Mine card real-time scheduling method and platform based on unmanned driving technology

Abstract

The invention discloses a mine card real-time scheduling method and platform based on an unmanned driving technology, and relates to the technical field of unmanned mine card scheduling, and the method comprises the following steps: obtaining an actual production plan, road condition data, first vehicle unloading time and second vehicle loading time of a mining area; inputting the road condition data into a road condition time model to obtain first total time of the first vehicle and second total time of the second vehicle output by the road condition time model; based on the first vehicle unloading time, the second vehicle loading time, the first total time and the second total time, constructing a target scheduling model by taking the transportation task, the minimum transportation time and the environmental protection influence as targets; and outputting a mine card scheduling instruction based on the target scheduling model in response to the actual production plan. The problems that mine card transportation time is not accurate due to weather or other factors, road blockage is likely to occur, and tasks cannot be completed in time can be solved as much as possible.

Description

Technical Field

[0001] The present invention relates to the technical field of dispatching unmanned mining trucks, and in particular to a real-time dispatching method and platform for mining trucks based on unmanned driving technology. Background Art

[0002] With the rapid development of automation and intelligent technology, unmanned driving technology has been widely used in many industries, especially in the field of mine transportation. Mining trucks are one of the indispensable equipment in mining operations, responsible for transporting ore from the mining point to the processing area or storage area. Traditional mining truck scheduling mainly relies on manual operation, which is inefficient and has safety hazards. In recent years, the development of unmanned driving technology has provided a new solution for real-time scheduling of mining trucks. Through the automated scheduling system, efficient and safe operation of mining trucks can be achieved. In addition, unmanned driving technology can also improve the production efficiency of mines, reduce operating costs, and reduce dependence on human resources.

[0003] The current scheduling methods used in mining areas can significantly improve the operating efficiency and safety of mining trucks while reducing operating costs. However, the natural conditions in mining areas are relatively harsh, and there may be many types of mining roads, such as asphalt roads, cement roads, dirt roads, and sand roads. The driving speed of mining trucks on different roads is affected differently by weather and other factors. Therefore, when production tasks are changed, they are easily affected by various factors such as weather, resulting in inaccurate grasp of time, road congestion, and inability to complete tasks in a timely manner, affecting the overall scheduling effect. Summary of the invention

[0004] The purpose of the present invention is to solve the problems in the prior art and to propose a real-time scheduling method and platform for mining trucks based on unmanned driving technology.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A real-time dispatching method for mining trucks based on unmanned driving technology comprises the following steps:

[0007] Obtaining the actual production plan of the mining area, road condition data, unloading time of the first vehicle and loading time of the second vehicle; the first vehicle is any fully loaded mining vehicle in the mining area, and the second vehicle is any unloaded mining vehicle in the mining area;

[0008] The road condition data is input into a road condition time model, and a first total time of a first vehicle and a second total time of a second vehicle output by the road condition time model are obtained; the first total time is the time taken by the first vehicle from the loading area to the unloading area, and the second total time is the time taken by the second vehicle from the unloading area to the loading area; the road condition time model is trained based on sample road condition data and its corresponding time label results;

[0009] Based on the unloading time of the first vehicle, the loading time of the second vehicle, the first total time and the second total time, a target scheduling model is constructed with the transportation task, the minimum transportation time and the environmental impact as the objectives;

[0010] In response to the actual production plan, the mining truck scheduling instructions are output based on the target scheduling model.

[0011] The present invention provides a real-time dispatching method for mining trucks based on unmanned driving technology, wherein the training step of the road condition time model comprises:

[0012] Acquire historical road condition data, historical driving data and historical weather data of roads in the mining area environment; the historical road condition data includes historical asphalt road surface data, historical cement road surface data, historical gravel road surface data and historical dirt road surface data.

[0013] Dividing the historical traffic condition data into a plurality of historical traffic condition sub-data according to the traffic condition type; each historical traffic condition sub-data corresponds to a traffic condition type;

[0014] The multiple historical road condition sub-data are matched with the historical driving data and the historical weather data according to the timestamp to obtain the historical driving sub-data and the historical weather sub-data corresponding to the multiple historical road condition sub-data at the same time.

