Monitoring method for multi-source information fusion of strip mine transportation equipment based on automatic driving

Abstract

The invention discloses a strip mine transportation equipment multi-source information fusion monitoring method based on automatic driving, and the method comprises the following steps: 1, collecting a position signal and a vehicle speed signal of transportation equipment and a rotation angle of a steering wheel according to a sampling period; 2, performing data preprocessing on the to-be-fused data of the multi-source information; 3, constructing a support vector machine regression model; 4, establishing a particle swarm optimization algorithm model, and carrying out optimal solution solving on a penalty coefficient and a kernel function parameter of the support vector machine regression model to obtain an optimal parameter combination; and step five, the optimal parameter combination is substituted into the support vector machine regression model, so that the monitoring result is more accurate, and the travel route monitoring result of the transportation equipment is obtained. The method has the characteristics that the operation time is greatly shortened, the monitoring precision can be improved, and the accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of transportation equipment data fusion, and more particularly, to a monitoring method for multi-source information fusion of open-pit mine transportation equipment based on automatic driving. Background Art

[0003] Iron ore is one of the most widely used metals in our daily lives. It's used in many applications, from cars to knives. Because of its importance, the question of how to extract iron ore resources has become increasingly prominent. Iron ore mining involves carefully researching, mapping, and identifying the minerals needed to obtain them.

[0004] The diversity of iron ore production environment conditions poses huge challenges to iron ore production and harvesting. Therefore, with the development of science and technology, higher requirements are placed on iron ore production methods, and intelligent and mechanized iron ore has gradually replaced the traditional model.

[0005] Transport equipment on the path of autonomous driving will inevitably be affected by a series of factors such as the external environment, road conditions, load changes, its own vibration, and signal reception. Automated transport equipment generally uses multiple sensors to monitor its necessary influencing factors and effectively obtain data information at different time nodes and spatial dimensions. This will inevitably lead to the problem of multi-source information fusion during transmission and decision-making.

[0006] In the existing multi-source information fusion research, some use hierarchical structure and genetic neural network algorithm for processing, and some use grey least squares support vector machine algorithm for processing. However, these two fusion algorithms cannot achieve the expected accuracy, resulting in insufficient monitoring accuracy. In addition, due to the large amount of data, the calculation time is too long and the real-time performance is not enough. Summary of the Invention

[0007] The purpose of this invention is to design and develop a monitoring method for open-pit mine transportation equipment based on automatic driving and multi-source information fusion. Through the adaptive particle swarm optimization algorithm combined with the support vector machine regression model, the route of the transportation equipment can be accurately monitored, thereby avoiding the deviation of the transportation equipment route and improving the monitoring accuracy.

[0008] The technical solution provided by the present invention is:

[0009] A monitoring method for open-pit mine transportation equipment based on multi-source information fusion for autonomous driving includes the following steps:

[0010] Step 1: Collect the position signal, speed signal and steering wheel rotation angle of the transportation equipment according to the sampling period;

[0011] Step 2: Preprocess the data to be fused from multiple sources;

[0012] Step 3: Build a support vector machine regression model;

[0013] Step 4: Establish a particle swarm optimization algorithm model, solve the penalty coefficient and kernel function parameters of the support vector machine regression model for the optimal solution, and obtain the optimal parameter combination;

[0014] Step 5: Bring the optimal parameter combination into the support vector machine regression model to obtain the prediction result of the route of the transportation equipment.

[0015] Preferably, the multi-source information data to be fused includes:

[0016] Initial position signal, predicted position signal at the previous moment, real-time position signal at the previous moment, average vehicle speed signal, predicted vehicle speed signal at the previous moment, real-time vehicle speed signal at the previous moment, real-time maximum steering wheel rotation angle, predicted steering wheel rotation angle at the previous moment, and real-time steering wheel rotation angle at the previous moment.

[0017] Preferably, the position signal, vehicle speed signal and steering wheel rotation angle of the transportation equipment are all monitored by sensors.

[0018] Preferably, the data preprocessing includes removing abnormal values ​​and processing default values.

[0019] Preferably, the removal of outliers specifically includes the following steps:

[0020] Step 1: Arrange the data collected by the sensor in ascending order;

[0021] Step 2: Obtain high-end statistics and low-end statistics based on the sample size;

[0022] Step 3: When the confidence value At different times, the number of samples corresponds to different critical values :

[0023] when , ,but For outliers, remove ;when , ,but For outliers, remove , otherwise there are no outliers;

[0024] Where, is a high-end statistic, is a low-end statistic, and .

[0025] Preferably, the default value processing specifically includes:

[0026] At known coordinates and Coordinates to interpolate between satisfy:

[0027] .

