Dynamic scheduling method, system, storage medium, and program

Abstract

A dynamic scheduling method, a system, a storage medium, and a program. The dynamic scheduling method comprises: acquiring a road condition and environmental parameter of an open-pit mine, and operational data and performance parameters of a plurality of traffic participants; generating a graph data structure on the basis of the road condition and environmental parameter of the open-pit mine and the operational data and performance parameters of the plurality of traffic participants, wherein each node in the graph data structure comprises relevant information of each traffic participant; inputting the graph data structure into a neural network model to obtain a probability distribution of a collision occurring between each node in the graph data structure and the other nodes; and adjusting a driving policy of at least one traffic participant among the plurality of traffic participants based on the probability distribution of a collision occurring between each node in the graph data structure and the other nodes.

Description

Dynamic scheduling methods, systems, storage media, and programs

[0001] Cross-reference to related applications

[0002] This application is based on and claims priority to CN application No. 202411854552.9, filed on December 16, 2024, the disclosure of which is incorporated herein by reference in its entirety. Technical Field

[0003] This disclosure relates to the field of vehicle scheduling technology, and in particular to a dynamic scheduling method, system, storage medium, and program. Background Technology

[0004] In the field of open-pit mining, with the development of technology, unmanned driving technology has become an important means to improve operational efficiency, reduce labor costs, and enhance operational safety. Summary of the Invention

[0005] According to one aspect of this disclosure, a dynamic scheduling method is provided, comprising: acquiring road conditions and environmental parameters of an open-pit mine, as well as operational data and performance parameters of multiple traffic participants; generating a graph data structure based on the road conditions and environmental parameters of the open-pit mine and the operational data and performance parameters of the multiple traffic participants, wherein each node in the graph data structure contains relevant information of each traffic participant; inputting the graph data structure into a neural network model to obtain a probability distribution of collisions between each node in the graph data structure and other nodes; and adjusting the driving strategy of at least one of the multiple traffic participants based on the probability distribution of collisions between each node in the graph data structure and other nodes.

[0006] In some embodiments, generating a graph data structure based on the road conditions and environmental parameters of the open-pit mine and the operational data and performance parameters of the multiple traffic participants includes: generating path points of the trajectories of the multiple traffic participants based on the road conditions and environmental parameters of the open-pit mine and the operational data and performance parameters of the multiple traffic participants; determining path relationships between different path points based on the path points of the trajectories of the multiple traffic participants; and obtaining the graph data structure based on the path points of the trajectories of the multiple traffic participants and the path relationships between different path points, wherein the path relationships serve as edges between different nodes of the graph data structure.

[0007] In some embodiments, the plurality of traffic participants include a plurality of mining trucks; based on the road conditions and environmental parameters of the open-pit mine and the operating data and performance parameters of the plurality of traffic participants, waypoints for the trajectories of the plurality of traffic participants are generated, including: generating a task route for each mining truck based on the scheduling task of each mining truck; generating the expected speed of each mining truck based on the load information of each mining truck; and generating waypoints for the planned trajectory of each mining truck based on the task route, expected speed, and speed limit information of different roads.

[0008] In some embodiments, the road conditions include: traffic conditions and speed limit information of multiple roads in the open-pit mine; the environmental parameters include: weather condition parameters of the open-pit mine; the operating data includes: current location information, speed information, acceleration information and load information of each mining truck; and the performance parameters include: the driving speed of each mining truck when unloaded and the driving speed when fully loaded.

[0009] In some embodiments, the plurality of traffic participants further includes one or more electric shovels; the operational data includes: the speed, orientation, and historical position of the one or more electric shovels; the path points of the trajectories of the plurality of traffic participants are generated based on the road conditions and environmental parameters of the open-pit mine and the operational data and performance parameters of the plurality of traffic participants, and further includes: generating the future trajectories of the one or more electric shovels based on the speed, orientation, and historical position of the one or more electric shovels using a Kalman filter algorithm.

[0010] In some embodiments, the relevant information for each traffic participant includes: the identifier of each traffic participant, the coordinates of each waypoint of each traffic participant, the priority of each traffic participant's current task, the current speed information of each traffic participant, the current acceleration information of each traffic participant, the current load information of each traffic participant, the estimated time for each traffic participant to reach each node, and the path relationship between different waypoints.

[0011] In some embodiments, inputting the graph data structure into a neural network model to obtain the probability distribution of collisions between each node in the graph data structure and other nodes includes: processing the graph data structure sequentially through a graph convolutional network, a pooling layer, and a connection layer in the neural network model to obtain each node and other nodes that may collide with each node; inputting each node and other nodes that may collide with each node into a feedforward neural network of the neural network model to output a probability distribution vector of collisions between each node and other nodes; and calculating the probability distribution of collisions between each node and other nodes using a normalized exponential function based on the probability distribution vector of collisions between each node and other nodes.

[0012] In some embodiments, adjusting the driving strategy of at least one of the plurality of traffic participants based on the probability distribution of collisions between each node in the graph data structure and other nodes includes: adjusting the driving strategy of the at least one traffic participant based on the probability distribution of collisions between each node in the graph data structure and other nodes, so as to minimize the probability of the at least one traffic participant colliding with other traffic participants among the plurality of traffic participants and maximize the benefit of the at least one traffic participant.

[0013] In some embodiments, the at least one traffic participant includes at least one mining truck; the revenue of the at least one traffic participant is calculated based on the total load or total unload of each of the at least one mining trucks from the loading point to the unloading point, the distance of each of the at least one mining trucks from the loading point to the unloading point, the energy consumption per unit distance of each of the at least one mining trucks when fully loaded, the waiting time of each of the at least one mining trucks, and the energy consumption per unit time of each of the at least one mining trucks while waiting.

[0014] According to another aspect of this disclosure, a dynamic scheduling system is provided, comprising: a data acquisition unit for acquiring road conditions and environmental parameters of an open-pit mine, as well as operating data and performance parameters of multiple traffic participants; a generation unit for generating a graph data structure based on the road conditions and environmental parameters of the open-pit mine and the operating data and performance parameters of the multiple traffic participants, wherein each node in the graph data structure contains relevant information of each traffic participant; a probability acquisition unit for inputting the graph data structure into a neural network model to obtain a probability distribution of collisions between each node in the graph data structure and other nodes; and an adjustment unit for adjusting the driving strategy of at least one of the multiple traffic participants based on the probability distribution of collisions between each node in the graph data structure and other nodes.

[0015] According to another aspect of this disclosure, a dynamic scheduling system is provided, comprising: a memory; and a processor coupled to the memory, the processor being configured to execute the dynamic scheduling method as described above based on instructions stored in the memory.

[0016] According to another aspect of this disclosure, a computer-readable storage medium is provided having computer instructions stored thereon that, when executed by a processor, implement the dynamic scheduling method as described above.

[0017] According to another aspect of this disclosure, a computer program is provided, comprising: instructions that, when executed by a processor, cause the processor to perform the dynamic scheduling method as described above.

[0018] Other features and advantages of this disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0019] The accompanying drawings, which form part of this specification, illustrate embodiments of this disclosure and, together with the specification, serve to explain the principles of this disclosure.