[0015] Inputting the historical driving sub-data and the historical weather sub-data corresponding to the plurality of historical road condition sub-data at the same time into a pre-trained neural network for model training to obtain a road condition time model;

[0016] The present invention provides a real-time dispatching method for mining trucks based on unmanned driving technology, wherein the optimization objective function of the road condition time model is:

[0017]

[0018] Where n represents the number of data samples; T represents the time span; represents the actual road condition value of the i-th sample at time t; Represents the traffic condition prediction value of the i-th sample of the model at time T; represents the change in vehicle driving data in the i-th sample at time t, Δt represents the time interval, Indicates the change in traffic data at time t; represents the weather data vector corresponding to the i-th sample, m represents the dimension of the weather data vector, represents the covariance of weather data and road condition data, Represent the expected values ​​of weather data and road condition data respectively; θ krepresents the parameters of the model; γ1, γ2, γ3 and γ4 represent the corresponding weight coefficients respectively.

[0019] The present invention provides a real-time scheduling method for mining trucks based on unmanned driving technology, wherein the first total time and the second total time also include traffic light waiting time and road obstacle avoidance time, wherein the traffic light waiting time is the time taken by the first vehicle or the second vehicle to wait at all traffic lights during driving; and the road obstacle avoidance time is the total time taken by the first vehicle or the second vehicle to avoid roadblocks.

[0020] The present invention provides a real-time scheduling method for mining trucks based on unmanned driving technology, which constructs a target scheduling model based on the unloading time of the first vehicle, the loading time of the second vehicle, the first total time and the second total time, with the transportation task, the minimum transportation time and the environmental impact as the goals, including:

[0021] Determining constraints based on the first vehicle unloading time, the second vehicle loading time, the first total time, and the second total time; the constraints include transportation task constraints, vehicle load constraints, vehicle speed constraints, loading and unloading time constraints, and environmental constraints;

[0022] Determine an objective function based on the constraints according to the transportation task, the minimum transportation time and the environmental impact;

[0023] A target scheduling model is constructed based on the target function.

[0024] The present invention provides a real-time scheduling method for mining trucks based on unmanned driving technology, wherein the transportation task constraint formula is:

[0025] Σ α Σ β x αβ q β ≥Q ti

[0026] ; The vehicle load constraint formula is:

[0027] x αβ q β ≤C α

[0028] ; The vehicle speed constraint formula is:

[0029]

[0030] ; The loading and unloading time constraint formula is:

[0031]

[0032] ; The environmental protection constraint formula is:

[0033]

[0034] ; where x αβ Indicates whether the αth mining truck is assigned to the task of transporting from the loading area β to the unloading area; f α represents the departure time of the αth mining truck; q α represents the actual load of the αth mining truck; q β represents the amount of ore that can be loaded from the loading area β, Q ti represents the total amount of ore that needs to be transported according to the production plan within the time period ti; C α represents the rated load of the αth mining truck; v αβ represents the speed of the αth mining truck on section β, Indicates the maximum permissible speed for a road section based on road conditions and vehicle performance; and Denote the working time and end time of loading and unloading area β, respectively, αβ represents the travel time of the αth mining truck from the departure point to the loading and unloading area β; e α (d αβ ,v αβ ) represents the travel distance d of the αth mining truck αβ and speed v αβ The relevant exhaust emission function, E max Indicates the maximum tail gas emission allowed by the mining area environmental requirements; s β represents the dust generation coefficient of the road type corresponding to the loading area β, S max Indicates the maximum allowable dust generation.

[0035] The present invention provides a real-time dispatching method for mining trucks based on unmanned driving technology, wherein the formula after the objective function is optimized is:

[0036]

[0037]

[0038] ; Among them, p1 represents the penalty coefficient of the transportation task, Task represents the penalty item for completing the transportation task; Time represents the item for minimizing the transportation time; Envi represents the penalty item for environmental impact, and p2 represents the penalty coefficient for environmental impact; and Represents the weight coefficient.

[0039] A real-time dispatching platform for mining trucks based on unmanned driving technology, including:

[0040] Acquisition unit: used to acquire the actual production plan of the mining area, road condition data, unloading time of the first vehicle and loading time of the second vehicle; the first vehicle is any fully loaded mining vehicle in the mining area, and the second vehicle is any unloaded mining vehicle in the mining area;

[0041] A calculation unit is used to input the road condition data into a road condition time model to obtain a first total time of a first vehicle and a second total time of a second vehicle output by the road condition time model; the first total time is the time taken by the first vehicle from the loading area to the unloading area, and the second total time is the time taken by the second vehicle from the unloading area to the loading area; the road condition time model is trained based on sample road condition data and its corresponding time label results;

[0042] Model building unit: used for building a target scheduling model based on the unloading time of the first vehicle, the loading time of the second vehicle, the first total time and the second total time, with the transportation task, the minimum transportation time and the environmental impact as the objectives;

[0043] Instruction output unit: used to output mining truck scheduling instructions based on the target scheduling model in response to the actual production plan.

