mining area in real time, extracts features and learns the bumpiness-related data according to historical conditions, estimates the bumpiness of the current road section, optimizes the estimated value by adding historical conditions, and improves the accuracy of road condition prediction by taking into account the influence of the road surface historical conditions on the current conditions. Finally, corresponding traffic decisions and road maintenance measures are taken for different road conditions, and real-time detection of road conditions is achieved, so that the unmanned driving operation road surface is maintained at an ideal level, operation interruptions are reduced, safety hazards of unmanned driving are reduced, and the stability and operation efficiency of the unmanned mining truck system are improved; multi-sensor data is used to detect ground conditions, ground spatial features are obtained by laser radar, the vehicle's own state is obtained by IMU and wheel speed sensor, and road condition judgment is performed by integrating the features of each sensor, so that the road bumpiness characteristics can be more effectively represented; by quantifying the road condition, the bumpiness state of the road surface is represented in percentage form, and the bumpiness of the road surface can be accurately predicted by decision tree regression, providing more detailed data support for subsequent decision-making processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 The figure is a flow chart of a mining area road maintenance decision method based on road surface state estimation in embodiment 1 of the present invention;

[0049] Figure 2 The figure is a flow chart of a mining area road maintenance decision method based on road surface state estimation in embodiment 2 of the present invention;

[0050] Figure 3 The figure shows the architecture diagram of the mining area road maintenance decision system based on road surface state estimation in Example 3 of the present invention. DETAILED DESCRIPTION

[0051] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0052] Example 1

[0053] like Figure 1 As shown, this embodiment provides a mining area road maintenance decision method based on road surface state estimation, including:

[0054] Acquire the real-time relevant parameters of the road surface state collected by the vehicle driving on the current area of ​​the mining area road surface, and pre-process the real-time relevant parameters;

[0055] Perform feature extraction on the preprocessed parameters to obtain relevant feature sets;

[0056] Performing feature learning on the relevant feature set according to historical samples to obtain a road surface condition estimation value of the current area, wherein the historical samples include historical relevant features and their corresponding road surface condition values, and the road surface condition values ​​represent the degree of road surface bumps;

[0057] Optimizing the estimated value of the road surface condition according to the historical state of the current area;

[0058] Make corresponding dispatch instructions for vehicle traffic and road maintenance based on the optimized road condition estimation value.

[0059] The present invention detects the road bumpiness caused by multiple driving in a fixed operation area during unmanned operation in a mining area. It extracts features from road surface state-related data collected when a vehicle is driving on a road section, learns the features, regresses the bumpiness state, and optimizes the road surface state prediction value using historical bumpiness information on the same road section. This achieves accurate estimation of the same road surface state, and takes different measures for different road surface states, thereby reducing safety hazards, enhancing system stability, and improving operation efficiency.

[0060] Example 2

[0061] like Figure 2 As shown, based on Example 1, this example also makes the following design.

[0062] The road surface condition detection refers to detecting the road surface condition of each unit area, that is, the bumpiness parameter of each unit area. The road surface condition parameter value is a percentage from 0 to 100. If the road surface condition is completely flat and there is no bumpiness, the road surface condition value is 0%, which also means that the bumpiness of the road surface is 0%. If the road surface condition is extremely poor, the road surface is uneven, and there are a large number of ruts and fallen rocks, the road surface condition value is 100%, which also means that the bumpiness of the road surface is 100%.

[0063] The mining area road maintenance decision-making method based on road surface state estimation is specifically divided into the following steps.

[0064] Step 1: Read sensor data collected by sensors installed on the vehicle during driving, and pre-process the sensor data, including lidar point cloud data, IMU data, and wheel speed sensor data.

[0065] For the lidar data, point cloud straight-through filtering, radius filtering and voxel filtering are performed to obtain the ideal point cloud data of the target area to remove noise. Then, the CSF cloth simulation filtering algorithm is used to perform ground segmentation on the ideal point cloud, and the original point cloud is divided into two parts: ground point cloud and non-ground point cloud to obtain the ground point cloud.

[0066] After segmenting the ground points, collect ground point clouds over a continuous period of time when the vehicle is traveling in a certain area. ,The transformation relationship between each frame of the ground point cloud is obtained through ICP point cloud registration, and then the point cloud is stitched to obtain a continuous ground point cloud set in this area.