[0020] This disclosure will become clearer with reference to the accompanying drawings and the following detailed description, wherein:

[0021] Figure 1 is a flowchart illustrating a dynamic scheduling method according to some embodiments of the present disclosure;

[0022] Figure 2 is a schematic diagram showing a site map of a mining area according to some embodiments of the present disclosure;

[0023] Figure 3 is a schematic diagram showing a site map of a mining area according to some other embodiments of the present disclosure;

[0024] Figure 4 is a schematic diagram illustrating the diagram structure according to some embodiments of the present disclosure;

[0025] Figure 5 is a schematic diagram illustrating the diagram structure according to some other embodiments of the present disclosure;

[0026] Figure 6 is a schematic diagram illustrating the graph convolutional network extraction of relationships according to some embodiments of the present disclosure;

[0027] Figure 7 is a schematic block diagram illustrating the structure of a dynamic scheduling system according to some embodiments of the present disclosure;

[0028] Figure 8 is a schematic block diagram illustrating the structure of a dynamic scheduling system according to other embodiments of the present disclosure;

[0029] Figure 9 is a schematic block diagram illustrating the structure of a dynamic scheduling system according to other embodiments of the present disclosure. Detailed Implementation

[0030] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the present disclosure.

[0031] At the same time, it should be understood that, for ease of description, the dimensions of the various parts shown in the accompanying drawings are not drawn according to actual scale.

[0032] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit this disclosure or its application or use.

[0033] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0034] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0035] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0036] The complex traffic environment of open-pit mines, including varied terrain, dense transport vehicles, and frequent overlapping operations, poses a significant challenge to the stable operation of unmanned driving systems. The inventors of this disclosure have discovered that, in related technologies, static planning-based transportation scheduling systems struggle to adapt to these dynamic changes, especially in the face of emergencies or urgent tasks, often leading to substantial safety hazards.

[0037] In view of this, embodiments of the present disclosure provide a dynamic scheduling method to reduce safety hazards during transportation in open-pit mines.

[0038] Figure 1 is a flowchart illustrating a dynamic scheduling method according to some embodiments of the present disclosure.

[0039] In step S102, the road conditions and environmental parameters of the open-pit mine, as well as the operation data and performance parameters of multiple traffic participants, are obtained.

[0040] For example, traffic participants may include mining trucks and / or electric shovels (i.e., electric excavators). As another example, traffic participants may also include water trucks or pedestrians.

[0041] In some embodiments, road conditions include: traffic conditions and speed limit information for multiple roads in the open-pit mine. For example, whether any road in the open-pit mine is impassable due to weather conditions, or whether any road requires an adjustment to the maximum speed of mining trucks due to weather conditions.

[0042] In some embodiments, environmental parameters include weather condition parameters for the open-pit mine. Weather condition parameters may include conditions such as sunny, sandstorm, fog, rain, and snow. For example, reduced visibility due to sandstorms or fog may necessitate vehicle speed reduction.

[0043] In some embodiments, the operational data includes the current location, speed, acceleration, and load information of each mining truck. For example, the number of available mining trucks of different models, their load capacity, and the real-time location, speed, and acceleration of the mining trucks can be obtained.

[0044] In some embodiments, performance parameters include the unloaded speed and the fully loaded speed of each mining truck. For example, the unloaded and fully loaded speeds of different models of mining trucks can be obtained.

[0045] In other embodiments, the number of transfer and unloading points in the mining area on that day can also be obtained, as well as the working status and efficiency of the electric shovels. For example, it can be obtained which electric shovels are available and the time required for the electric shovels to fill different types of mining trucks.

[0046] In other embodiments, the above method may further include: obtaining the mineral reserves and production plan for the day to determine the task priorities for the day.

[0047] Assuming a mining company's main product is iron ore, the following is the relevant data for that day:

[0048] Mineral reserves: (1) High-grade iron ore reserves: 1 million tons; (2) Medium-grade iron ore reserves: 1.5 million tons; (3) Low-grade iron ore reserves: 2 million tons.

[0049] Production plan: (1) 500,000 tons of high-grade iron ore needs to be produced to meet the needs of high-value customers; (2) 700,000 tons of medium-grade iron ore needs to be produced for general market demand; (3) 300,000 tons of low-grade iron ore needs to be produced for low-value market demand or as a by-product.

[0050] Based on the data above, the following are the steps to determine task priorities:

[0051] Step 1: Analyze market demand and profit margins. For example, high-grade iron ore has strong market demand, high prices, and the highest profit margins, so it has the highest priority; medium-grade iron ore has stable market demand, moderate prices, and medium profit margins, so it has the next highest priority; low-grade iron ore has low market demand, low prices, and the lowest profit margins, so it has the lowest priority.

[0052] Step 2: Consider reserves and production targets. For example, high-grade iron ore reserves are relatively small, but production targets are high, so production needs to be prioritized; medium-grade iron ore reserves and production targets are relatively balanced, so production can be carried out according to plan; low-grade iron ore reserves are sufficient, but production targets are low, so production can be carried out on the premise of ensuring the production of other grades of ore.

[0053] Step 3: Determine the task priorities for the day. For example, prioritize the production of high-grade iron ore to ensure the needs of high-value customers are met; secondly, ensure the production of medium-grade iron ore to meet general market demand; and finally, schedule the production of low-grade iron ore based on remaining resources and time.

[0054] In step S104, a graph data structure is generated based on the road conditions and environmental parameters of the open-pit mine, as well as the operational data and performance parameters of multiple traffic participants. Here, each node in the graph data structure contains relevant information for each traffic participant.

[0055] In some embodiments, step S104 includes: generating path points of the trajectories of multiple traffic participants based on road conditions and environmental parameters of the open-pit mine and the operation data and performance parameters of multiple traffic participants; determining the path relationships between different path points based on the path points of the trajectories of multiple traffic participants; and obtaining a graph data structure based on the path points of the trajectories of multiple traffic participants and the path relationships between different path points, wherein the path relationships are the edges between different nodes of the graph data structure.

[0056] For example, the process of generating graph data structures is described in detail below with reference to Figures 2 to 5.

[0057] For example, as shown in Figure 2, assume there are L loading areas and UL unloading areas (L and UL are positive integers) in the mining area. Loading area A is equipped with m electric shovels, and loading area B is equipped with n electric shovels (m and n are positive integers). Parking areas are provided for charging and maintenance of mining trucks and electric shovels, with a total of T idle mining trucks (T is a positive integer). These mining trucks are numbered according to their load capacity. For example, T1-1 represents mining truck No. 1 with a load capacity of 100 tons, T1-2 represents mining truck No. 2 with a load capacity of 100 tons, T2-1 corresponds to mining truck No. 1 with a load capacity of 200 tons, and so on. As another example, S represents the electric shovels used for shoveling materials, and each electric shovel has a unique number. As shown in Figure 3, the arrowed lines in the figure represent roads between different locations. When mining trucks are fully charged and start their day's work, they mainly move along predetermined routes. The routes marked with solid lines have a higher risk of collisions due to frequent use, while the routes marked with dashed lines are less frequently chosen.