[0044] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps of the above-mentioned method for real-time scheduling of mining trucks based on unmanned driving technology are implemented.

[0045] A non-transitory computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned method for real-time scheduling of mining trucks based on unmanned driving technology.

[0046] Compared with the prior art, the present invention has the following advantages:

[0047] The present invention provides a real-time scheduling method and platform for mining trucks based on unmanned driving technology. The method obtains road condition data in the mining area and inputs the road condition data into a road condition time model for calculation, thereby accurately determining the vehicle driving time of the mining truck under different road conditions, and trying to avoid the problems of inaccurate mining truck transportation time, road congestion, and failure to complete tasks in time caused by weather or other factors. Then, the predicted precise time is combined with the mining truck loading and unloading time, transportation tasks, transportation time, and environmental impact to construct a scheduling model to achieve multi-objective optimization processing, further solving the problem of difficulty in balancing and optimizing multiple objectives in the scheduling process, so that the dynamic change requirements of tasks can be quickly responded to in the mining area scheduling, the impact on transportation tasks can be minimized as much as possible, and the production operation risks can be reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0049] Figure 1 A schematic flow chart of a method for real-time dispatching of mining trucks based on unmanned driving technology provided by an embodiment of the present invention;

[0050] Figure 2 A schematic diagram of the structure of a real-time dispatching platform for mining trucks based on unmanned driving technology provided by an embodiment of the present invention;

[0051] Figure 3 This is a schematic diagram of the structure of the electronic device proposed by the present invention. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] Combine the following Figure 1 - Figure 3 The present invention describes a real-time dispatching method and platform for mining trucks based on unmanned driving technology.

[0054] Figure 1 The figure is a flow chart of a method for real-time dispatching of mining trucks based on unmanned driving technology provided by the present invention. Figure 1 As shown, the following steps are included:

[0055] Step 101, obtaining the actual production plan of the mining area, road condition data, unloading time of the first vehicle and loading time of the second vehicle; the first vehicle is any fully loaded mining car in the mining area, and the second vehicle is any unloaded mining car in the mining area.

[0056] When obtaining the actual production plan of the mining area, detailed production plan data can be obtained from the production management information system of the mining area, including daily, weekly and monthly ore mining targets, as well as mining vehicles. Specifically, each mining truck in the mining area is equipped with a unique identification, and the mining truck can be allocated according to the unique identification. In addition, the actual production plan of the mining area can also be manually input into the production management system by the staff to cope with sudden production plan needs.

[0057] For road condition data, firstly, we obtain all road data corresponding to the actual production plan through the high-precision map in the mining area, and then collect road condition data based on the sensors arranged in all road data. The sensors include laser flatness meters for detecting road flatness, meteorological sensors for detecting road temperature, humidity, precipitation and wind speed, AC flow sensors for detecting road flow conditions, and visual image acquisition equipment for collecting obstacles, water accumulation and cracks on the road. The sensors detect road conditions in real time and upload the detection data to the cloud server or the terminal control room in the mining area. After that, the collected data is processed to obtain the required road condition data.

[0058] The unloading time of the first vehicle and the loading time of the second vehicle can be obtained from historical data, such as the same specifications of mining trucks and the same specifications of excavating equipment, based on historical data, or according to the loading space of the mining truck and the excavation volume per unit time of the excavating equipment, after calculation, the loading time of each mining truck for the excavating equipment and the unloading time used by the mining truck itself when unloading can be obtained.

[0059] Step 102, input the road condition data into the road condition time model, and obtain the first total time of the first vehicle and the second total time of the second vehicle output by the road condition time model; the first total time is the time taken by the first vehicle from the loading area to the unloading area, and the second total time is the time taken by the second vehicle from the unloading area to the loading area; the road condition time model is trained based on the sample road condition data and its corresponding time label results.

[0060] The first total time and the second total time also include the traffic light waiting time and the road obstacle avoidance time. The traffic light waiting time is the time taken by the first vehicle or the second vehicle to wait at all traffic lights during the driving process; the road obstacle avoidance time is the total time taken by the first vehicle or the second vehicle to avoid road obstacles. By setting the traffic light waiting time and the road obstacle avoidance time, the accuracy of the first total time and the second total time can be further improved when the first total time and the second total time are used for subsequent scheduling calculations.