[0067] For IMU data, including three-axis angular velocity and three-axis acceleration, the three-axis acceleration is collected during driving in a certain area. , where N represents the number of time points at which vehicles travel in this area, , f represents the data acquisition frequency of IMU, represents the i-th moment when the vehicle travels in this area, represents the x-axis acceleration, represents the acceleration on the y-axis, Represents the z-axis acceleration, forming an IMU data matrix based on time.

[0068] For the wheel speed sensor data, the rotation speeds of the four wheels are read respectively, and a wheel speed data matrix is ​​generated based on time. The first dimension of the matrix is ​​the rotation speed of the four wheels at a certain moment, and the second dimension is the number of times the data is collected at a frequency f within the time t of passing through the area.

[0069] Step 2: Perform feature extraction on the sensor data matrix provided after data preprocessing to obtain features related to bumps in the sensor data of vehicles in a certain area.

[0070] For ground point clouds, PointPillars is used for encoding, and a plane grid is established according to the area size. The ground area of ​​the point cloud is , set the grid size to , forming a wide , Gao Wei The ground point cloud is projected into the corresponding grid, and the features of the point cloud inside each grid are extracted.

[0071] For each grid, the features of the internal point cloud are calculated, mainly including: counting the z coordinates of the point cloud in the grid, calculating the standard deviation of the point cloud in the z direction, indicating the discrete degree of the ground height, calculating the angle between the point cloud normal vector and the z direction, indicating the degree of convexity of the local road surface, calculating the FPFH feature descriptor of the point cloud, and describing the local geometric information of the ground point cloud.

[0072] The three features of the above point cloud are aggregated to form an expression of the road surface characteristics, which is expressed as ,in, is the range of the point cloud, and C is the feature of the point cloud.

[0073] For the IMU inertial measurement unit data, PCA principal component analysis is performed on the IMU data at each moment. First, the covariance matrix is ​​constructed, and then the eigenvalues ​​and eigenvectors of the covariance matrix are calculated. The eigenvectors of the principal components are used as the features of the IMU data. The IMU features are encoded according to the time dimension to form the feature matrix of the IMU.

[0074] For the wheel speed sensor, the standard deviation of the wheel speed of each wheel during the driving time in a certain area is calculated to indicate the stability of the wheel rotation. The first-order derivative or differential of the wheel speed is calculated to reflect the speed change rate of the wheel. A characteristic matrix with time as the dimension is also formed for the stability and speed change rate of the wheel rotation.

[0075] Step 3: Perform feature learning on the acquired sensor features and output the estimated value of the road state.

[0076] The random forest regression method is used to regress the features. First, multiple decision trees are constructed. Multiple sample sets are generated from the original training data through replacement sampling. Each sample set is used to train a decision tree. At the same time, when constructing the nodes of each tree, only a part of the features are randomly selected from all the features for optimal division.

[0077] Use the trained decision tree to predict the road state value of the acquired sensor features. Each decision tree will give a prediction value based on the input feature vector (features of IMU data, lidar data, and wheel speed sensor data), that is, an estimate of the degree of road bumpiness. The results of all decision trees are averaged to obtain the final regression value.

[0078] Step 4: Optimize the predicted value based on the historical status of the current area.

[0079] Since the road surface state is correlated with the road surface state at this time in history, the predicted road surface state is optimized by combining the historical state. For the degree of bumpiness at the historical time, a weight factor is introduced to indicate its influence on the current predicted value. The exponential decay weight factor is used to indicate the influence of the state of this area in the past on the current state.

[0080] set up For time The corresponding weight factor is, is the decay rate ( ), since the road surface changes little over time during the unmanned driving process, λ is set to 0.8, is the current time point, and n is the number of historical time points. Then for each historical time point t (where ), weight factor It can be calculated as: .

[0081] Then the weight factor is normalized so that the sum of the weights of the historical data points is 1. The normalized weight factor for: .

[0082] The current road condition prediction value obtained by random forest regression is IP as the initial prediction value, and the prediction value is adjusted using the weight factor. The adjusted prediction value is AP:

[0083] ,

[0084] Among them, AP is the optimized road state value, and IP is the state value regressed by random forest based on the current ground features. is the road surface state value at historical time t, is an additional tuning parameter that controls the degree to which historical data influences the initial forecast value.

