Collision avoidance methods, devices and electronic equipment for unmanned mining vehicles

Abstract

This application provides a collision avoidance method, device, and electronic device for unmanned driving in mines, relating to the field of unmanned driving technology. The method includes: real-time acquisition of a global traffic information map, first state information of a target vehicle, and second state information of surrounding vehicles; the global traffic information map is used to characterize the position and state of all vehicles in the mining area; based on the first state information, second state information, and the global traffic information map, a first expansion path corresponding to the target vehicle and second expansion paths corresponding to other vehicles are generated; both the first and second expansion paths are safe driving areas generated after the corresponding paths are expanded; if a collision point exists between the first expansion path and any of the second expansion paths, the target vehicle is controlled to avoid a collision with a potentially colliding vehicle. This method can reduce computational complexity, thereby improving the real-time performance of collision avoidance scheduling and reducing the risk of vehicle collisions.

Description

Technical Field

[0001] This application belongs to the field of unmanned driving technology, specifically relating to a collision avoidance method, device and electronic equipment for unmanned driving in mines. Background Technology

[0002] Unmanned mining trucks, as a typical intelligent transportation tool, are widely used in heavy industries such as mines and open-pit coal mines, primarily for transporting minerals. Due to the complex terrain and often harsh geological and environmental conditions in mines, traditional transportation methods cannot meet the requirements of efficiency and safety. In recent years, with the development of automation, artificial intelligence, and Internet of Things technologies, unmanned mining trucks have gradually become core equipment in the mining transportation field.

[0003] In the operation of unmanned mining trucks, how to handle the right-of-way issues when multiple mining trucks run parallel or intersect in the same mine tunnel, and how to make accurate driving decisions based on real-time data, have become bottlenecks restricting their further development.

[0004] Current autonomous driving systems are based on deep reinforcement learning models. While they can achieve basic path planning and obstacle avoidance, they suffer from poor real-time performance and high computational latency in complex scenarios, especially on mining roads, leading to a higher risk of collisions. Summary of the Invention

[0005] The technical problem to be solved by this application is to address the above-mentioned shortcomings of the existing technology by providing a collision avoidance method, device and electronic equipment for unmanned mining vehicles. Using this unmanned mining vehicle collision avoidance method can reduce computational latency and improve real-time performance, thereby reducing the risk of vehicle collisions.

[0006] In a first aspect, embodiments of this application provide a collision avoidance method for unmanned mining operations, including: The system acquires real-time global traffic information maps, the first state information of the target vehicle, and the second state information of surrounding vehicles. The global traffic information map is used to represent the position and state of all vehicles in the mining area. Surrounding vehicles refer to vehicles around the target vehicle. Based on the first state information, the second state information, and the global traffic information map, a first expansion path corresponding to the target vehicle and a second expansion path corresponding to other vehicles are generated; both the first expansion path and the second expansion path are safe driving areas generated after the corresponding paths are expanded. If a collision point exists between the first expansion path and any second expansion path, the target vehicle is controlled to avoid colliding with a potential collision vehicle; the potential collision vehicle is any other vehicle on the second expansion path where a collision point exists.

[0007] In some embodiments of the first aspect, real-time acquisition of a global traffic information map, first state information of the target vehicle, and second state information of surrounding vehicles includes: The vehicle-cloud communication system obtains a global traffic information map in real time from the cloud platform; the global traffic information map is generated by the cloud platform based on the status information uploaded by all vehicles in the mining area. Obtain the first state information of the target vehicle; Real-time second-state information of surrounding vehicles is obtained through vehicle-to-everything (V2X) communication.

[0008] In some embodiments of the first aspect, generating a first expansion path corresponding to the target vehicle and a second expansion path corresponding to other vehicles based on first state information, second state information, and a global traffic information map includes: Determine the first path of the target vehicle from the first state information; Based on the first state information, the first path is normalized to generate a first expanded path; Determine the second path for other vehicles based on the second state information and the global traffic information map; Based on the second state information and the global traffic information map, the corresponding second path is normalized to generate the corresponding second expanded path.

[0009] In some embodiments of the first aspect, controlling the target vehicle to avoid a collision with a potential collision vehicle includes: The collision scenario is determined based on the point of collision, the first state information, and the third state information corresponding to the vehicles that may collide. Based on the collision scenario, the target vehicle is controlled to avoid collisions with other vehicles that may collide.

[0010] In some implementations of the first aspect, the collision scenario is a longitudinal same-direction conflict scenario; the longitudinal same-direction conflict scenario includes a following vehicle scenario. Based on the collision scenario, control the target vehicle to avoid collisions with potential collision vehicles, including: Control the target vehicle to maintain a safe distance from vehicles that may collide, in order to avoid a collision.

[0011] In some embodiments of the first aspect, the collision scenario is a head-on conflict scenario; head-on conflict scenarios include passing scenarios. Based on the collision scenario, control the target vehicle to avoid collisions with potential collision vehicles, including: The vehicles that need to give way are determined according to the first preset priority rule; If the vehicle that needs to yield is the target vehicle, then the target vehicle is controlled to perform the yielding operation before reaching the collision point in order to avoid a collision.

[0012] In some implementations of the first aspect, the collision scenario is a lateral intersection conflict scenario; the lateral intersection conflict scenario includes intersection scenarios and same-direction merging scenarios. Based on the collision scenario, control the target vehicle to avoid collisions with potential collision vehicles, including: Determine whether the target vehicle needs to yield or accelerate to overtake based on the second preset priority rule; If it is determined that the target vehicle needs to yield or overtake, the target vehicle will be controlled to perform the corresponding yielding or overtaking maneuver before reaching the collision point in order to avoid a collision.

[0013] In some embodiments of the first aspect, if a collision point exists between the first expansion path and any second expansion path, before controlling the target vehicle to avoid a collision with a potentially colliding vehicle, the method further includes: Determine whether there is a spatial intersection between the first expansion path and all second expansion paths; If spatial intersection is determined, then based on the first state information, the second state information, and the global traffic information map, it is determined whether spatial intersection also has temporal intersection. If it is determined that time overlap exists simultaneously, then a collision point is determined to exist between the first expansion path and any second expansion path.