[0061] After obtaining the road condition data used in the actual production plan, the road condition in the future can be predicted in real time based on the constructed road condition time model, and then the first total time of the first vehicle and the second total time of the second vehicle required for subsequent mining truck scheduling can be output. The first total time and the second total time are both obtained by summarizing several branch times used under different stages of road conditions.

[0062] Specifically, the training steps of the traffic time model include:

[0063] Obtain historical road condition data, historical driving data and historical weather data of the roads in the mining area environment; historical road condition data includes historical asphalt pavement data, historical cement pavement data, historical gravel pavement data and historical dirt road pavement data. For the acquisition of historical asphalt pavement data, historical cement pavement data, historical gravel pavement data and historical dirt road pavement data, the above-mentioned sensors and visual acquisition equipment are also used for acquisition, and they can also be input into the mining area road map system after manual investigation. As for the historical driving data, it is obtained through the on-board sensors of the mining trucks in the mining area, including speed sensors, acceleration sensors, steering angle sensors and load sensors, etc. The data on the mining trucks are uploaded to the cloud at a certain time interval, and the driving data are accompanied by corresponding timestamps to facilitate subsequent matching with historical road condition data and historical weather data. As for the historical weather data, it is collected from the meteorological station in the mining area, including temperature, humidity, air pressure, wind speed, wind direction and precipitation data, etc.

[0064] The historical road condition data is divided into a plurality of historical road condition sub-data according to the road condition type; each historical road condition sub-data corresponds to a road condition type; by dividing the historical road condition sub-data into a plurality of historical road condition sub-data, the data can be classified and stored to facilitate subsequent use.

[0065] According to the timestamp, multiple historical road condition sub-data are matched with historical driving data and historical weather data to obtain historical driving sub-data and historical weather sub-data corresponding to multiple historical road condition sub-data at the same time. When matching data, firstly, a data management data is constructed to retrieve the historical driving data and historical weather data corresponding to the historical road condition sub-data according to the time information, and then the retrieved data is combined. At the same time, when combining the data, the retrieved data is normalized to unify the different data into a similar data range, so as to facilitate the subsequent training of the neural network.

[0066] The historical driving sub-data and historical weather sub-data corresponding to multiple historical road condition sub-data at the same time are input into the pre-trained neural network for model training to obtain the road condition time model; during model training, the matched data is divided into a training set and a validation set, and the pre-trained recurrent neural network is trained using the training set, and appropriate hyperparameters such as the learning rate, number of training rounds, and batch size are set. During the training process, the weights and biases of the neural network are continuously adjusted through the back propagation algorithm to minimize the loss function. The validation set is used to monitor the performance of the model, and the training is stopped when the loss on the validation set no longer decreases or overfitting occurs.

[0067] Correspondingly, the optimization objective function of the traffic time model is:

[0068]

[0069]

[0070] Where n represents the number of data samples; T represents the time span; represents the actual road condition value of the i-th sample at time t, such as the road surface flatness index or damage degree value, Represents the traffic condition prediction value of the i-th sample of the model at time T; represents the change in vehicle driving data in the i-th sample at time t, Δt represents the time interval, Indicates the change in traffic data at time t; represents the weather data vector corresponding to the i-th sample, m represents the dimension of the weather data vector, represents the covariance of weather data and road condition data, Represent the expected values ​​of weather data and road condition data respectively; θ k represents the parameters of the model; γ1, γ2, γ3 and γ4 represent the corresponding weight coefficients. It is the L2 regularization term of the model parameters, which is used to prevent the model from overfitting. By limiting the size of the model parameters, the model's fitting effect on the training data and its generalization ability on unknown data are balanced, avoiding the overfitting of the training data due to the model being too complex, thereby ensuring the effectiveness and stability of the model in practical applications.

[0071] Through the design of the optimization objective function, the model can reasonably evaluate the prediction errors under various road conditions on the one hand, avoiding the deviation of error evaluation due to the difference in the size of the data itself; in addition, it can measure the consistency of the vehicle driving change trend and the road condition change trend by calculating the sum of the absolute values ​​of the difference between the vehicle driving data change rate and the road condition data change rate, so as to enable the model to learn the inherent dynamic correlation between the two, so as to more accurately predict the road condition changes according to the vehicle driving conditions or adjust the vehicle driving strategy according to the road conditions; on the other hand, by evaluating the degree of deviation from the linear correlation between weather data and road condition data, the model can better capture the complex impact of weather on road conditions, rather than just a simple linear correlation, which helps to improve the reliability of road condition prediction under different weather conditions.