[0058] Next, as shown in Figure 4, the planned trajectories are set for the mining trucks T1-1, T1-2, T1-3, T1-4, T2-1, T2-2, T2-3 and T2-4 at a certain moment, and the predicted trajectory is set for the electric shovel S4.

[0059] As described above, multiple traffic participants can include multiple mining trucks. For example, based on the road conditions and environmental parameters of the open-pit mine, as well as the operational data and performance parameters of multiple traffic participants, waypoints for the trajectories of multiple traffic participants are generated. This includes: generating the task route for each mining truck based on its scheduling task (e.g., a dynamic scheduling task); generating the expected speed (i.e., the expected actual speed) for each mining truck based on its load information; and generating waypoints for the planned trajectory of each mining truck based on its task route, expected speed, and speed limit information for different roads. Thus, by generating these waypoints, the planned trajectory of each mining truck can be generated.

[0060] For example, in the process of planning the trajectory, the tasks and constraints can be clearly defined, as described in (1) to (3) below.

[0061] (1) Determine the dynamic scheduling task. The path information generated by the dynamic scheduling system can be obtained, which usually includes a series of waypoints or road segments, as well as related task requirements (e.g., arrival time, load limit, etc.).

[0062] (2) Determine road speed limits. Speed ​​limit information for different roads can be obtained based on electronic maps or real-time traffic information.

[0063] (3) Determine the load and expected speed. Based on the vehicle's load and dynamic performance, the actual driving speed of the vehicle on different road sections can be expected. Dynamic parameters such as vehicle acceleration and deceleration can be used as a reference.

[0064] For example, the trajectory planning process includes the following steps a1 to a4.

[0065] In step a1, the path is refined. The path generated by the dynamic scheduling task is refined into a series of specific road segments, each with a clear start point, end point, road speed limit, and expected speed.

[0066] In step a2, speed planning. On each road segment, based on the road speed limit and the expected speed, a target speed curve for the vehicle on that segment is determined. This can be referenced to the vehicle's rate of acceleration change to avoid sudden acceleration and deceleration, thereby improving ride comfort.

[0067] In step a3, time planning, the estimated travel time for the vehicle on each road segment can be calculated based on the target speed curve and segment length. This helps ensure that the vehicle arrives on time as required by the mission.

[0068] In step a4, trajectory generation. The results of velocity planning and time planning are combined to generate the planned trajectory of the vehicle. This can include a series of position, velocity, and time information, which can be represented as a continuous curve or a series of discrete points.

[0069] Thus, trajectory planning was performed through the steps a1 to a4 described above.

[0070] Furthermore, multiple traffic participants may also include one or more electric shovels. Operational data includes the speed, orientation, and historical location of one or more electric shovels. For example, based on the road conditions and environmental parameters of the open-pit mine, as well as the operational data and performance parameters of multiple traffic participants, waypoints for the trajectories of multiple traffic participants are generated. This also includes generating the future trajectories of one or more electric shovels based on their speed, orientation, and historical location using a Kalman filter algorithm. In other words, the predicted trajectory of the electric shovel (e.g., electric shovel S4) is a preliminary future trajectory generated using Kalman filtering based on the shovel's speed, orientation, and historical location.

[0071] Here, the Kalman filter algorithm is a recursive state estimation method that updates the optimal estimate of the state using the system model and measurements. In a linear dynamic system, the system's state transitions and observation models are linear, and both system noise and observation noise are Gaussian distributed and independent of each other. The Kalman filter algorithm is known to those skilled in the art and will not be described in detail here.

[0072] As shown in Figure 4, assume that the planned trajectory data of each traffic participant can be represented as a sequence of coordinate points: P ij ={(x i1 ,y i1 ),(x i2 ,y i2 ),…,(x ij ,y ij ),…,(x im ,y im )}, i=1,2,…,n, (1)

[0073] Among them, P ij Let x represent the waypoint of the i-th traffic participant at time j (i.e., the j-th waypoint of the i-th traffic participant), (x ij ,y ij Let represent the coordinates of the i-th traffic participant at time j, where i, j, and n are positive integers.

[0074] In addition, task priorities are set for the current task of each mining truck according to the task plan for the day. For example, if the demand for ore in unloading area A is more urgent, then the task to unloading area A has a relatively higher priority. The initial path points of the mining trucks are extracted into a graph structure, as shown in Figure 5. The data of each node in the graph contains the following information: H={ID,P,Pr,V,A,L,T}, (2)

[0075] Wherein, ID is the identifier of each traffic participant (e.g., mining trucks or other dynamic traffic participants) (i.e., the identifier is a unique identifier for each traffic participant), P is the coordinate of each waypoint of each traffic participant, Pr is the priority of the current task of each traffic participant, V is the current speed information of each traffic participant, A is the current acceleration information of each traffic participant, L is the current load information of each traffic participant, and T is the estimated time for each traffic participant to arrive at each node (i.e., the timestamp corresponding to each node).

[0076] The planned trajectory of the mining truck can be represented as a graph data structure (i.e., graph structure) G = (H, E), where H represents the set of nodes in the graph, that is, each coordinate point is considered a node in the graph, and E represents the set of edges in the graph, that is, the edges between nodes, which represent the path relationships between nodes. For example, if the first node and the second node are adjacent on the planned trajectory of the mining truck, then there is an edge between them. For example, the graph structure can be represented by an n×n adjacency matrix A, where if there is an edge from node i to node j, then A... ij =1, if there is no edge from node i to node j, then A ij=0. Therefore, the adjacency matrix A is a matrix consisting of 0s and 1s.

[0077] Each node represents a discrete coordinate point in the planned trajectory of a mining truck or other traffic participant. Specifically, the coordinate position of each node can be represented as (x... ij ,y ij ), where x ij and y ij These represent the x-coordinate and y-coordinate of the node, respectively.

[0078] Thus, each node in the graph data structure contains relevant information about each traffic participant. This information includes: the participant's identifier, the coordinates of each waypoint, the priority of the participant's current task, the participant's current speed, acceleration, load, the estimated time for each participant to reach each node (i.e., the timestamp corresponding to each node), and the path relationships between different waypoints.

[0079] In step S106, the graph data structure is input into the neural network model to obtain the probability distribution of each node in the graph data structure colliding with other nodes.

[0080] In some embodiments, step S106 includes: processing the graph data structure sequentially through a graph convolutional network, a pooling layer, and a connection layer in a neural network model to obtain each node and other nodes that may collide with each node; inputting each node and other nodes that may collide with each node into a feedforward neural network of the neural network model to output a probability distribution vector of each node colliding with other nodes; and calculating the probability distribution of each node colliding with other nodes by using a normalized exponential function based on the probability distribution vector of each node colliding with other nodes.

[0081] In practical applications, there are numerous traffic participants, and the nodes representing paths contain many elements. Furthermore, depending on the number of traffic participants, each scheduling task can have different graph structures at different times. Therefore, Graph Convolutional Networks (GCNs) can be used to learn and propagate node features. Through multiple layers of graph convolutional operations, node features are gradually aggregated and updated, thereby capturing the relationships and dynamic information between traffic participants in subsequent processing. GCNs learn the feature representation of each node, which includes information about the node itself and its relationships with neighboring nodes. These features can be used to predict the collision probability between traffic participants. Here, graph convolutional networks are a technique known to those skilled in the art.