[0028] Preferably, the data preprocessing further includes normalizing the data to be fused from the multi-source information:

[0029] ;

[0030] Where, is the data output after normalization processing, is the original data, For variables The minimum value of For variables The maximum value of .

[0031] Preferably, the support vector machine regression model satisfies:

[0032] ;

[0033] ;

[0034] ;

[0035] Where, is the kernel function, corresponding to and The inner product mapped to the feature space is used to replace the complex operations in the feature space. is the kernel function parameter, σ is the standard deviation of the kernel function, also known as the scale parameter, and b is the bias term, also known as the intercept term.

[0036] Preferably, the particle swarm optimization algorithm model includes:

[0037] The speed and position updates of the particle swarm optimization algorithm model satisfy:

[0038] ;

[0039] ;

[0040] Where, The number of iterations is Time Particles in The speed in dimensional space, is the inertia weight, The number of iterations is Time Particles in The speed in dimensional space, is the first acceleration factor, To be distributed in Random numbers in the interval, The number of iterations is Time Particles in The optimal solution obtained in the dimensional space, The number of iterations is Time Particles in The position vector in the dimensional space, is the second acceleration factor, To be distributed in Random numbers in the interval, The number of iterations is When the group The optimal solution in dimensional space, The number of iterations is Time Particles in Position vector in dimensional space;

[0041] Among them, the inertia weight satisfies:

[0042]

[0043] Where, is the minimum weight, is the maximum weight, is the fitness value of the particle at any time, is the minimum fitness value in the particle swarm, is the average fitness value of the particle swarm;

[0044] The first acceleration factor and the second acceleration factor satisfy:

[0045] ;

[0046] Where, 、 They are and The initial value of 、 They are and The ending value of .

[0047] Preferably, the parameters of the particle swarm optimization algorithm model are initialized as:

[0048] The population size is 50, the number of iterations is 1000, the initial value of the inertia weight is 0.6, and the initial values ​​of the first acceleration factor and the second acceleration factor are both 2.

[0049] The beneficial effects of the present invention are:

[0050] The present invention designs and develops a multi-source information fusion route monitoring method for open-pit mine transportation equipment based on automatic driving. Through the combination of an adaptive particle swarm optimization algorithm and a support vector machine regression model, the calculation process is simplified, the calculation time is greatly shortened, and at the same time the monitoring precision and accuracy are improved. The route of the transportation equipment can be determined in advance, and then whether the route of the transportation equipment meets expectations can be judged, thereby achieving the purpose of revising the route in advance and optimizing the automatic driving of the transportation equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 The figure is a flow chart of the route monitoring method of multi-source information fusion for open-pit mine transportation equipment based on autonomous driving according to the present invention. DETAILED DESCRIPTION

[0054] The present invention is described in further detail below so that those skilled in the art can implement the invention with reference to the description.

[0055] like Figure 1 As shown, the present invention provides a multi-source information fusion route monitoring method for open-pit mine transportation equipment based on automatic driving, which includes the following steps:

[0056] Step 1: Collect the position signal, speed signal and steering wheel rotation angle of the transport equipment according to the sampling period, wherein the position signal, speed signal and steering wheel rotation angle are the data to be fused of the multi-source information;

[0057] Among them, the data to be fused of the multi-source signals specifically include: initial position signal, predicted position signal at the previous moment, real-time position signal at the previous moment, average vehicle speed signal, predicted vehicle speed signal at the previous moment, real-time vehicle speed signal at the previous moment, real-time maximum steering wheel rotation angle, predicted steering wheel rotation angle at the previous moment and real-time steering wheel rotation angle at the previous moment.

[0058] Step 2: preprocessing the data to be fused from the multi-source information;

[0059] Wherein, the data preprocessing includes: elimination of abnormal values ​​and default value processing;

[0060] The removal of outliers includes:

[0061] The data collected by the sensor are arranged from small to large as follows: , different sample sizes lead to different statistics, as follows:

[0062] exist hour, , ;

[0063] exist hour, , ;

[0064] exist hour, , ;

[0065] exist hour, , ;

[0066] Where, is the sample size, is the first high-end statistic, is the first low-end statistic, is the second highest statistic, is the second lowest statistic, is the third high-end statistic, It is the third lowest statistic. It is the fourth high-end statistic, is the fourth lowest statistic, which is convenient for summarization. 、 、 、 Unified summary , 、 、 and Unified summary ,and and correspond to each other;

[0067] When the confidence value At different times, the number of samples corresponds to different critical values ,when , ,but For outliers, remove ;when , ,but For outliers, remove , otherwise there is no outlier; critical value The specific values ​​of are shown in Table 1.