[0072] Step 103, based on the unloading time of the first vehicle, the loading time of the second vehicle, the first total time and the second total time, a target scheduling model is constructed with transportation tasks, minimum transportation time and environmental impact as objectives;

[0073] It includes: determining constraints based on the unloading time of the first vehicle, the loading time of the second vehicle, the first total time and the second total time; the constraints include transportation task constraints, vehicle load constraints, vehicle speed constraints, loading and unloading time constraints and environmental protection constraints.

[0074] When the transportation task is constrained, it is to ensure that the ore transportation task in the mine production plan can be completed smoothly. αβ Indicates whether the αth mining truck is assigned to the task of transporting from the loading area β to the unloading area; where x αβ =1 means allocation, x αβ =0 means unassigned. β represents the amount of ore that can be loaded from the loading area β, Q ti It represents the total amount of ore that needs to be transported according to the production plan within the time period ti, so the transportation task constraint formula is determined as:

[0075] ∑ α ∑ β x αβ q β ≥Q ti

[0076] This constraint ensures that the total transportation volume completed by all mining trucks must at least reach the total ore transportation volume required by the production plan, ensuring that production is not restricted by the transportation link.

[0077] When performing vehicle load constraints, since each mining truck has a rated load, in order to ensure the safe driving and normal operation of the vehicle, C α represents the rated load of the αth mining truck, and the vehicle load constraint is set as:

[0078] x αβ q β ≤C α

[0079] Under this constraint, ensure that mining trucks do not exceed the limit, and avoid problems such as increased vehicle wear, reduced braking performance, and road damage caused by overloading.

[0080] When constraining vehicle speed, the vehicle speed on the mining road is limited by many factors, including road conditions (such as road type, slope, flatness) and vehicle performance (such as power, braking system). αβ represents the speed of the αth mining truck on section β, Indicates that the maximum permissible speed of a road section is determined based on road conditions and vehicle performance, and the vehicle speed constraint is set as:

[0081]

[0082] Under this constraint, it helps to reasonably control vehicle speed, improve driving safety, and reduce the risk of accidents caused by speeding. It can also prevent vehicles from causing excessive damage to roads when driving at high speeds under unsuitable road conditions, and protect road infrastructure in mining areas.

[0083] When setting loading and unloading time constraints, the working time limits of the loading and unloading areas and the actual efficiency of vehicle loading and unloading operations should be taken into consideration. and They represent the working time and end time of loading and unloading area β, respectively, α represents the departure time of the αth mining truck, D αβ represents the travel time of the αth mining truck from the departure point to the loading and unloading area β, and the loading and unloading time constraint is:

[0084]

[0085] This constraint ensures that the mining truck can arrive at the loading and unloading area within the normal working hours to carry out loading and unloading operations, avoiding time waste or operation delays caused by arriving too early or too late, and improving the time utilization efficiency of the entire transportation process.

[0086] When carrying out environmental protection constraints, the purpose is to consider the impact on the environment and set environmental protection constraints during the mining truck dispatch process as environmental protection requirements are increasingly increasing. First of all, exhaust emissions are an important aspect of the impact of mining trucks on the environment. α (d αβ ,v αβ ) represents the travel distance d of the αth mining truck αβ and speed v αβ The relevant exhaust emission function, E max represents the maximum tail gas emission allowed by the mining area environmental requirements; then the tail gas emission constraint is:

[0087]

[0088] ; By controlling the total amount of tail gas emissions, we can reduce the pollution of mining truck transportation to the air quality in the mining area and surrounding areas, and protect the environment and employee health as much as possible.

[0089] In addition, environmental protection constraints also include dust generation constraints, because dust generation is also an environmental problem in the transportation process in the mining area, especially on roads with poor road conditions such as dirt roads or gravel roads. β represents the dust generation coefficient of the road type corresponding to the loading area β, S max represents the maximum allowed dust generation. The dust generation constraint is:

[0090]

[0091] This constraint can prompt the dispatching system to minimize the number of trips on dust-prone sections or control the driving speed when arranging the routes and tasks of mining trucks, thereby reducing the impact of dust on the mining environment and improving the ecological environment of the mining area.