[0082] Graph convolutional networks (GCNNs) are an adapted version of convolutional neural networks for graph encoding, used for tasks involving graph data, such as node classification, graph classification, and link prediction. In embodiments of this disclosure, GCNNs can be used to predict link relationships, which, after further processing, can then be used to predict collisions between different traffic participants.

[0083] For example, given a graph with n nodes, its structure can be represented by an n×n adjacency matrix A. If there is an edge from node i to node j, then A... ij =1. In an L-layer GCN (where L is a positive integer), if the input vector of the i-th node in the l-th layer is represented as... The output vector is represented as The graph convolution operation can then be written as:

[0084] Among them, W (l) Let b be a linear transformation matrix. (l) σ is the bias term, and σ is a nonlinear function (e.g., the ReLU function). W (l) and b (l) The parameters that the model learns and adjusts automatically during training.

[0085] In some cases, data can be normalized before being fed into each non-linear layer of a graph convolutional network, and a self-loop can be added to each node in the graph.

[0086] in, I is an n×n identity matrix. It is the degree of node i in the graph.

[0087] By superimposing this operation on L layers, a deep GCN network can be obtained, as shown in Figure 6. To represent the vector input to the GCN network, and using This represents the vector output from the GCN network.

[0088] Let H = [h1,…,h] n ] represents a path, where h i Let H be the i-th path point, and let it correspond to two intervals in the path: s =[h s1 ,…,h sr ] and H o =[h o1 ,…,h ou As shown in Figure 6, h sent For any node among all nodes, Hs represents the node connected to node h. sentFor the preceding set of nodes, Ho can represent the node h. sent The subsequent nodes, all nodes within the intervals Hs and Ho, are related to node h. sent Nodes where collisions may occur.

[0089] Thus, as shown in Figure 6, the graph data structure is processed sequentially through a Graph Convolutional Network (GCN), pooling layers, and concatenation layers (e.g., fully connected layers) in a neural network model to obtain each node and other nodes that may collide with each node. In the neural network model, the goal of relation extraction is to predict whether the relationship between entities is r∈R (where R is a predefined set of relations, i.e., conflicting) or "no relation".

[0090] In Figure 6, the left side shows the overall architecture, and the right side shows the detailed graph convolution computation of the path points. After applying an L-layer GCN to the path vectors, a hidden representation of each collision conflict relationship can be obtained. These collision conflict relationships are directly affected by the path points of several edges in the graph structure. To utilize these representations to extract relationships, the following sentence representation can be obtained (see the left side of Figure 6): h sent =f(h (L) )=f"GCN(h (0) (5)

[0091] Among them, h (l) Let h be the implicit representation of the output vector of the l-th layer of the GCN, and f be the max pooling function that maps the n output vectors to the relation vector. Path points that are closer together in the graph are usually the core of relation extraction. Therefore, as shown below, from h... (l) The corresponding representation h was obtained s The corresponding representation h can be obtained in a similar way. o .

[0092] Representations for classification are obtained by connecting path points, and these are then fed into a feedforward neural network (FFNN): h final =FFNN([h sent h s h o (7)

[0093] Then this h final Finally, the input is a linear layer, and then the Softmax function (i.e., the normalized exponential function) is used to perform calculations to obtain the probability distribution of the relationship, which is the probability distribution of conflicts between different nodes.

[0094] In other words, each node and other nodes that may collide with each node are input into the feedforward neural network of the neural network model to output the probability distribution vector of each node colliding with other nodes. Then, based on the probability distribution vector of each node colliding with other nodes, the probability distribution of each node colliding with other nodes is calculated by the normalized exponential function (Softmax function).

[0095] Here, the probability distribution vector is a vector whose elements are real numbers. The Softmax function is a mathematical function mainly used to convert a set of arbitrary real numbers into real numbers representing a probability distribution. Its core function is to map a set of real values ​​to the interval (0,1) and ensure that the sum of these values ​​is 1, which can be interpreted as a probability distribution. Here, the Softmax function calculates the probability value of each node colliding with other nodes, and these probability values ​​form the probability distribution.

[0096] In the GCN model, hidden layers with multiple nodes (e.g., 200 nodes) are used as output feedforward layers. For example, in the experiments, two GCN layers and two feedforward layers (FFNN) were used. ReLU was used as the activation function for all nonlinear layers in the GCN, and standard max pooling was used for all pooling layers.

[0097] For training, stochastic gradient descent is used with an initial learning rate of 1.0 and an exponentially decaying learning rate Lr of 0.001. Based on the learned node feature representations, a regressor can be used to predict the probability of collisions.

[0098] In step S108, the driving strategy of at least one of the multiple traffic participants is adjusted based on the probability distribution of each node in the graph data structure colliding with other nodes.

[0099] In some embodiments, step S108 includes: adjusting the driving strategy of at least one traffic participant based on the probability distribution of each node in the graph data structure colliding with other nodes, so as to minimize the probability of at least one traffic participant colliding with other traffic participants among multiple traffic participants and maximize the benefit of at least one traffic participant.

[0100] For example, for graph structures with a high probability of collision, ensuring that high-priority scheduling tasks remain unchanged can be achieved through several adjustment strategies to avoid collisions. For example, (1) adjust the scheduling tasks of mining trucks with low task priorities (e.g., mining trucks with task priorities below the priority threshold, or mining trucks with lower priorities than other mining trucks) to change the planned trajectory of the mining truck, such as delaying its departure time. Another example is to make the low-priority mining truck wait in place and calculate its waiting time. Yet another example is to adjust its planned path if there are alternative routes available, and calculate the additional time required for the extra route it travels.

[0101] When the graph structure has no collision risk and is relatively sparse, the scheduling task of mining trucks can be increased to improve the overall work efficiency. For example, as shown in Figure 4, there are 3 electric shovels in loading area A. When mining truck T1-2 arrives at loading area A, if each mining truck needs 1 electric shovel to load its materials, and there are idle mining trucks at the parking point, then the scheduling task of mining truck T1-2 will be increased on the path of mining truck T1-2.

[0102] In some embodiments, after each dynamic adjustment of the scheduling task, the graph structure can be regenerated, and several graph structures at different times can be generated each time. For example, five graph structures can be generated each time, one for the current time, one for 10 seconds later, one for 20 seconds later, one for 30 seconds later, and one for 40 seconds later. The process of step S106 is repeated until the driving strategy with the lowest risk (e.g., no risk) and the highest benefit is obtained, that is, the final scheduling scheme is obtained.

[0103] The aforementioned at least one traffic participant includes at least one mining truck.

[0104] In some embodiments, the revenue of the at least one traffic participant is calculated based on the total load or total unload of each of the at least one mining trucks from the loading point to the unloading point, the distance of each of the at least one mining trucks from the loading point to the unloading point, the energy consumption per unit distance when each of the at least one mining trucks is fully loaded, the waiting time of each of the at least one mining trucks, and the energy consumption per unit time when each of the at least one mining trucks is waiting.