[0014] Based on the same inventive concept, in a second aspect, embodiments of this application also provide a collision avoidance device for unmanned mining operations, comprising: The acquisition module is used to acquire the global traffic information map, the first state information of the target vehicle, and the second state information of the surrounding vehicles in real time; the global traffic map is used to represent the position and status of all vehicles in the mining area; the surrounding vehicles are the vehicles around the target vehicle. The generation module is used to generate a first expansion path corresponding to the target vehicle and a second expansion path corresponding to other vehicles based on the first state information, the second state information and the global traffic information map; both the first expansion path and the second expansion path are safe driving areas generated after the corresponding paths are expanded. The control module is used to control the target vehicle to avoid a collision with a potential collision vehicle if there is a collision point between the first expansion path and any second expansion path; the potential collision vehicle is another vehicle corresponding to the second expansion path where the collision point exists.

[0015] In some embodiments of the second aspect, the acquisition module is specifically used for: The global traffic information map is obtained in real time from the cloud platform through vehicle-to-cloud communication; the global traffic information map is generated by the cloud platform based on the status information uploaded by all vehicles in the mining area; the first status information of the target vehicle is obtained; and the second status information of the surrounding vehicles is obtained in real time through vehicle-to-everything (V2X) communication.

[0016] In some embodiments of the second aspect, the generation module is specifically used for: The first path of the target vehicle is determined from the first state information; the first path is normalized based on the first state information to generate a first expanded path; the second paths of other vehicles are determined based on the second state information and the global traffic information map; the corresponding second paths are normalized based on the second state information and the global traffic information map to generate corresponding second expanded paths.

[0017] In some embodiments of the second aspect, when controlling the target vehicle to avoid a collision with a potentially colliding vehicle, the control module is specifically used for: The collision scenario is determined based on the collision point, the first state information, and the third state information corresponding to the possible collision vehicle; the target vehicle is controlled to avoid collision with the possible collision vehicle based on the collision scenario.

[0018] In some implementations of the second aspect, the collision scenario is a longitudinal same-direction conflict scenario; the longitudinal same-direction conflict scenario includes a following vehicle scenario. When the control module controls the target vehicle to avoid a collision with a potential collision vehicle based on the collision scenario, it is specifically used for: Control the target vehicle to maintain a safe distance from vehicles that may collide, in order to avoid a collision.

[0019] In some embodiments of the second aspect, the collision scenario is a head-on conflict scenario; head-on conflict scenarios include passing scenarios. When the control module controls the target vehicle to avoid a collision with a potential collision vehicle based on the collision scenario, it is specifically used for: The vehicle that needs to yield is determined according to the first preset priority rule; if the vehicle that needs to yield is the target vehicle, the target vehicle is controlled to perform the yielding operation before reaching the collision point in order to avoid a collision.

[0020] In some implementations of the second aspect, the collision scenario is a lateral intersection conflict scenario; the lateral intersection conflict scenario includes intersection scenarios and same-direction merging scenarios. When the control module controls the target vehicle to avoid a collision with a potential collision vehicle based on the collision scenario, it is specifically used for: The second preset priority rule determines whether the target vehicle needs to yield or accelerate to overtake. If it is determined that the target vehicle needs to yield or overtake, the target vehicle is controlled to perform the corresponding yielding or accelerating operation before reaching the collision point in order to avoid a collision.

[0021] In some embodiments of the second aspect, the apparatus further includes: The judgment module is used to determine whether there is a spatial intersection between the first expansion path and all second expansion paths; if a spatial intersection is determined to exist, it is determined whether a temporal intersection also exists based on the first state information, the second state information and the global traffic information map; if a temporal intersection also exists, it is determined that there is a collision point between the first expansion path and any second expansion path.

[0022] Based on the same inventive concept, in a third aspect, embodiments of this application also provide an electronic device, including: a memory and a processor; The memory stores the instructions that the computer executes; The processor executes computer-executable instructions stored in memory to implement a collision avoidance method for unmanned mining operations, as described in any of the first aspects.

[0023] Based on the same inventive concept, in a fourth aspect, embodiments of this application also provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the mine unmanned driving collision avoidance method as described in any of the first aspects.

[0024] According to the collision avoidance method, device, and electronic equipment for unmanned mining provided in this application, a first expansion path corresponding to the target vehicle and a second expansion path corresponding to other vehicles can be generated by using a global traffic information map characterizing the position and status of all vehicles in the mining area, and by using the first state information of the target vehicle and the second state information of surrounding vehicles acquired in real time. Since the first and second expansion paths are simple geometric shapes, by determining whether there is a collision point between the first and any second expansion path, and when a collision point exists, the target vehicle is controlled to avoid collisions with potentially colliding vehicles. Thus, through the aforementioned simple and efficient geometric calculations, the computational complexity can be greatly reduced, thereby improving the real-time performance of collision avoidance scheduling and reducing the risk of vehicle collisions. Attached Figure Description

[0025] Figure 1 This illustration shows a flowchart of a collision avoidance method for unmanned mining operations provided in an embodiment of this application. Figure 2 This illustration shows another flowchart of the collision avoidance method for unmanned mining operations provided in an embodiment of this application; Figure 3 This illustration shows another flowchart of the collision avoidance method for unmanned mining operations provided in this application embodiment; Figure 4 This diagram illustrates the expansion path provided in an embodiment of this application. Figure 5 This diagram illustrates the existence of collision points between expansion paths provided in the embodiments of this application. Detailed Implementation

[0026] To enable those skilled in the art to better understand the technical solutions of this application, the application will be further described in detail below with reference to the accompanying drawings and embodiments.

[0027] The features and exemplary embodiments of various aspects of this application will now be described in detail. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only configured to explain this application and are not configured to limit this application. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples of this application.

[0028] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0029] As described in the background section, current autonomous driving systems are based on deep reinforcement learning models. While they can achieve basic path planning and obstacle avoidance, they suffer from poor real-time performance and high computational latency in complex scenarios, especially on mining roads, leading to a higher risk of collisions and requiring further optimization.