[0068] Table 1 Critical value

[0069]

[0070] The default value processing includes:

[0071] In this embodiment, the default value is processed by linear interpolation. The current interpolation value can be obtained based on the data before and after the interpolation point. Specifically, the coordinates are known. and According to the following formula, we can get Internal insertion Values ​​and their corresponding :

[0072] ;

[0073] Where, The value is known and can be directly obtained according to the formula value, which is the default value of the current position.

[0074] Data preprocessing also includes normalizing the data:

[0075] Since the types of the 9 data are different, in order to eliminate the impact of the dimensions between the data and improve the monitoring accuracy, all the data are normalized as follows , the calculation formula is:

[0076] ;

[0077] Where, is the data output after normalization processing, is the original data, For variables The minimum value of For variables The maximum value of .

[0078] Data preprocessing can not only eliminate abnormal values ​​and reduce data fluctuations, but also transmit effective values ​​to reduce energy consumption and improve network transmission rate and transmission quality.

[0079] Step 3: Build a support vector machine regression model and initialize the parameters in the model:

[0080] The training sample set is , ,in, The corresponding input vector is The output vector of , therefore, the regression function is:

[0081] ;

[0082] Where, is the normal vector, is the feature vector after the input vector is mapped, is the offset term;

[0083] The training of the regression model is to find the optimal normal vector and offset term. Get as close as possible , this process is finally transformed into a convex optimization problem, which is in the form of:

[0084] ;

[0085] ;

[0086] Where, is the penalty coefficient, is the slack variable, is an insensitive loss function;

[0087] By introducing Lagrange multipliers, the original problem is transformed into a dual problem, and the calculation results are:

[0088] ;

[0089] Where, 、 is the Lagrange coefficient.

[0090] In the nonlinear case, a suitable kernel function is introduced to map the samples into a high-dimensional space and transform it into a linear one. The support vector machine regression model satisfies:

[0091] ;

[0092] ;

[0093] ;

[0094] Where, is the kernel function, corresponding to and The inner product mapped to the feature space is used to replace the complex operations in the feature space, simplify the calculation process, and shorten the calculation time. is the kernel function parameter;

[0095] Parameter initialization specifically includes: the penalty factor range is , the kernel function parameter range is .

[0096] Step 4: Establish a particle swarm optimization algorithm model to find the optimal solution for the penalty coefficient and kernel function parameters of the support vector machine regression model:

[0097] 1. The speed and position updates of the particle swarm optimization algorithm model satisfy:

[0098] ;

[0099] ;

[0100] Where, The number of iterations is Time Particles in The speed in dimensional space, is the inertia weight, The number of iterations is Time Particles in The speed in dimensional space, is the first acceleration factor, To be distributed in Random numbers in the interval, The number of iterations is Time Particles in The optimal solution obtained in the dimensional space, The number of iterations is Time Particles in The position vector in the dimensional space, is the second acceleration factor, To be distributed in Random numbers in the interval, The number of iterations is When the group The optimal solution in dimensional space, The number of iterations is Time Particles in Position vector in dimensional space;

[0101] Among them, the inertia weight satisfies:

[0102]

[0103] Where, is the minimum weight, is the maximum weight, is the fitness value of the particle at any time, is the minimum fitness value in the particle swarm, is the average fitness value of the particle swarm;

[0104] A large first acceleration factor will increase the iteration time, while a large second acceleration factor will cause the particles to converge to the local optimum too early, leading to premature algorithm maturation. Therefore, adjusting the two acceleration factors so that they change adaptively during the optimization process is conducive to convergence to the global optimum. The two acceleration factors satisfy:

[0105] ;

[0106] Where, 、 They are and The initial value of 、 They are and The ending value of .

[0107] 2. Initialize the particle swarm:

[0108] In the particle swarm optimization algorithm, the penalty coefficient and kernel function parameters are mapped to particles, the population size is 50, the number of iterations is 1000, the initial value of the inertia weight is 0.6, and the initial values ​​of the first acceleration factor and the second acceleration factor are both 2.

[0109] 3. The mean square error function of the support vector machine regression model is used as the fitness function to calculate the fitness of each particle, update the individual extreme value and the global optimal value until the iteration termination condition is met and the optimal parameter combination is obtained.

[0110] Step 5: Bring the optimal parameter combination into the support vector machine regression model to make the monitoring results more accurate and obtain the optimal results for monitoring the route of the transportation equipment.

[0111] The present invention designs and develops a multi-source information fusion monitoring method for open-pit mine transportation equipment based on automatic driving. By combining an adaptive particle swarm optimization algorithm with a support vector machine regression model, the calculation process is simplified, the calculation time is greatly shortened, and at the same time, the monitoring precision and accuracy can be improved.

[0112] Although the embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the description and implementation methods. They can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily implemented. Therefore, without departing from the general concept defined by the claims and the scope of equivalents, the present invention is not limited to the specific details and embodiments shown and described herein.