[0085] Step 5: Make a decision based on the road surface condition value of the detection area to determine whether the area needs maintenance.

[0086] Setting Thresholds , If the degree of turbulence , it means the ground is flat and does not require maintenance. , it means that the road surface is somewhat bumpy. The remote dispatch center sends a decision signal to the unmanned mining truck to slow down when entering this area. If there is an idle grader on site, the grader will be dispatched to perform ground maintenance work after the mining truck passes through this area. , it means that the road is too bumpy and impassable. At this time, the vehicle chooses to drive around the obstacle, and the remote dispatch center needs to dispatch a grader to repair the road condition. When the grader is finished, the road condition and maintenance request of this area are initialized.

[0087] Example 3

[0088] like Figure 3 As shown, this embodiment provides a mining area road maintenance decision system based on road surface state estimation, including:

[0089] Parameter preprocessing module: used to obtain the real-time relevant parameters of the road surface state collected by the vehicle when driving on the current area of ​​the mining area road surface, and preprocess the real-time relevant parameters;

[0090] Feature extraction module: used to extract features from preprocessed parameters to obtain relevant feature sets;

[0091] A road surface state estimation module is used to perform feature learning on the relevant feature set according to historical samples to obtain a road surface state estimation value of the current area, wherein the historical samples include historical relevant features and their corresponding road surface state values, and the road surface state value represents the degree of road bumps;

[0092] An estimation value optimization module: used for optimizing the road surface state estimation value according to the historical state of the current area;

[0093] Pavement maintenance decision module: used to make corresponding vehicle traffic and pavement maintenance scheduling instructions based on the optimized pavement condition estimation value.

[0094] Example 4

[0095] This embodiment provides a computer storage medium on which a computer program is stored. When the computer program is executed by a processor, the mining area road maintenance decision-making method based on road surface state estimation as described in any step of Example 2 is implemented.

[0096] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0097] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0098] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0099] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0100] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which all fall within the protection of the present invention.

Claims

1. A mining area pavement maintenance decision-making method based on pavement state estimation, characterized in that: include: Acquire the real-time relevant parameters of the road surface state collected by the vehicle driving on the current area of ​​the mining area road surface, and pre-process the real-time relevant parameters; Perform feature extraction on the preprocessed parameters to obtain relevant feature sets; Performing feature learning on the relevant feature set according to historical samples to obtain a road surface condition estimation value of the current area, wherein the historical samples include historical relevant features and their corresponding road surface condition values, and the road surface condition values ​​represent the degree of road surface bumps; Optimizing the estimated value of the road surface condition according to the historical state of the current area; Make corresponding dispatch instructions for vehicle traffic and road maintenance based on the optimized road condition estimation value.

2. The mining area road maintenance decision-making method based on road condition estimation according to claim 1 is characterized in that: The real-time relevant parameters of the road surface state collected by the vehicle in the current area of ​​the mining area road are obtained, including: collecting point cloud data, IMU data and wheel speed data respectively during the vehicle driving process through the laser radar, IMU inertial measurement unit and wheel speed sensor installed on the vehicle.

3. The mining area road maintenance decision-making method based on road condition estimation according to claim 2 is characterized in that: The preprocessing of the real-time related parameters comprises: Perform spatial processing on the point cloud data of the current area collected by the laser radar to obtain a ground point cloud set of the current area; The inertial data and wheel speed data of the vehicle traveling in the current area are respectively processed in time series to obtain time series IMU data and time series wheel speed data.