[0030] Example 1

[0031] The collision avoidance method for unmanned mining operations provided in this application is applied to an electronic device, which can be a computer or a device within a computer for implementing the collision avoidance method. The computer can be a server, a server cluster, or a terminal; this application does not specifically limit its application in this regard. The following description uses the example of the collision avoidance method for unmanned mining operations being executed by an electronic device.

[0032] like Figure 1 As shown, the collision avoidance method for unmanned mining provided in this application embodiment may include steps S101 to S103.

[0033] S101. Real-time acquisition of the global traffic information map, the first state information of the target vehicle, and the second state information of surrounding vehicles. The global traffic information map is used to represent the position and status of all vehicles in the mining area. Surrounding vehicles refer to vehicles around the target vehicle.

[0034] For example, the global traffic information map can be generated by a cloud platform based on the fusion of status information uploaded by all vehicles in the mining area, including information such as the position, speed, path, and heading angle of all vehicles. In this case, the global traffic information map can be obtained directly from the cloud platform, thereby improving the efficiency of global traffic information map acquisition.

[0035] The target vehicle is the control entity for collision avoidance, and it can be the self-driving vehicle corresponding to the method of this embodiment. The first state information is the same as the aforementioned state information, and may also include information such as the target vehicle's position, speed, path, and heading angle.

[0036] Surrounding vehicles are those around the target vehicle. They can be determined through vehicle-to-vehicle communication. The specific definition of surrounding vehicles can be selected according to the actual application, such as when the distance between them and the target vehicle is within a preset distance range, when there is a business connection, or when the paths intersect.

[0037] S102. Generate a first expansion path corresponding to the target vehicle and a second expansion path corresponding to other vehicles based on the first state information, the second state information, and the global traffic information map. Both the first and second expansion paths are safe driving areas generated after the corresponding paths are expanded.

[0038] For example, the first expansion path is generated by normal expansion of the first path of the target vehicle. Normal expansion can be implemented based on the target vehicle's width, the target vehicle's location environment, the target vehicle's task, and the influence of other vehicles on the target vehicle. That is, path expansion can be based on global reference line planning and local planning.

[0039] Similarly, the second expansion path is generated by normal expansion of the second path of other vehicles, and the expansion method is the same as that of the target vehicle.

[0040] For example, "other vehicles" can refer to all other vehicles besides the target vehicle, or it can refer to vehicles that are spatially or task-related to the target vehicle and may require determination of whether to collide with it. In practical applications, if a second expansion path is determined for vehicles that are spatially or task-related to the target vehicle and may require determination of whether to collide with it, the computational complexity can be further reduced and the computational efficiency improved.

[0041] For example, the first expansion path and the second expansion path can be referenced. Figure 4 The blue and red bars indicate safe driving areas.

[0042] S103. If a collision point exists between the first expansion path and any second expansion path, then control the target vehicle to avoid a collision with a potential collision vehicle. A potential collision vehicle is any other vehicle on the second expansion path where a collision point exists.

[0043] For example, controlling the target vehicle to avoid a collision with a potential vehicle can be done by accelerating, decelerating, or stopping to yield. The specific method depends on the relationship between the target vehicle and the potential vehicle. For instance, if the collision point occurs in the oncoming lane, the vehicle can stop to yield before the collision point (in a wide area or at a road intersection) to avoid a collision.

[0044] According to the collision avoidance method for unmanned vehicles in mines provided in this application, a first expansion path corresponding to the target vehicle and a second expansion path corresponding to other vehicles can be generated by using a global traffic information map characterizing the position and status of all vehicles in the mining area, and by using the first state information of the target vehicle and the second state information of surrounding vehicles acquired in real time. Since the first and second expansion paths are simple geometric shapes, by determining whether there is a collision point between the first and any second expansion path, and when a collision point exists, the target vehicle is controlled to avoid collisions with potentially colliding vehicles. Thus, through the aforementioned simple and efficient geometric calculations, the computational complexity can be greatly reduced, thereby improving the real-time performance of collision avoidance scheduling and reducing the risk of vehicle collisions.

[0045] Example 2

[0046] like Figure 2 As shown, the collision avoidance method for unmanned mining provided in this application embodiment is further described based on the collision avoidance method for unmanned mining provided in embodiment 1 of this application, and may include steps S201 to S205.

[0047] S201. Obtain the global traffic information map in real time from the cloud platform via vehicle-to-cloud communication. The global traffic information map is generated by the cloud platform based on the status information uploaded by all vehicles in the mining area.

[0048] For example, vehicle-to-cloud communication refers to high-bandwidth, low-latency two-way data interaction between vehicles and cloud service platforms via mobile communication networks. Vehicle-to-cloud communication can improve the efficiency of obtaining real-time global traffic information maps from the cloud platform, thereby improving the efficiency of subsequent vehicle collision avoidance.

[0049] For example, the target vehicle will also upload its own status information to the cloud platform in real time.

[0050] S202, Obtain the first state information of the target vehicle.

[0051] For example, the initial state information of the target vehicle can be obtained in real time through sensors.

[0052] S203: Real-time acquisition of the second status information of surrounding vehicles through vehicle-to-everything (V2X) communication.

[0053] For example, vehicle-to-everything (V2X) can improve the efficiency of acquiring second-state information of surrounding vehicles.

[0054] For example, when unmanned mining trucks are engaged in transportation operations in a mining environment, there is a need for multiple trucks to work together. In this environment, enabling each truck to obtain real-time status information from other vehicles in the vicinity is crucial to ensuring driving safety and operational efficiency. To achieve this goal, this embodiment uses V2X communication and vehicle-to-cloud communication technologies to achieve parallel information exchange.

[0055] V2X communication enables mining trucks to directly exchange information with other mining trucks in the vicinity, obtaining status information of surrounding vehicles, including but not limited to local paths, location coordinates, speed, heading angle, driving area, driving mode, and load status. This information helps mining trucks perceive the dynamics of other vehicles in real time, predict potential path intersections and collision risks, and make corresponding behavioral adjustments.