[0092] The objective function based on constraints is determined according to the transportation task, the minimum transportation time and the environmental impact; the formula after the objective function is optimized is:

[0093]

[0094] ; Among them, p1 represents the penalty coefficient of the transportation task, which is used to adjust the influence of the unfinished task volume on the objective function. When the actual transportation volume is less than the planned transportation volume, the penalty term will increase according to the square of the difference, so that the model tends to avoid this situation; Task represents the penalty term for completing the transportation task. If the transportation task is not completed according to the production plan, this item will produce a larger penalty value, prompting the model to give priority to meeting the production task requirements; Time represents the transportation time minimization item. By minimizing this item, the model will try to arrange the driving route and departure time of the mining truck to shorten the overall transportation time and improve transportation efficiency; Envi represents the environmental impact penalty term, and p2 represents the environmental impact penalty coefficient, which is used to adjust the influence of exhaust emissions and dust generation on the objective function; and represents the weight coefficient, which is used to balance the importance of transportation tasks, minimum transportation time and environmental impact in the target scheduling model, and These coefficients can be adjusted dynamically based on the actual needs of the mining area, environmental protection policy requirements, operating costs and other factors. For example, during periods of tight production tasks, the coefficients can be appropriately increased. value to ensure that production tasks are completed first; when environmental pressure is high, increase ratio to strengthen environmental control; and in pursuit of efficient transportation, it can be increased weight to shorten transportation time.

[0095] Through the target scheduling model and its objective function constructed above, we can comprehensively consider various factors such as transportation tasks, transportation time and environmental impact on the basis of meeting various constraints, provide a scientific and reasonable decision-making basis for the scheduling of mining trucks in the mining area, and realize the efficient, low-consumption and environmentally friendly operation of the mining area transportation system.

[0096] Construct a target scheduling model based on the objective function.

[0097] Step 104: In response to the actual production plan, output the mining truck scheduling instruction based on the target scheduling model.

[0098] In actual work, based on the constructed target scheduling model and according to the actual production plan of the mining area, the model is solved through optimization algorithms (such as genetic algorithms, particle swarm optimization algorithms, etc.), the task allocation of each mining truck (including loading area, unloading area, departure time, driving route, etc.) is determined, and the mining truck scheduling instructions are output. These instructions will be transmitted to the control system of the unmanned mining truck in real time through the wireless communication network. The mining truck will automatically perform the transportation task according to the instructions to realize the automatic and intelligent scheduling of the mining truck. During the transportation process, the system will also continuously monitor the road condition data, vehicle status data and changes in the production plan. If an abnormality or adjustment is found (such as road congestion due to sudden accidents, temporary changes in production plans, etc.), the target scheduling model will be re-optimized in time and the mining truck scheduling instructions will be updated to ensure the efficient, stable and sustainable operation of the entire mining area transportation system.

[0099] Figure 2 The structure diagram of a real-time dispatching platform for mining trucks based on unmanned driving technology provided by the present invention is as follows: Figure 2 As shown, the platform includes: an acquisition unit 10: used to acquire the actual production plan of the mining area, road condition data, the unloading time of the first vehicle and the loading time of the second vehicle; the first vehicle is any fully loaded mine car in the mining area, and the second vehicle is any unloaded mine car in the mining area; a calculation unit 20: used to input the road condition data into the road condition time model, and obtain the first total time of the first vehicle and the second total time of the second vehicle output by the road condition time model; the above-mentioned first total time is the time taken by the first vehicle from the loading area to the unloading area, and the second total time is the time taken by the second vehicle from the unloading area to the loading area; the road condition time model is obtained by training based on sample road condition data and its corresponding time label results; a model construction unit 30: used to construct a target scheduling model based on the unloading time of the first vehicle, the loading time of the second vehicle, the first total time and the second total time, with the transportation task, the minimum transportation time and the environmental impact as the target; an instruction output unit 40: used to output the mining truck scheduling instruction based on the target scheduling model in response to the actual production plan.

[0100] Figure 3 is a schematic diagram of the structure of the electronic device provided by the present invention, such as Figure 3As shown, the electronic device may include: a processor (processor) 310, a communication interface (CommunicationsInterface) 320, a memory (memory) 330 and a communication bus 340, wherein the processor 33, the communication interface 320, and the memory 330 communicate with each other through the communication bus 340. The processor 33 can call the logic instructions in the memory 330, and can execute a real-time scheduling method and platform for mining trucks based on unmanned driving technology, the method comprising: obtaining the actual production plan of the mining area, road condition data, the unloading time of the first vehicle and the loading time of the second vehicle; the first vehicle is any fully loaded mining truck in the mining area, and the second vehicle is any unloaded mining truck in the mining area; the road condition data is input into the road condition time model to obtain the first total time of the first vehicle and the second total time of the second vehicle output by the road condition time model; the first total time is the time taken by the first vehicle from the loading area to the unloading area, and the second total time is the time taken by the second vehicle from the unloading area to the loading area; the road condition time model is trained based on sample road condition data and its corresponding time label results; based on the unloading time of the first vehicle, the loading time of the second vehicle, the first total time and the second total time, a target scheduling model is constructed with transportation tasks, minimum transportation time, and environmental impact as the goals; in response to the actual production plan, based on the target scheduling model, a mining truck scheduling instruction is output.