4. The mining area road maintenance decision-making method based on road condition estimation according to claim 3 is characterized in that: The feature extraction of the preprocessed parameters to obtain a relevant feature set includes: Use pointpillars to encode the ground point cloud set to obtain the point cloud grid map of the current area, and calculate the features of the point cloud inside each grid to obtain the discrete degree of road height, the convexity of the local road surface and the local geometric information of the point cloud, and generate the road surface features according to the calculated point cloud features ,in is the range of the point cloud, and C is the feature of the point cloud; Perform PCA principal component analysis on the IMU data at each moment in the time series IMU data, use the eigenvector of the principal component as the IMU data feature, and encode the IMU data feature according to the time dimension to generate an IMU feature matrix; Calculate the standard deviation of the wheel speed and the first-order derivative or difference of the wheel speed of each wheel during the driving time in the current area according to the time-series wheel speed data, obtain the wheel rotation stability and the wheel speed change rate, and encode the wheel rotation stability and the wheel speed change rate according to the time dimension to generate a wheel speed feature matrix; The road surface characteristics, IMU feature matrix and wheel speed feature matrix of the current area are combined into a relevant feature set of the road surface state of the current area.

5. The mining area road maintenance decision-making method based on road condition estimation according to claim 1 is characterized in that: The performing feature learning on the relevant feature set according to the historical samples to obtain the estimated value of the road surface state in the current area includes: Inputting the relevant feature set into a plurality of pre-trained road surface condition prediction models to obtain a plurality of road surface condition prediction values; averaging the plurality of road surface condition prediction values ​​to obtain a road surface condition estimation value; The method for acquiring the plurality of road surface condition prediction models includes: By training each pre-constructed decision tree with each historical sample in a pre-constructed historical sample set, a plurality of pavement condition prediction models are obtained, wherein the pavement condition value in the historical sample is in the form of percentage, wherein the pavement condition value in the historical sample is in the form of percentage.

6. The mining area road maintenance decision-making method based on road condition estimation according to claim 5 is characterized in that: During the construction of the historical sample set, the historical data related to the road surface status of each area of ​​the mining area road surface are sampled with replacement; During the construction of each decision tree, some relevant features are randomly selected to perform optimal division on the nodes.

7. The mining area road maintenance decision-making method based on road condition estimation according to claim 1 is characterized in that: The optimizing the estimated value of the road surface state according to the historical state of the current area includes: Get the influence weight of each historical state of the current region on the current state. The calculation formula of the influence weight is: ， in, is the weight factor corresponding to time t, is the attenuation rate, , is the current moment, n is the number of historical moments, ; The influence weight is normalized and the calculation formula is as follows: ， in, is the normalized weight factor; The road surface condition estimation value is adjusted by the normalized weight factor to obtain an optimized road surface condition estimation value, and the calculation formula is as follows: ， Among them, IP represents the initial road surface state estimation value, AP represents the optimized road surface state estimation value, To adjust the parameters used to control the impact of historical data on the initial road condition estimate, Represents the estimated value of the road surface condition corresponding to time t.

8. The mining area road maintenance decision-making method based on road condition estimation according to claim 1 is characterized in that: The step of making corresponding vehicle traffic and road maintenance scheduling instructions according to the optimized road condition estimation value includes: The optimized road surface condition estimation value is compared with a preset bumpiness threshold value. and For comparison, if the degree of turbulence , indicating that the road surface is flat and does not require maintenance. , indicating that the road surface is not bumpy, a decision signal is sent to the unmanned mine truck to slow down the current area, and the idle grader on site is dispatched to perform ground maintenance work in the current area after the unmanned mine truck passes. , indicating that the road surface is bumpy and undulating, a decision signal is sent to the unmanned mining truck to bypass the current area, and the grader is dispatched to repair the road condition in time.

9. A mining area road maintenance decision system based on road condition estimation, characterized in that: include: Parameter preprocessing module: used to obtain the real-time relevant parameters of the road surface state collected when the vehicle is driving on the current area of ​​the mining area road surface, and preprocess the real-time relevant parameters; Feature extraction module: used to extract features from preprocessed parameters to obtain relevant feature sets; A road surface state estimation module is used to perform feature learning on the relevant feature set according to historical samples to obtain a road surface state estimation value of the current area, wherein the historical samples include historical relevant features and their corresponding road surface state values, and the road surface state value represents the degree of road bumps; An estimation value optimization module: used for optimizing the road surface state estimation value according to the historical state of the current area; Pavement maintenance decision module: used to make corresponding vehicle traffic and pavement maintenance scheduling instructions based on the optimized pavement condition estimation value.

10. A computer storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the mining area road maintenance decision-making method based on road surface state estimation as described in any one of claims 1 to 8 is implemented.

Images