[0056] Vehicle-to-Cloud Communication: Operating in parallel with V2X communication, mining trucks upload their status information to the cloud platform via vehicle-to-cloud communication. The cloud platform aggregates status data from multiple mining trucks to form a global traffic information map. The cloud platform not only aggregates and processes data but also pushes real-time global information within the mining area, ensuring that each mining truck has access to the latest surrounding environment and traffic conditions. Vehicle-to-cloud communication relies on base stations and antennas to achieve full coverage within the mining area, eliminating communication blind spots.

[0057] These two communication technologies work in parallel, ensuring that the mining truck can rely on both real-time information from surrounding vehicles and global traffic data when making path crossing judgments and decisions. In this way, the unmanned mining truck can effectively avoid information lag and data inconsistency, improving the real-time performance and reliability of collision avoidance.

[0058] S204. Generate a first expansion path corresponding to the target vehicle and a second expansion path corresponding to other vehicles based on the first state information, the second state information and the global traffic information map.

[0059] For example, the second state information and the global traffic information map can be fused to obtain updated state information for other vehicles. The updated state information for other vehicles is more accurate than either the second state information or the global traffic information map alone.

[0060] By using the updated status information of other vehicles, a normal collision is performed to obtain the second expansion path corresponding to the other vehicles.

[0061] In some implementations, S204 may be specifically as follows: The first path of the target vehicle is determined from the first state information.

[0062] The first path is normalized based on the first state information to generate the first expanded path.

[0063] The second path for other vehicles is determined based on the second state information and the global traffic information map.

[0064] Based on the second state information and the global traffic information map, the corresponding second path is normalized to generate the corresponding second expanded path.

[0065] For example, normal expansion can be implemented based on the target vehicle's width, the target vehicle's location environment, the target vehicle's mission, and the influence of other vehicles on the target vehicle.

[0066] The second-state information and the global traffic information map can be merged first, and then the second path for other vehicles can be determined. Similarly, the second-state information and the global traffic information map can be merged first, and then the corresponding second path can be normalized.

[0067] In some implementations, a collision point determination process is included before step S205, as detailed below: Determine whether there is a spatial intersection between the first expansion path and all second expansion paths.

[0068] If a spatial intersection is determined, then the first state information, the second state information, and the global traffic information map are used to determine whether a temporal intersection also exists.

[0069] If it is determined that time overlap exists simultaneously, then a collision point is determined to exist between the first expansion path and any second expansion path.

[0070] For example, spatial intersection indicates the existence of path intersection points. Temporal intersection requires determining whether the target vehicle and other vehicles arrive at the path intersection point at the same time or within a potential collision time range, based on the first state information, second state information, and speed and position information in the global traffic information map. If so, it means that temporal intersection also exists, and the collision point is the corresponding path intersection point and the corresponding collision time.

[0071] In this embodiment, the determination of whether a collision point exists determines the vehicle's driving safety in complex environments.

[0072] Path dilation: Path dilation is a fundamental step in determining whether paths intersect. When calculating path intersections, the path of each vehicle is first dilated in the normal direction, resulting in an approximate rectangular path. This dilation process is based on global reference line planning and local planning. The area after path dilation represents the safe driving zone for vehicles. The size of the dilation is set according to the vehicle size and the specific environment of the mining area to ensure sufficient safety space is considered during collision detection.

[0073] Rectangular collision detection: The expanded path is approximately rectangular in shape, which makes collision detection simple and efficient.

[0074] For example, the Boost Geometry library (a function library providing geometric objects and algorithms) is used to determine collisions in the expanded path rectangles. This library offers mature and efficient geometric algorithms that can quickly determine whether two rectangles intersect. By calculating the spatial relationship between the path rectangles of the two vehicles, the risk of path intersection can be determined.

[0075] Collision point calculation: If the rectangles of two paths intersect, the specific points where a collision may occur can be calculated (the collision area is as follows). Figure 5 (As shown in green). Collision points are two-dimensional coordinate points where temporal and spatial conflicts may occur after paths intersect. By accurately calculating these collision points, the collision location can be determined, providing a basis for subsequent behavioral decisions.

[0076] Collision point calculation can incorporate a time estimation model. This model determines whether two vehicles will arrive at the path intersection simultaneously (i.e., time intersection) and collide by accurately calculating the estimated time for vehicles to arrive at the intersection. This allows for dynamic adjustment of vehicle priorities to ensure safe passage.

[0077] Specifically, the time estimation model works as follows: First, based on the relative positions and speeds of the two mining trucks, the arrival times at the intersection are predicted. This prediction model considers the dynamic driving states of the vehicles, including current speed, acceleration, and position.

[0078] After obtaining the estimated arrival time, compare the arrival times of your own vehicle with those of other vehicles to determine whether a collision is likely. If the estimated arrival times of the two vehicles overlap, then a preset priority rule (such as "lighter load yields to heavier load" or "first come, first served") is used to determine which mining truck has priority to pass.

[0079] If the arrival times of the two vehicles differ significantly, both vehicles are allowed to pass through the intersection simultaneously without colliding.

[0080] This time estimation model effectively compensates for the lack of time dimension judgment in traditional path intersection judgment methods, ensuring efficiency and safety in multi-vehicle collaboration. Especially in high-density, multi-vehicle collaborative operation environments such as mining areas, it can quickly respond and adjust decisions to avoid collisions due to poor real-time performance.

[0081] For example, if a potential collision risk is identified, a warning can be triggered to alert the user of the specific risk.

[0082] S205. If there is a collision point between the first expansion path and any second expansion path, then control the target vehicle to avoid colliding with the possible collision vehicle.

[0083] In some implementations, controlling the target vehicle to avoid a collision with a potential collision vehicle includes: The collision scenario is determined based on the point of collision, the first state information, and the third state information corresponding to the vehicles that may collide.

[0084] Based on the collision scenario, the target vehicle is controlled to avoid collisions with other vehicles that may collide.

[0085] For example, if the collision point occurs at an intersection, the association between the target vehicle and the vehicle that may collide can be determined based on the first state information and the third state information. For instance, if the target vehicle is going straight at the intersection and the vehicle that may collide is turning left at the intersection, then the collision scenario is an intersection scenario.