[0101] In addition, the logic instructions in the above-mentioned memory 330 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0102] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by the computer, the computer can execute a real-time scheduling method and platform for mining trucks based on unmanned driving technology provided by the above methods, the method comprising: obtaining an actual production plan of a mining area, road condition data, and an unloading time of a first vehicle and a loading time of a second vehicle; the first vehicle is any fully loaded mining truck in the mining area, and the second vehicle is any unloaded mining truck in the mining area; the road condition data is input into a road condition time model to obtain a first total time of the first vehicle and a second total time of the second vehicle output by the road condition time model; the first total time is the time taken by the first vehicle from the loading area to the unloading area, and the second total time is the time taken by the second vehicle from the unloading area to the loading area; the road condition time model is trained based on sample road condition data and its corresponding time label results; based on the unloading time of the first vehicle, the loading time of the second vehicle, the first total time and the second total time, a target scheduling model is constructed with transportation tasks, minimum transportation time, and environmental impact as targets; in response to the actual production plan, based on the target scheduling model, a mining truck scheduling instruction is output.

[0103] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the processor executes the above-mentioned method and platform for real-time scheduling of mining trucks based on unmanned driving technology, the method comprising: obtaining an actual production plan of the mining area, road condition data, unloading time of a first vehicle and loading time of a second vehicle; the first vehicle is any fully loaded mining truck in the mining area, and the second vehicle is any unloaded mining truck in the mining area; the road condition data is input into a road condition time model to obtain a first total time of the first vehicle and a second total time of the second vehicle output by the road condition time model; the first total time is the time taken by the first vehicle from the loading area to the unloading area, and the second total time is the time taken by the second vehicle from the unloading area to the loading area; the road condition time model is trained based on sample road condition data and its corresponding time label results; based on the unloading time of the first vehicle, the loading time of the second vehicle, the first total time and the second total time, a target scheduling model is constructed with transportation tasks, minimum transportation time and environmental impact as the goals; in response to the actual production plan, based on the target scheduling model, a mining truck scheduling instruction is output.

[0104] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, i.e., they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative effort.

[0105] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0106] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A real-time dispatching method for mining trucks based on unmanned driving technology, characterized in that: The following steps are involved: Obtaining the actual production plan of the mining area, road condition data, unloading time of the first vehicle and loading time of the second vehicle; the first vehicle is any fully loaded mining vehicle in the mining area, and the second vehicle is any unloaded mining vehicle in the mining area; Input the road condition data into a road condition time model to obtain a first total time of the first vehicle and a second total time of the second vehicle output by the road condition time model; the first total time is the time taken by the first vehicle from the loading area to the unloading area, and the second total time is the time taken by the second vehicle from the unloading area to the loading area; The traffic time model is obtained by training based on sample traffic data and its corresponding time label results; Based on the unloading time of the first vehicle, the loading time of the second vehicle, the first total time and the second total time, a target scheduling model is constructed with the transportation task, the minimum transportation time and the environmental impact as the objectives; In response to the actual production plan, the mining truck scheduling instructions are output based on the target scheduling model.

2. According to claim 1, a real-time dispatching method for mining trucks based on unmanned driving technology is characterized in that: The training step of the traffic time model includes: Acquire historical road condition data, historical driving data and historical weather data of roads in the mining area environment; the historical road condition data includes historical asphalt road surface data, historical cement road surface data, historical gravel road surface data and historical dirt road surface data; Dividing the historical traffic condition data into a plurality of historical traffic condition sub-data according to the traffic condition type; each historical traffic condition sub-data corresponds to a traffic condition type; Matching the plurality of historical road condition sub-data with the historical driving data and the historical weather data according to the timestamps to obtain the historical driving sub-data and the historical weather sub-data corresponding to the plurality of historical road condition sub-data at the same time; The historical driving sub-data and historical weather sub-data corresponding to the multiple historical road condition sub-data at the same time are input into the pre-trained neural network for model training to obtain a road condition time model.