[0105] For example, after each new scheduling task is generated, calculate the total revenue E of t (t is a positive integer) mining trucks. total for

[0106] in, Let be the total cargo load or total unload of the k-th mining truck from loading point i to unloading point j. Let Q be the distance from loading point i to unloading point j for the k-th mining truck. f,kLet t be the energy consumption per kilometer (i.e., energy consumption per unit distance) of the kth mining truck when fully loaded. k Let Q be the waiting time for the kth mining truck. w,k Let be the energy consumption per unit time of the k-th mining truck while it is waiting, where k is a positive integer. Here, t k ·Q w,k This represents the energy loss of the k-th mining truck due to waiting. It should be noted that these parameters can vary depending on the type of the k-th mining truck, its load capacity, the loading task, and whether waiting is required.

[0107] In specific scheduling scenarios for mining truck operations, given that scheduling tasks often have highly defined priorities and execution objectives, when the collision risk introduced by a new scheduling task is at a low threshold and the expected benefits are maximized, a dynamic scheduling strategy combined with refined trajectory prediction technology can be used to achieve collaborative optimization. This process fully considers the unique nature of mining truck operations: compared to general commercial vehicle environments, mining trucks enjoy higher road usage priority within mining areas, leading other traffic participants to be more inclined to take proactive avoidance measures when faced with potential path conflicts, such as planning detours in advance or waiting in place, to ensure the uninterrupted and efficient operation of mining trucks. Therefore, by utilizing the collaborative prediction mechanism of scheduling tasks to accurately predict and optimize the travel trajectories of other traffic participants, precise adjustments to detour or avoidance strategies can be achieved.

[0108] This provides a dynamic scheduling method according to some embodiments of the present disclosure. The dynamic scheduling method includes: acquiring road conditions and environmental parameters of an open-pit mine, as well as operational data and performance parameters of multiple traffic participants; generating a graph data structure based on the road conditions and environmental parameters of the open-pit mine, and the operational data and performance parameters of multiple traffic participants, wherein each node in the graph data structure contains relevant information about each traffic participant; inputting the graph data structure into a neural network model to obtain the probability distribution of collisions between each node in the graph data structure and other nodes; and adjusting the driving strategy of at least one of the multiple traffic participants based on the probability distribution of collisions between each node in the graph data structure and other nodes. This dynamic scheduling method can reduce safety hazards and improve transportation efficiency.

[0109] Dynamic task scheduling, a key technology for resolving traffic congestion and optimizing resource allocation in open-pit mines, can dynamically adjust the allocation of different vehicles and time periods based on real-time traffic conditions, operational needs, and safety requirements. This mechanism not only effectively alleviates traffic pressure and reduces vehicle waiting time but also allows for flexible scheduling based on task priority, ensuring timely execution of critical operations. Crucially, dynamic scheduling management also fully considers the real-time location and movement status of other traffic participants (such as electric shovels and other mining trucks), reducing conflicts and interference among them through intelligent decision-making, thereby significantly improving the flexibility and efficiency of task scheduling.

[0110] In the complex and ever-changing traffic environment of open-pit mines, accurately predicting the movement trajectories of other traffic participants is crucial for efficient and safe transportation. Trajectory prediction methods in related technologies are often based on fixed models or historical data, making it difficult to cope with frequently changing factors in real-world environments. However, the trajectory prediction model in this embodiment, which incorporates dynamic scheduling management, can dynamically adjust the parameters of the prediction model based on real-time updated right-of-way data, making the prediction results closer to reality. For example, when a road is congested due to an urgent scheduling task, the probability of other traffic participants frequently using that road decreases. Therefore, the trajectory prediction takes this change into account, improving prediction accuracy.

[0111] Furthermore, the mutual influence among traffic participants is also a factor in trajectory prediction. Dynamic scheduling management achieves collaborative prediction and optimization by comprehensively considering the movement trends, speed changes, and potential conflict points of each traffic participant. This collaborative mechanism can not only provide early warnings of potential collision risks but also maximize the utilization of road resources through intelligent scheduling, thereby improving the overall efficiency and safety of mine transportation.

[0112] Therefore, the application of dynamic scheduling management and trajectory collaborative prediction methods provides strong support for unmanned driving technology in open-pit mines. Through real-time perception, intelligent decision-making, and collaborative optimization, the aforementioned dynamic scheduling methods can solve as many of the challenges faced by transportation scheduling systems in related technologies as possible, injecting new vitality into the sustainable development of open-pit mines.

[0113] The embodiments of this disclosure provide a dynamic scheduling and trajectory collaborative prediction method for open-pit mines. By combining dynamic scheduling algorithms and intelligent trajectory prediction algorithms, it achieves intelligent scheduling and collision prevention of mining trucks, maximizing the safety, efficiency, and cost-effectiveness of automated operations in mining scenarios. In this method, the probability of collisions is predicted based on the planned trajectories of mining trucks, improving operational safety; operational priorities are set according to the daily ore material plan, improving operational efficiency; unified pre-scheduling of all mining trucks provides a fully unmanned operation system, saving human resources; dynamically adjusting the planned trajectories over time further enhances the rationality and timeliness of decision-making; and incorporating trajectory predictions of other participants into the scheduling task further improves prediction accuracy.

[0114] Unlike other vehicle types, the method in this disclosure is designed for autonomous driving of mining trucks in open-pit mine scenarios. For mining trucks, ensuring the total transport volume of all mining trucks is a hard target. Therefore, the tasks of mining trucks can be prioritized according to the demand for ore.

[0115] In some embodiments of the present disclosure, determining whether a mining truck scheduling task is reasonable and whether there are conflicts in the planned trajectory requires not only considering two-dimensional trajectory coordinates but also multiple information dimensions such as time. When there are many mining trucks and the task is complex, the prediction accuracy is higher when using a neural network.

[0116] When dynamically adjusting scheduling tasks, the core indicator should still be "maximizing the total transportation volume." For scheduling tasks with a high probability of collision, a refined scheduling adjustment strategy can be adopted: while ensuring the stable execution of high-priority scheduling tasks, the operation schedule of low-priority mining trucks can be dynamically adjusted through intelligent algorithms. Specific strategies include: first, flexibly adjusting the scheduling sequence of low-priority mining trucks, such as appropriately delaying their departure time to avoid peak hours; second, implementing a waiting-in-place strategy, accurately calculating waiting time to minimize the impact on the overall transportation plan; and third, using path optimization algorithms to find and plan alternative, low-conflict-risk routes for low-priority mining trucks, while comprehensively considering additional time and safety to ensure that the adjusted plan is both efficient and safe. Through such dynamic adjustments, the transportation efficiency of mining trucks can be maximized while ensuring safety, thus meeting the total volume requirements for mineral transportation.

[0117] Furthermore, when predicting the future trajectories of other traffic participants, the interactive effects of mining truck scheduling tasks can be considered. By monitoring and analyzing mining truck scheduling tasks and future planned routes in real time, the future trajectories of other traffic participants can be accurately predicted and optimized, improving the accuracy of the prediction model and thus achieving intelligent coordination and efficient management of global traffic flow. This design significantly enhances the robustness and practicality of the system, providing strong support for the intelligent and refined management of traffic in open-pit mines.