[0086] For example, if the collision occurs on a road, the association between the target vehicle and the vehicle that may collide can be determined based on the first state information and the third state information. For instance, if the target vehicle is following the vehicle that may collide, then the collision scenario is a following scenario.

[0087] In some implementations, the collision scenario is a longitudinal, same-direction conflict scenario. Longitudinal, same-direction conflict scenarios include following-vehicle scenarios.

[0088] The process of controlling the target vehicle to avoid a collision with a potential collision vehicle based on the collision scenario can be specifically described as follows: Control the target vehicle to maintain a safe distance from vehicles that may collide, in order to avoid a collision.

[0089] For example, longitudinal same-direction conflict scenarios are mainly following scenarios. When the target vehicle is following another vehicle, it is necessary to ensure that the speed is not too high and to maintain a safe distance to avoid collision.

[0090] In some implementations, the collision scenario is a head-on conflict scenario. Head-on conflict scenarios include passing scenarios.

[0091] The process of controlling the target vehicle to avoid a collision with a potential collision vehicle based on the collision scenario can be specifically described as follows: The vehicles that need to give way are determined according to the first preset priority rule.

[0092] If the vehicle that needs to yield is the target vehicle, then the target vehicle is controlled to perform the yielding operation before reaching the collision point in order to avoid a collision.

[0093] For example, oncoming conflict scenarios also include situations where an oncoming vehicle is encountered while overtaking. The first preset priority rule can determine which vehicle should yield based on factors such as business importance and load level (lighter load yields to heavier load).

[0094] In some implementations, the collision scenario is a lateral intersection conflict scenario. Lateral intersection conflict scenarios include intersection scenarios and same-direction merging scenarios.

[0095] Based on the collision scenario, control the target vehicle to avoid collisions with potential collision vehicles, including: The second preset priority rule determines whether the target vehicle needs to give way or accelerate to overtake.

[0096] If it is determined that the target vehicle needs to yield or overtake, the target vehicle will be controlled to perform the corresponding yielding or overtaking maneuver before reaching the collision point in order to avoid a collision.

[0097] Examples of lateral intersection conflict scenarios include cross / T intersection scenarios, traffic divergence (exit) scenarios, parking lot intersection scenarios, and pedestrian crossing scenarios.

[0098] For example, the second preset priority rule can determine whether a target vehicle needs to yield or overtake based on factors such as business importance, load, vehicle speed, and surrounding vehicle conditions. If the business importance is high, there are few surrounding vehicles, and the vehicle's load is low, an acceleration operation can be performed to achieve speed-up and overtaking.

[0099] The first and second preset priority rules can be the same or different, and both can be based on the following rules: 1) General Driving Rules: Manned vehicles must yield to unmanned vehicles: In the mining area, if an unmanned mining truck and a manned vehicle cross paths, the manned vehicle must give way to the unmanned mining truck.

[0100] Empty unmanned trucks yield to loaded unmanned trucks: In the mining area, when an empty mining truck meets a loaded mining truck, the empty mining truck must yield to the loaded mining truck to ensure that the loaded mining truck has priority in passing.

[0101] Between unmanned mining trucks in the same state: The order of arrival at the entrance follows a "first-come, first-served" principle. If two unmanned mining trucks in the same state meet at an intersection, the order of arrival time at the intersection determines which truck proceeds first.

[0102] The general rule is: stop at red lights, go at green lights. In intelligently controlled sections of the mining area, the movement of mining trucks is controlled by traffic light signals. When the light is red, vehicles must stop and wait. When the light is green, vehicles can proceed. A yellow light may require vehicles to slow down or stop and wait, depending on the situation.

[0103] General rules for straight, left, and right turns: At intersections in mining areas, the rules for straight, left, and right turns take precedence. For example, vehicles going straight usually have priority over vehicles turning left or right.

[0104] Emergency Vehicle Priority: Different vehicle priorities are set, with vehicles performing urgent tasks (such as emergency vehicles) having the highest priority, followed by unmanned mining trucks, manned vehicles, auxiliary production vehicles, and light vehicles. This priority rule ensures the smooth operation of emergency operations and critical transportation tasks within the mining area.

[0105] 2) Specific environmental rules: Narrow and Long Roads: In narrow and long road sections within the mining area, intelligent control devices are deployed at both entrances and exits, and vehicles follow traffic light rules. Entry is prohibited when the light is red, and is permitted when the light is green. During a yellow light, vehicles may be required to slow down or stop, depending on the specific circumstances, to ensure smooth and safe traffic flow. This rule avoids traffic congestion in narrow road sections and ensures that vehicles proceed according to priority.

[0106] Mixed traffic at intersections: At intersections in the mining area, traffic control is implemented according to standard traffic rules, such as auxiliary roads yielding to main roads, turning roads yielding to straight traffic, level roads yielding to ramps, and outside roads yielding to inside roads. Rule configuration is supported, allowing for the customization of specific yielding rules based on the mining area environment.

[0107] In this embodiment, appropriate decision-making schemes are automatically matched according to the actual needs of vehicles in the mining area, ensuring that unmanned vehicles can operate efficiently and safely in different scenarios. Roads, areas, and work sites in mining areas are usually structured, and different environments have different requirements for the driving of mining trucks. Therefore, it is necessary to dynamically adjust the decision-making strategy according to the actual situation in the mining area.

[0108] For example, the scene matching process can be as follows: First, real-time environmental data within the mining area is acquired, including road types (such as loading and unloading areas, narrow sections, intersections, etc.), obstacle information, and the location and status of other mining vehicles (based on third-state information). Based on this information and first-state information, the environmental scene in which the target vehicle is located is automatically identified, such as loading and unloading areas, narrow roads, intersections, etc., and collisions may occur.

[0109] For different environmental scenarios, such as in loading and unloading areas, priority is given to parking and waiting. In narrow lanes, traffic strategies are adjusted based on traffic light control and traffic priority. At intersections, the order of arrival and priority rules (such as "first come, first served") determine which mining truck goes first.

[0110] Secondly, for collision scenarios (which may occur in the aforementioned environmental scenarios), common scenario types include: Following scenario: When the vehicle's path is similar to and the distance to the vehicle in front is close, it is identified as a following scenario. At this time, the distance to the vehicle in front is monitored to ensure that the vehicle maintains a safe following distance and avoids rear-end collisions.