3. The real-time dispatching method for mining trucks based on unmanned driving technology according to claim 2 is characterized in that: The optimization objective function of the traffic time model is: Where n represents the number of data samples; T represents the time span; represents the actual road condition value of the i-th sample at time t; Represents the traffic condition prediction value of the i-th sample of the model at time T; represents the change in vehicle driving data in the i-th sample at time t, Δt represents the time interval, Indicates the change in traffic data at time t; represents the weather data vector corresponding to the i-th sample, m represents the dimension of the weather data vector, represents the covariance of weather data and road condition data, Represent the expected values ​​of weather data and road condition data respectively; θ k represents the parameters of the model; γ1, γ2, γ3 and γ4 represent the corresponding weight coefficients respectively.

4. The real-time dispatching method for mining trucks based on unmanned driving technology according to claim 1 is characterized in that: The first total time and the second total time also include traffic light waiting time and road obstacle avoidance time. The traffic light waiting time is the time taken by the first vehicle or the second vehicle to wait at all traffic lights during driving; the road obstacle avoidance time is the total time taken by the first vehicle or the second vehicle to avoid roadblocks.

5. The real-time dispatching method for mining trucks based on unmanned driving technology according to claim 1 is characterized in that: The target scheduling model is constructed based on the unloading time of the first vehicle, the loading time of the second vehicle, the first total time and the second total time, with the transportation task, the minimum transportation time and the environmental impact as the goals, including: Determining constraints based on the first vehicle unloading time, the second vehicle loading time, the first total time, and the second total time; the constraints include transportation task constraints, vehicle load constraints, vehicle speed constraints, loading and unloading time constraints, and environmental constraints; Determine an objective function based on the constraints according to the transportation task, the minimum transportation time and the environmental impact; A target scheduling model is constructed based on the target function.

6. A real-time dispatching method for mining trucks based on unmanned driving technology according to claim 5, characterized in that: The transportation task constraint formula is: ∑ α ∑ β x αβ q β ≥Q ti ; The vehicle load constraint formula is: x αβ q β ≤C α ; The vehicle speed constraint formula is: ； The loading and unloading time constraint formula is: ； The environmental protection constraint formula is: ； Among them, x αβ Indicates whether the αth mining truck is assigned to the task of transporting from the loading area β to the unloading area; f α represents the departure time of the αth mining truck; q α represents the actual load of the αth mining truck; q β represents the amount of ore that can be loaded from the loading area β, Q ti represents the total amount of ore that needs to be transported according to the production plan within the time period ti; C α represents the rated load of the αth mining truck; v αβ represents the speed of the αth mining truck on section β, Indicates the maximum permissible speed for a road section based on road conditions and vehicle performance; and Denote the working time and end time of loading and unloading area β, respectively, αβ represents the travel time of the αth mining truck from the departure point to the loading and unloading area β; e α (d αβ , v αβ ) represents the travel distance d of the αth mining truck αβ and speed v αβ The relevant exhaust emission function, E max Indicates the maximum tail gas emission allowed by the mining area environmental requirements; s β represents the dust generation coefficient of the road type corresponding to the loading area β, S max Indicates the maximum allowable dust generation.

7. A real-time dispatching method for mining trucks based on unmanned driving technology according to claim 6, characterized in that: The formula after the objective function is optimized is: ； Among them, p1 represents the penalty coefficient of the transportation task, Task represents the penalty item for completing the transportation task; Time represents the item for minimizing the transportation time; Envi represents the penalty item for environmental impact, and p2 represents the penalty coefficient for environmental impact; and Represents the weight coefficient.

8. A real-time dispatching platform for mining trucks based on unmanned driving technology, characterized in that: include: Acquisition unit: used to acquire the actual production plan of the mining area, road condition data, the unloading time of the first vehicle and the loading time of the second vehicle; The first vehicle is any fully loaded mine vehicle in the mining area, and the second vehicle is any unloaded mine vehicle in the mining area; A calculation unit is used to input the road condition data into a road condition time model to obtain a first total time of the first vehicle and a second total time of the second vehicle output by the road condition time model; the first total time is the time taken by the first vehicle from the loading area to the unloading area, and the second total time is the time taken by the second vehicle from the unloading area to the loading area; The traffic time model is obtained by training based on sample traffic data and its corresponding time label results; Model building unit: used for building a target scheduling model based on the unloading time of the first vehicle, the loading time of the second vehicle, the first total time and the second total time, with the transportation task, the minimum transportation time and the environmental impact as the objectives; Instruction output unit: used to output mining truck scheduling instructions based on the target scheduling model in response to the actual production plan.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of a real-time scheduling method for mining trucks based on unmanned driving technology as described in any one of claims 1 to 7 are implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a real-time scheduling method for mining trucks based on unmanned driving technology as described in any one of claims 1 to 7 are implemented.

Images