[0118] Figure 7 is a schematic structural block diagram illustrating a dynamic scheduling system according to some embodiments of the present disclosure. As shown in Figure 7, the dynamic scheduling system includes: a data acquisition unit 702, a generation unit 704, a probability acquisition unit 706, and an adjustment unit 708.

[0119] The data acquisition unit 702 is used to acquire road conditions and environmental parameters of the open-pit mine, as well as the operation data and performance parameters of multiple traffic participants.

[0120] In some embodiments, road conditions include: traffic conditions and speed limit information of multiple roads in the open-pit mine; environmental parameters include: weather condition parameters of the open-pit mine; operational data include: current location information, speed information, acceleration information and load information of each mining truck; performance parameters include: the driving speed of each mining truck when unloaded and the driving speed when fully loaded.

[0121] The generation unit 704 is used to generate a graph data structure based on the road conditions and environmental parameters of the open-pit mine, as well as the operational data and performance parameters of multiple traffic participants. Each node in the graph data structure contains relevant information for each traffic participant.

[0122] In some embodiments, the generation unit 704 is used to generate path points of the trajectories of multiple traffic participants based on the road conditions and environmental parameters of the open-pit mine and the operation data and performance parameters of multiple traffic participants; determine the path relationships between different path points based on the path points of the trajectories of multiple traffic participants; and obtain a graph data structure based on the path points of the trajectories of multiple traffic participants and the path relationships between different path points, wherein the path relationships are the edges between different nodes of the graph data structure.

[0123] In some embodiments, the multiple traffic participants include multiple mining trucks.

[0124] In some embodiments, the generation unit 704 is used to generate a task route for each mining truck based on the scheduling task of each mining truck, generate an expected speed for each mining truck based on the load information of each mining truck, and generate waypoints for the planned trajectory of each mining truck based on the task route, expected speed and speed limit information of different roads.

[0125] In some embodiments, the multiple traffic participants may also include one or more electric shovels.

[0126] In some embodiments, operational data includes the speed, orientation, and historical location of one or more electric shovels.

[0127] In some embodiments, the generation unit 704 is used to generate the future trajectory of one or more electric shovels based on the speed, orientation, and historical position of one or more electric shovels using a Kalman filter algorithm.

[0128] In some embodiments, the relevant information for each traffic participant includes: the identifier of each traffic participant, the coordinates of each waypoint of each traffic participant, the priority of each traffic participant's current task, the current speed information of each traffic participant, the current acceleration information of each traffic participant, the current load information of each traffic participant, the estimated time for each traffic participant to arrive at each node (i.e., the timestamp corresponding to each node), and the path relationship between different waypoints.

[0129] The probability acquisition unit 706 is used to input the graph data structure into the neural network model to obtain the probability distribution of each node in the graph data structure colliding with other nodes.

[0130] In some embodiments, the probability acquisition unit 706 is used to process the graph data structure sequentially through the graph convolutional network, pooling layer and connection layer in the neural network model to obtain each node and other nodes that may collide with each node. Each node and other nodes that may collide with each node are input into the feedforward neural network of the neural network model to output the probability distribution vector of each node colliding with other nodes. Based on the probability distribution vector of each node colliding with other nodes, the probability distribution of each node colliding with other nodes is calculated by a normalized exponential function.

[0131] The neural network model disclosed herein includes a graph convolutional network (GCN), which may include an input layer, a graph convolutional layer, an activation function, etc.

[0132] (1) Input layer

[0133] The input to GCN is a graph consisting of nodes and edges between them. Nodes represent entities in the graph, and edges represent the connections between nodes. Each node has a feature vector representing its characteristic information. The graph can be represented by an adjacency matrix, where each element represents the connection between nodes.

[0134] (2) Graph Convolutional Layer

[0135] The core of GCN is the graph convolutional layer, which updates the feature representation of a node by aggregating information from its neighbors. In each graph convolutional operation, the feature vector of a node is weighted and summed with the feature vectors of its neighbors to obtain the aggregated feature vector. This process can be represented by matrix multiplication.

[0136] (3) Activation function

[0137] After each graph convolutional layer, a nonlinear activation function can be applied to introduce a nonlinear transformation. Commonly used activation functions include ReLU, Sigmoid, and tanh. Activation functions can increase the network's expressive power and nonlinear fitting ability.

[0138] Generative Networks (GCNs) typically consist of multiple graph convolutional layers, each updating the feature representations of nodes. Through the stacking of multiple layers, GCNs can gradually capture deeper levels of graph structure information. The training process for GCNs usually employs supervised learning, optimizing network parameters by minimizing a loss function. Commonly used loss functions include cross-entropy loss and mean squared error loss.

[0139] The neural network model disclosed herein also includes pooling layers. Pooling layers are used to aggregate and reduce the overall features of the graph. Commonly used pooling methods include max pooling and average pooling for graphs.

[0140] Furthermore, the neural network model disclosed herein also includes an output layer for generating the final prediction result. The form of the output layer depends on the specific task and can be a classifier, regressor, or generator. For example, a regressor can be used to predict the probability of a collision.

[0141] As mentioned earlier, GCN is a deep learning model that utilizes graph structures for information propagation and feature learning. Through multi-layer graph convolution operations, the neural network model can easily capture the relationships and dynamic information between nodes, effectively predicting the probability of collisions.

[0142] The adjustment unit 708 is used to adjust the driving strategy of at least one of multiple traffic participants based on the probability distribution of each node colliding with other nodes in the graph data structure.

[0143] In some embodiments, the adjustment unit 708 is used to adjust the driving strategy of at least one traffic participant based on the probability distribution of each node in the graph data structure colliding with other nodes, so as to minimize the probability of at least one traffic participant colliding with other traffic participants among multiple traffic participants and maximize the benefit of at least one traffic participant.

[0144] In some embodiments, at least one traffic participant includes at least one mining truck.

[0145] In some embodiments, the revenue of at least one traffic participant is calculated based on the total load or total unload of each of the at least one mining trucks from the loading point to the unloading point, the distance of each of the at least one mining trucks from the loading point to the unloading point, the energy consumption per unit distance when each of the at least one mining trucks is fully loaded, the waiting time of each of the at least one mining trucks, and the energy consumption per unit time when each of the at least one mining trucks is waiting.

[0146] Thus, a dynamic scheduling system according to some embodiments of the present disclosure is provided. In the dynamic scheduling system, a data acquisition unit is used to acquire road conditions and environmental parameters of the open-pit mine, as well as the operating data and performance parameters of multiple traffic participants; a generation unit is used to generate a graph data structure based on the road conditions and environmental parameters of the open-pit mine, as well as the operating data and performance parameters of multiple traffic participants, wherein each node in the graph data structure contains relevant information about each traffic participant; a probability acquisition unit is used to input the graph data structure into a neural network model to obtain the probability distribution of collisions between each node in the graph data structure and other nodes; and an adjustment unit is used to adjust the driving strategy of at least one of the multiple traffic participants based on the probability distribution of collisions between each node in the graph data structure and other nodes. This dynamic scheduling system can reduce safety hazards and improve transportation efficiency.

[0147] Embodiments of this disclosure provide a dynamic scheduling method and system that enables collaborative prediction of dynamic right-of-way and trajectory in open-pit mines.