[0111] Meeting Scenario (Two-Lane): This scenario is defined as a meeting scenario when the paths of two vehicles intersect and the meeting occurs on a two-lane road. In this scenario, intelligent algorithms determine the relative speed and distance between the two vehicles to ensure they pass the intersection smoothly. Simultaneously, the speed and driving strategy of each vehicle (e.g., a mining truck) are adjusted based on lane width, direction of travel, and vehicle status. For example, the target vehicle may accelerate, slow down to yield, or stop to yield.

[0112] Intersection Scenario: At intersections in mining areas, private vehicles may encounter conflicts with other mining trucks. The system will assess the traffic flow at the intersection and determine whether to slow down or stop. Combined with traffic light control, the system will automatically adjust the driving strategy of the mining trucks.

[0113] Loading Area: In the loading area of ​​the mining area, mining trucks need to work in coordination with other mining trucks or loading machinery. The decision to wait for or avoid other mining trucks is based on location and task priority.

[0114] Unloading Zone: In the unloading zone of the mining area, the mining truck needs to complete the unloading operation. In this scenario, the mining truck is instructed to stop or adjust its direction of travel to ensure the smooth progress of the operation.

[0115] Two-lane to single-lane transition scenario (for merging in the same direction): In a mining area where a two-lane road transitions to a single-lane road, when two mining trucks are traveling in the same direction and merging into a single lane, priority is determined based on lane capacity, arrival order, and vehicle status. Subsequent vehicles are instructed to wait or slow down to avoid traffic congestion and accidents.

[0116] Narrow Road Meeting Scenario (Oncoming Conflict): In narrow sections or single-lane scenarios within a mining area, when two mining trucks are traveling in opposite directions, the path expansion results determine which truck needs to stop and wait, and which truck can proceed. In this scenario, vehicles are prohibited from entering when the light is red, are allowed to pass when the light is green, and the situation is determined based on the circumstances when the light is yellow.

[0117] After completing scenario analysis and generating behavioral decisions, control commands are generated based on the decision results. These control commands include: Following Control: In following scenarios, the mining truck is instructed to maintain a safe distance from the vehicle in front, avoiding excessive acceleration or getting too close. Deceleration Command: In scenarios involving oncoming traffic or intersections where a potential collision is detected, a deceleration command is issued, requiring the vehicle to reduce speed to avoid a collision. Stop Command: In certain situations, such as intersections or loading / unloading areas, the mining truck is instructed to stop and wait for other mining trucks to pass or to complete its task. Normal Driving: In scenarios without collision risk, the mining truck is allowed to continue driving normally, maintaining its current speed.

[0118] Meanwhile, all control commands and the operating status of the mining card will be uploaded to the cloud platform in real time for monitoring, data recording, and subsequent analysis.

[0119] The collision avoidance method for unmanned mining vehicles in this embodiment has the following effects: Through real-time V2X communication and vehicle-to-cloud communication, the location information and dynamic status of each vehicle can be quickly obtained, thereby optimizing vehicle path planning and decision-making. Simultaneously, through precise path intersection judgment and collision point calculation, unnecessary traffic delays are effectively avoided, ensuring that mining trucks can complete transportation tasks within the mining area more efficiently. Especially in complex mining environments, precise real-time decision control enables multi-vehicle collaboration, reducing waiting times and unnecessary speed adjustments, thereby improving operational efficiency and shortening task completion time.

[0120] Secondly, through path expansion and collision detection mechanisms, potential collision risks can be monitored and predicted in real time, allowing for proactive behavioral adjustments and reducing the risk of collisions between vehicles in the mining area. Simultaneously, by considering the actual mining environment, different collision scenarios (such as following other vehicles, meeting oncoming traffic, intersections, etc.) are automatically identified, and driving behavior decisions are generated based on preset rules to ensure that mining vehicles maintain a safe distance in changing environments, thus preventing traffic accidents.

[0121] In summary, the method of this embodiment not only improves the working efficiency of unmanned mining trucks, but also enhances their safety in complex mining environments.

[0122] To better understand the collision avoidance method for unmanned mining operations provided in this application, more specific embodiments will be used for illustration below. In this embodiment, a mining truck is used as an example.

[0123] like Figure 3 As shown, the architecture of this embodiment includes an information transmission link, a planning module, and a decision-making module.

[0124] V2X and Cloud Platform Forwarding of Multi-Vehicle Messages: In the operating environment of unmanned mining trucks, each mining truck exchanges real-time information with other surrounding vehicles (such as OBUs (On-Board Units)), infrastructure, and cloud platforms through V2X communication technology. V2X communication is mainly used to enable direct real-time communication between mining trucks and surrounding vehicles to obtain key status information about other vehicles (such as local paths (e.g., 200m), other vehicles' trajectories, positioning coordinates, speed, heading angle, etc.).

[0125] Simultaneously, the mining trucks also upload their own status information (such as their trajectory (200m) and status) to the cloud platform via vehicle-to-cloud communication. The cloud platform aggregates the location information and status data from multiple mining trucks to form a global traffic information map. Through this parallel communication method, the mining trucks can not only directly perceive the dynamics of surrounding vehicles, but also obtain global information about other vehicles in the area with the help of the cloud platform. The cloud platform's processing and data fusion will provide comprehensive information support for subsequent path intersection judgment, scenario analysis, and behavioral decision-making.

[0126] The planning module performs path intersection determination between the target vehicle and other vehicles (whether there are intersection points or collision areas): Based on the vehicle status information obtained in the previous step, the planning module performs intersection judgments on the paths of its own vehicle and other vehicles. The specific operations include the following steps: Path expansion: By expanding the safe area of ​​the vehicle path, the influence range of the vehicle path is calculated. This expanded area is used to assess the relative position of the target vehicle and other vehicles, ensuring that even in cases of incomplete intersection, the vehicle can avoid potential collision risks.