Claims

1. A monitoring method for open-pit mine transportation equipment based on multi-source information fusion for autonomous driving, characterized in that: The steps include: Step 1: collecting data to be fused from multiple sources of information according to a sampling period, wherein the data to be fused from multiple sources of information includes a position signal of the transport equipment, a vehicle speed signal, and a steering wheel rotation angle; Step 2: Preprocess the data to be fused from multiple sources; Step 3: Build a support vector machine regression model; Step 4: Establish a particle swarm optimization algorithm model, solve the penalty coefficient and kernel function parameters of the support vector machine regression model for the optimal solution, and obtain the optimal parameter combination; Step 5: Bring the optimal parameter combination into the support vector machine regression model to obtain the route monitoring results of the transportation equipment.

2. The method for monitoring multi-source information fusion of open-pit mine transportation equipment based on automatic driving according to claim 1, characterized in that: The position signal, speed signal and steering wheel rotation angle of the transport equipment in step 1 include: Initial position signal, predicted position signal at the previous moment, real-time position signal at the previous moment, average vehicle speed signal, predicted vehicle speed signal at the previous moment, real-time vehicle speed signal at the previous moment, real-time maximum steering wheel rotation angle, predicted steering wheel rotation angle at the previous moment, and real-time steering wheel rotation angle at the previous moment.

3. The method for monitoring multi-source information fusion of open-pit mine transportation equipment based on automatic driving according to claim 1, characterized in that: The position signal, vehicle speed signal and steering wheel rotation angle of the transport equipment are all monitored by sensors.

4. The method for monitoring multi-source information fusion of open-pit mine transportation equipment based on automatic driving according to claim 1, characterized in that: The second step is to perform data preprocessing on the data to be fused from multiple sources of information, including the elimination of abnormal values ​​and the processing of default values.

5. The method for monitoring multi-source information fusion of open-pit mine transportation equipment based on automatic driving according to claim 4, characterized in that: The elimination of outliers specifically includes the following steps: Step 1: Arrange the data collected by the sensor in ascending order; Step 2: Obtain high-end statistics and low-end statistics based on the sample size; Step 3: When the confidence interval At different times, the number of samples corresponds to different critical values : when , ,but For outliers, remove ;when , ,but For outliers, remove , otherwise there are no outliers; Where, is a high-end statistic, is a low-end statistic, and .

6. The method for monitoring multi-source information fusion of open-pit mine transportation equipment based on automatic driving according to claim 4, characterized in that: The default value processing specifically includes: At known coordinates and Coordinates to interpolate between satisfy: 。 7. The method for monitoring open-pit mine transportation equipment based on automatic driving and multi-source information fusion according to claim 1, characterized in that: The pre-processing of the data to be fused from the multi-source information further includes normalizing the data to be fused from the multi-source information: ； Where, is the result data output after normalization processing, For different original data, For variables The minimum value of For variables The maximum value of .

8. The method for monitoring open-pit mine transportation equipment based on automatic driving and multi-source information fusion according to claim 1, characterized in that: The support vector machine regression model satisfies: ； ； ； Where, is the kernel function, corresponding to and The inner product mapped to the feature space is used to replace the complex operations in the feature space. is the kernel function parameter, σ is the standard deviation of the kernel function, and b is the bias term.

9. The method for monitoring multi-source information fusion of open-pit mine transportation equipment based on automatic driving according to claim 8, characterized in that: The particle swarm optimization algorithm model specifically includes: The speed and position updates of the particle swarm optimization algorithm model satisfy: ； ； Where, The number of iterations is Time Particles in The speed in dimensional space, is the inertia weight, The number of iterations is Time Particles in The speed in dimensional space, is the first acceleration factor, To be distributed in Random numbers in the interval, The number of iterations is Time Particles in The optimal solution obtained in the dimensional space, The number of iterations is Time Particles in The position vector in the dimensional space, is the second acceleration factor, To be distributed in Random numbers in the interval, The number of iterations is When the group The optimal solution in dimensional space, The number of iterations is Time Particles in Position vector in dimensional space; Among them, the inertia weight satisfies: ； Where, is the minimum weight, is the maximum weight, is the fitness value of the particle at any time, is the minimum fitness value in the particle swarm, is the average fitness value of the particle swarm; The first acceleration factor and the second acceleration factor satisfy: ； Where, 、 They are and The initial value of 、 They are and The ending value of .

10. The method for monitoring open-pit mine transportation equipment based on automatic driving and multi-source information fusion according to claim 9, characterized in that: The parameters of the particle swarm optimization algorithm model are initialized as: The population size is 50, the number of iterations is 1000, the initial value of the inertia weight is 0.6, and the initial values ​​of the first acceleration factor and the second acceleration factor are both 2.

Images