[0148] First, the planned trajectory of each mining truck and the predicted trajectory of other traffic participants are transformed into a graph data structure, where each node represents a discrete coordinate point on the path, providing a solid data foundation for subsequent collision and conflict prediction.

[0149] Secondly, the aforementioned method and system cleverly utilize graph convolutional networks to deeply learn the complex relationships between mining trucks and between mining trucks and traffic participants, thereby accurately predicting potential path conflicts. This method fully leverages the inherent characteristics of graph data, significantly improving the accuracy and efficiency of path conflict prediction, and providing strong protection for the safe operation of mining trucks.

[0150] Furthermore, the aforementioned methods and systems propose a series of intelligent dynamic scheduling strategies to flexibly respond to complex and ever-changing mining operation environments. When allocating scheduling tasks, factors such as the number of electric shovels, waiting time, collision risk, path length, and scheduling efficiency are considered. For scheduling tasks with high collision risk, the strategies ensure the stable execution of high-priority tasks and take multiple measures to avoid conflicts: first, adjusting the scheduling of low-priority mining trucks, such as delaying their departure; second, having low-priority vehicles wait in place to optimize waiting time; and third, exploring alternative routes to balance additional time consumption and safety. In sparse environments without collision risk, scheduling tasks can be intelligently increased to maximize resource utilization and improve overall operational efficiency. These strategies collectively constitute an efficient and safe dynamic scheduling system for mining operations.

[0151] Furthermore, the aforementioned methods and systems can integrate an interactive prediction model of the dispatching trajectory of mining trucks with that of other traffic participants. By monitoring and analyzing the driving paths and speeds of mining trucks in real time, the future trajectories of other traffic participants can be accurately predicted and optimized. This model not only considers the impact of mining trucks as the dominant factor on other vehicles or equipment, but also inversely assesses the feedback effect of the potential behaviors of other traffic participants on the dispatching efficiency and safety of mining trucks, thereby achieving intelligent coordination and efficient management of global traffic flow. This design can improve traffic flow and safety in mining operation areas, providing solid technical support for the automation and intelligent transformation of mines.

[0152] Furthermore, the aforementioned methods and systems also take into account the impact of ore demand on operational priorities. By rationally setting operational priorities, it ensures that high-demand ore can be mined, loaded, and transported first, thereby effectively meeting the immediate needs of the production line, reducing inventory backlog, and improving the responsiveness and flexibility of the entire mining operation chain.

[0153] In summary, the embodiments of this disclosure construct a comprehensive and intelligent mine operation scheduling and traffic management system. This system deeply integrates advanced data processing, graph neural network analysis, intelligent scheduling strategies, and dynamic traffic prediction models, achieving comprehensive optimization from mining truck scheduling to overall traffic management. This system can not only accurately predict and reduce potential conflicts between mining trucks and other traffic participants, ensuring operational safety, but also flexibly adjust operational priorities according to ore demand, improving resource utilization efficiency and overall operational efficiency.

[0154] The graph structure uses discrete points representing the planned trajectories of mining trucks and the predicted trajectories of other traffic participants as nodes. For example, the planned trajectory points are trajectory data for the next 7 seconds, sampled at a frequency of 10Hz, resulting in 70 nodes. Supervised training is performed using a Geographic Network (GCN). The GCN updates the feature representations of nodes by aggregating information from local neighbor nodes on the graph, thereby leveraging the relationships between nodes for information propagation and feature learning.

[0155] The embodiments of this disclosure provide a method and system for dynamic scheduling and trajectory collaborative prediction in open-pit mines, which ensures that mining trucks can operate safely while improving operational efficiency and saving resources.

[0156] For example, multi-sensor fusion can improve the accuracy and robustness of detecting other traffic participants. By combining multiple sensors such as GPS (Global Positioning System), LiDAR, millimeter-wave radar, cameras, and V2X (vehicle-to-everything) technology, high-precision positioning and real-time perception of traffic participants (such as mining trucks, pedestrians, and electric shovels) in mining environments can be achieved, which is especially suitable for complex environments with high dust levels and large changes in lighting in mining settings.

[0157] Based on task requirements and the status of mining trucks, a dynamic optimization algorithm is used to allocate and schedule tasks, and the priority of mining trucks is dynamically set to optimize resource allocation. This achieves intelligent task allocation and priority management in the method and system.

[0158] By combining high-precision maps with real-time task management, scheduling routes are planned, and the trajectories of other traffic participants are predicted to ensure route safety. This achieves the path planning and trajectory prediction capabilities of the method and system.

[0159] In the methods and systems of this disclosure, all trajectory points can be generated as a graph structure, and deep learning can be used to more accurately determine the probability of conflict. The system predicts the potential conflict situation in the loading and unloading area within a future timeframe based on the mining truck scheduling path and the future trajectories of other traffic participants in the loading and unloading area. Scheduling tasks are dynamically adjusted based on the conflict situation, and predicted trajectories are collaboratively optimized according to task priorities. The system considers the possibility of collisions while allocating scheduling tasks, improving the safety of mining trucks. The system can operate in an unmanned mode, requiring no human intervention. Adjusting the priority of scheduling tasks based on task status ensures maximum efficiency for unmanned operations.

[0160] In summary, the methods and systems of the embodiments of this disclosure can dynamically adjust the allocation of scheduling tasks for different time periods based on real-time traffic conditions and operational needs, ensuring the efficiency and safety of mine transportation operations. By utilizing real-time updated scheduling paths and considering the interactive effects between traffic participants, the predicted trajectories of other traffic participants (such as electric shovels, transportation equipment, etc.) are dynamically adjusted, thereby more accurately predicting the future movement trajectories of other traffic participants, achieving collaborative prediction and optimization, and improving the efficiency and safety of mine transportation.

[0161] Figure 8 is a schematic structural block diagram illustrating a dynamic scheduling system according to other embodiments of the present disclosure. The dynamic scheduling system includes a memory 810 and a processor 820. Wherein:

[0162] The memory 810 can be a disk, flash memory, or any other non-volatile storage medium. The memory is used to store the instructions in the embodiment corresponding to FIG1.

[0163] The processor 820 is coupled to the memory 810 and can be implemented as one or more integrated circuits, such as a microprocessor or microcontroller. The processor 820 executes instructions stored in the memory, which can reduce security risks and improve transportation efficiency.

[0164] In some embodiments, as shown in FIG9, the dynamic scheduling system 900 may also include a memory 910 and a processor 920. The processor 920 is coupled to the memory 910 via a BUS bus 930. The dynamic scheduling system 900 may also be connected to an external storage device 950 via a storage interface 940 to access external data, and may also be connected to a network or another computer system (not shown) via a network interface 960, which will not be described in detail here.

[0165] In this embodiment, storing data instructions in a memory and then processing the instructions with a processor can reduce safety risks and improve transportation efficiency.

[0166] In another embodiment, this disclosure also provides a computer-readable storage medium (e.g., a non-transitory computer-readable storage medium) having computer program instructions stored thereon that, when executed by a processor, implement the steps of the method in the embodiment corresponding to FIG1. ​​Those skilled in the art will understand that embodiments of this disclosure can be provided as methods, apparatus, or computer program products. Therefore, this disclosure can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this disclosure can take the form of a computer program product embodied on one or more computer-usable non-transitory storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0167] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in one or more flowchart illustrations and / or one or more block diagrams.