[0127] Collision point calculation: After path expansion, such as Figure 4 As shown in the diagram, the blue and red bars represent the expansion paths of the two vehicles, while the x and y axes represent coordinate values. Points where collisions may occur are calculated. These collision points are dynamically calculated based on vehicle state information such as current position, speed, heading angle, and path shape to determine possible spatial intersections, i.e., path intersections and temporal intersections. Path intersections are shown as follows: Figure 5 The green area shows the collision points, which are the coordinate points surrounding the green area.

[0128] After calculating the collision point, the decision-making module analyzes the collision scenario and generates a decision action based on the scenario type. This process relies on the collision point calculation results from the previous step, as well as the status information of the vehicle and other vehicles (such as load status, speed, etc.).

[0129] The decision-making module mainly performs the following operations: Scene judgment: The system divides the current environment into different scenes, including "following another vehicle", "meeting another vehicle", "intersection", "loading area", "unloading area", etc. The behavioral requirements of the vehicle are different in these scenes.

[0130] Following another vehicle: If your vehicle's path is the same as the vehicle in front and they intersect, and the distance between them is relatively short, then it is considered a following vehicle scenario. In this case, your vehicle needs to maintain a safe distance from the vehicle in front and avoid excessive acceleration.

[0131] Meeting oncoming traffic: If two vehicles' paths intersect and they meet at the intersection within the same time period, it is considered a meeting oncoming traffic scenario. A first preset priority rule (such as light vehicles yielding to heavy vehicles, and turning vehicles yielding to straight-going vehicles) determines which vehicle needs to yield.

[0132] Intersection scenario: In an intersection scenario, it is necessary to determine whether there are other vehicles or obstacles to ensure that the vehicle will not collide when entering the intersection.

[0133] Loading and unloading area: In the loading or unloading area, your vehicle may need to stop and wait or give way to other vehicles.

[0134] Distance and Time Calculation: By calculating the relative distance between the vehicle and other vehicles, the system estimates whether a collision will occur at the intersection. Based on the relative positions along the paths and the directions of vehicle movement, it determines the potential intersection opportunities between vehicles and makes corresponding decisions.

[0135] Decision-making and Control: After determining the current scenario, the system generates control commands based on the decision results and sends them to the execution module. Specific control behaviors include: Following Control: If the system determines it is a following scenario, it will instruct the vehicle to maintain a safe distance from the vehicle in front and adjust its speed according to the speed difference. Deceleration Command: If the system determines it is about to enter an intersection or a meeting scenario, it will issue a deceleration command, requiring the vehicle to reduce its speed to ensure safe passage. Stop Command: In some complex scenarios, such as intersections or loading areas, the system may instruct the vehicle to stop and wait for other vehicles to pass. Normal Driving: When the system determines there is no intersection conflict in the current environment, it allows the vehicle to maintain its current speed and continue normal driving.

[0136] These control commands are fed back to the vehicle control system through the execution module. The mining truck adjusts its driving behavior according to the commands and uploads the status information to the cloud platform in real time for subsequent monitoring and analysis.

[0137] This embodiment of the method significantly reduces computational complexity by employing simple and efficient geometric operations such as path expansion and rectangular collision detection, ensuring the system can respond quickly in mining environments and meet the real-time decision-making requirements at the 100Hz level. More importantly, by introducing scene matching technology and a time estimation model, this embodiment of the method enables flexible decision generation based on mining-specific scenarios (such as loading and unloading areas and narrow passages), effectively avoiding computational latency issues. The system can quickly assess the relative positions between mining trucks, predict arrival times, and adjust priorities based on path intersections, ensuring efficiency and safety in multi-vehicle collaboration. These improvements make the present invention more accurate and efficient in complex mining environments, overcoming the computational latency and insufficient real-time performance of existing technologies.

[0138] It is understood that the various method embodiments mentioned above in this application can be combined with each other to form combined embodiments without violating the principle and logic. Due to space limitations, this application will not elaborate further. Those skilled in the art will understand that in the above methods of specific implementation, the specific execution order of each step should be determined by its function and possible internal logic.

[0139] Example 3

[0140] The collision avoidance device for unmanned mining operations provided in this application embodiment is located in an electronic device, and the unmanned mining collision avoidance device may include: The acquisition module is used to acquire, in real time, a global traffic information map, the first state information of the target vehicle, and the second state information of surrounding vehicles. The global traffic information map represents the position and status of all vehicles in the mining area. Surrounding vehicles refer to vehicles around the target vehicle.

[0141] The generation module is used to generate a first expansion path corresponding to the target vehicle and a second expansion path corresponding to other vehicles based on the first state information, the second state information, and the global traffic information map. Both the first and second expansion paths are safe driving areas generated after the corresponding paths are expanded.

[0142] The control module is used to control the target vehicle to avoid a collision with a potential collision vehicle if a collision point exists between the first expansion path and any second expansion path. The potential collision vehicle refers to other vehicles on the second expansion path where a collision point exists.

[0143] In some implementations, the acquisition module is specifically used for: Real-time global traffic information map is obtained from the cloud platform via vehicle-to-cloud communication. This global traffic information map is generated by the cloud platform based on the status information uploaded by all vehicles in the mining area. The first status information of the target vehicle is obtained. Second status information of surrounding vehicles is obtained in real-time via V2X communication.

[0144] In some implementations, the generation module is specifically used for: The first path of the target vehicle is determined from the first state information. The first path is then normalized based on the first state information to generate a first expanded path. Second paths of other vehicles are determined based on the second state information and the global traffic information map. The corresponding second paths are then normalized based on the second state information and the global traffic information map to generate corresponding second expanded paths.

[0145] In some implementations, when controlling the target vehicle to avoid a collision with a potentially colliding vehicle, the control module is specifically used for: The collision scenario is determined based on the collision point, the first state information, and the third state information corresponding to the potential collision vehicles. The target vehicle is then controlled to avoid a collision with the potential collision vehicles based on the collision scenario.

[0146] In some implementations, the collision scenario is a longitudinal, same-direction conflict scenario. Longitudinal, same-direction conflict scenarios include following-vehicle scenarios.

[0147] When the control module controls the target vehicle to avoid a collision with a potential collision vehicle based on the collision scenario, it is specifically used for: Control the target vehicle to maintain a safe distance from vehicles that may collide, in order to avoid a collision.