[0168] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means that implement the functions specified in one or more flowcharts and / or one or more block diagrams.

[0169] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions, which execute on the computer or other programmable apparatus, provide steps for implementing the functions specified in one or more flowcharts and / or one or more block diagrams.

[0170] In some embodiments of this disclosure, a computer program product is provided, which includes a computer program or instructions that, when executed by a processor, implement the dynamic scheduling method as described above.

[0171] In some embodiments of this disclosure, a computer program is also provided, comprising: instructions that, when executed by a processor, cause the processor to perform the dynamic scheduling method as described above.

[0172] This concludes the detailed description of the present disclosure. To avoid obscuring the concept of the disclosure, some details known in the art have not been described. Those skilled in the art will fully understand how to implement the technical solutions disclosed herein based on the above description.

[0173] While specific embodiments of this disclosure have been described in detail by way of example, those skilled in the art should understand that the examples are for illustrative purposes only and not intended to limit the scope of this disclosure. Those skilled in the art should understand that modifications can be made to the above embodiments without departing from the scope and spirit of this disclosure. The scope of this disclosure is defined by the appended claims.

Claims

1. A dynamic scheduling method, comprising: Acquire road conditions and environmental parameters in open-pit mines, as well as operational data and performance parameters of multiple traffic participants; A graph data structure is generated based on the road conditions and environmental parameters of the open-pit mine, as well as the operation data and performance parameters of the multiple traffic participants. Each node in the graph data structure contains relevant information about each traffic participant. The graph data structure is input into a neural network model to obtain the probability distribution of collisions between each node in the graph data structure and other nodes; and The driving strategy of at least one of the multiple traffic participants is adjusted based on the probability distribution of collisions between each node and other nodes in the graph data structure.

2. The dynamic scheduling method according to claim 1, wherein, A graph data structure is generated based on the road conditions and environmental parameters of the open-pit mine, as well as the operational data and performance parameters of the multiple traffic participants, including: Based on the road conditions and environmental parameters of the open-pit mine, as well as the operation data and performance parameters of the multiple traffic participants, path points of the trajectories of the multiple traffic participants are generated. Based on the waypoints of the trajectories of the multiple traffic participants, determine the path relationships between different waypoints; and The graph data structure is obtained based on the path points of the trajectories of the multiple traffic participants and the path relationships between different path points, wherein the path relationships are the edges between different nodes of the graph data structure.

3. The dynamic scheduling method according to claim 2, wherein: The multiple traffic participants include multiple mining trucks; Based on the road conditions and environmental parameters of the open-pit mine, as well as the operational data and performance parameters of the multiple traffic participants, waypoints for the trajectories of the multiple traffic participants are generated, including: Generate a task route for each mining truck based on its scheduling task; The projected speed of each mining truck is generated based on its load information; and Based on the mission route, expected speed, and speed limit information of each mining truck, waypoints for the planned trajectory of each mining truck are generated.

4. The dynamic scheduling method according to claim 2 or 3, wherein, The road conditions include: the traffic conditions and speed limit information of multiple roads in the open-pit mine; The environmental parameters include: weather conditions of the open-pit mine; The operational data includes: the current location, speed, acceleration, and load information of each mining truck; The performance parameters include: the speed of each mining truck when unloaded and the speed when fully loaded.

5. The dynamic scheduling method according to claim 3, wherein: The plurality of traffic participants also includes one or more electric shovels; The operational data includes: the speed, orientation, and historical location of the one or more electric shovels; Based on the road conditions and environmental parameters of the open-pit mine, as well as the operational data and performance parameters of the multiple traffic participants, the path points for the trajectories of the multiple traffic participants are generated, including: Based on the speed, orientation, and historical position of the one or more electric shovels, the future trajectory of the one or more electric shovels is generated using a Kalman filter algorithm.

6. The dynamic scheduling method according to any one of claims 2 to 5, wherein, The relevant information for each traffic participant includes: the identifier of each traffic participant, the coordinates of each waypoint of each traffic participant, the priority of each traffic participant's current task, the current speed information of each traffic participant, the current acceleration information of each traffic participant, the current load information of each traffic participant, the estimated time for each traffic participant to reach each node, and the path relationship between different waypoints.

7. The dynamic scheduling method according to any one of claims 1 to 6, wherein, The graph data structure is input into a neural network model to obtain the probability distribution of collisions between each node in the graph data structure and other nodes, including: The graph data structure is processed sequentially through the graph convolutional network, pooling layer and connection layer in the neural network model to obtain each node and other nodes that may collide with each node. Each node and other nodes that have a possibility of colliding with each node are input into the feedforward neural network of the neural network model to output the probability distribution vector of each node colliding with other nodes; and Based on the probability distribution vector of each node colliding with other nodes, the probability distribution of each node colliding with other nodes is calculated using a normalized exponential function.

8. The dynamic scheduling method according to any one of claims 1 to 7, wherein, Adjusting the driving strategy of at least one of the multiple traffic participants based on the probability distribution of collisions between each node and other nodes in the graph data structure includes: Based on the probability distribution of collisions between each node and other nodes in the graph data structure, the driving strategy of the at least one traffic participant is adjusted so that the probability of the at least one traffic participant colliding with other traffic participants among the plurality of traffic participants is minimized and the benefit of the at least one traffic participant is maximized.

9. The dynamic scheduling method according to claim 8, wherein: The at least one traffic participant includes at least one mining truck; The revenue of the at least one traffic participant is calculated based on the total load or total unload of each of the at least one mining trucks from the loading point to the unloading point, the distance of each of the at least one mining trucks from the loading point to the unloading point, the energy consumption per unit distance of each of the at least one mining trucks when fully loaded, the waiting time of each of the at least one mining trucks, and the energy consumption per unit time of each of the at least one mining trucks while waiting.

10. A dynamic scheduling system, comprising: The data acquisition unit is used to acquire road conditions and environmental parameters of the open-pit mine, as well as the operational data and performance parameters of multiple traffic participants. A generation unit is used to generate a graph data structure based on the road conditions and environmental parameters of the open-pit mine and the operation data and performance parameters of the multiple traffic participants, wherein each node in the graph data structure contains relevant information for each traffic participant; A probability acquisition unit is used to input the graph data structure into a neural network model to obtain the probability distribution of collisions between each node and other nodes in the graph data structure; and An adjustment unit is used to adjust the driving strategy of at least one of the plurality of traffic participants based on the probability distribution of collisions between each node and other nodes in the graph data structure.

11. A dynamic scheduling system, comprising: Memory; as well as A processor coupled to the memory, the processor being configured to execute the dynamic scheduling method as described in any one of claims 1 to 9 based on instructions stored in the memory.

12. A computer-readable storage medium having stored thereon computer instructions that, when executed by a processor, implement the dynamic scheduling method as described in any one of claims 1 to 9.

13. A computer program comprising: Instructions, when executed by a processor, cause the processor to perform the dynamic scheduling method as described in any one of claims 1 to 9.

Images