[0148] In some implementations, the collision scenario is a head-on conflict scenario. Head-on conflict scenarios include passing scenarios.

[0149] When the control module controls the target vehicle to avoid a collision with a potential collision vehicle based on the collision scenario, it is specifically used for: The vehicle that needs to yield is determined according to the first preset priority rule. If the vehicle that needs to yield is the target vehicle, the target vehicle is controlled to perform the yielding operation before reaching the collision point to avoid a collision.

[0150] In some implementations, the collision scenario is a lateral intersection conflict scenario. Lateral intersection conflict scenarios include intersection scenarios and same-direction merging scenarios.

[0151] When the control module controls the target vehicle to avoid a collision with a potential collision vehicle based on the collision scenario, it is specifically used for: The second preset priority rule determines whether the target vehicle needs to yield or accelerate to overtake. If it is determined that the target vehicle needs to yield or overtake, the target vehicle is controlled to perform the corresponding yielding or accelerating operation before reaching the collision point to avoid a collision.

[0152] In some implementations, the collision avoidance device for unmanned mining operations also includes: The judgment module is used to determine whether there is a spatial intersection between the first expansion path and all second expansion paths. If a spatial intersection is determined to exist, it then determines whether a temporal intersection also exists based on the first state information, the second state information, and the global traffic information map. If a temporal intersection is also determined to exist, a collision point is determined to exist between the first expansion path and any second expansion path.

[0153] The collision avoidance device for unmanned mining provided in this application has the beneficial effects and implementation methods of the collision avoidance methods for unmanned mining provided in Embodiments 1 and 2 of this application. For details, please refer to the specific descriptions of the collision avoidance methods for unmanned mining provided in Embodiments 1 and 2 above. This embodiment will not repeat them here.

[0154] Example 4

[0155] This application also provides an electronic device, which is intended to be various forms of devices with data processing capabilities, such as workbenches, servers, terminals, and other suitable computers. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present application described and / or claimed herein.

[0156] This electronic device includes a processor and memory. The various components are interconnected via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processor processes instructions that execute within the electronic device.

[0157] The memory is the non-transitory computer-readable storage medium provided in this application. The memory stores instructions executable by at least one processor to cause at least one processor to execute the mine unmanned driving collision avoidance method provided in this application. The non-transitory computer-readable storage medium of this application stores computer instructions for causing a computer to execute the mine unmanned driving collision avoidance method provided in this application.

[0158] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the mine unmanned driving collision avoidance method in the embodiments of this application. The processor executes various functional applications and data processing of the electronic device by running the non-transitory software programs, instructions, and modules stored in the memory, thereby implementing the mine unmanned driving collision avoidance method in the above method embodiments.

[0159] The electronic device provided in this application has the beneficial effects and implementation methods of the mine unmanned driving collision avoidance method provided in Embodiments 1 and 2 of this application. For details, please refer to the specific description of the mine unmanned driving collision avoidance method in Embodiments 1 and 2 above. This embodiment will not repeat the description here.

[0160] Example 5

[0161] This embodiment also provides a computer-readable storage medium storing a computer program thereon. When the computer program is executed by a processor, it implements the mine unmanned driving collision avoidance method in Embodiment 1 or Embodiment 2 above.

[0162] The computer-readable storage medium provided in this application embodiment has the beneficial effects and implementation methods of the collision avoidance method for unmanned mining vehicles in Embodiments 1 and 2 of this application. For details, please refer to the specific description of the collision avoidance method for unmanned mining vehicles in Embodiments 1 and 2 above. This embodiment will not repeat the description here.

[0163] As is known to those skilled in the art, computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable program instructions, data structures, program modules, or other data. Computer storage media includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), static random access memory (SRAM), flash memory or other memory technologies, portable compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, as is known to those skilled in the art, communication media typically contain computer-readable program instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0164] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0165] It is understood that the above embodiments are merely exemplary implementations used to illustrate the principles of this application, and this application is not limited thereto. For those skilled in the art, various modifications and improvements can be made without departing from the spirit and substance of this application, and these modifications and improvements are also considered to be within the scope of protection of this application.

Claims

1. A collision avoidance method for unmanned mining operations, characterized in that, include: Real-time acquisition of global traffic information map, first status information of target vehicle and second status information of surrounding vehicles; The global traffic information map is used to represent the location and status of all vehicles in the mining area; the surrounding vehicles are the vehicles around the target vehicle. Based on the first state information, the second state information, and the global traffic information map, a first expansion path corresponding to the target vehicle and a second expansion path corresponding to other vehicles are generated; both the first expansion path and the second expansion path are safe driving areas generated after the corresponding paths are expanded. If a collision point exists between the first expansion path and any of the second expansion paths, the target vehicle is controlled to avoid colliding with a potential collision vehicle; the potential collision vehicle is another vehicle corresponding to the second expansion path where a collision point exists.

2. The method according to claim 1, characterized in that, The real-time acquisition of the global traffic information map, the first state information of the target vehicle, and the second state information of surrounding vehicles includes: The global traffic information map is obtained in real time from the cloud platform via vehicle-to-cloud communication; the global traffic information map is generated by the cloud platform based on the status information uploaded by all vehicles in the mining area. Obtain the first state information of the target vehicle; The second status information of surrounding vehicles is obtained in real time through vehicle-to-everything (V2X) communication.

3. The method according to claim 1, characterized in that, The step of generating a first expansion path corresponding to the target vehicle and a second expansion path corresponding to other vehicles based on the first state information, the second state information, and the global traffic information map includes: Determine the first path of the target vehicle from the first status information; Based on the first state information, the first path is normal-expanded to generate the first expanded path; The second path of the other vehicles is determined based on the second status information and the global traffic information map; Based on the second state information and the global traffic information map, normal expansion is performed on the corresponding second path to generate the corresponding second expanded path.

4. The method according to claim 1, characterized in that, The control of the target vehicle to avoid collision with a potential collision vehicle includes: The collision scenario is determined based on the collision point, the first state information, and the third state information corresponding to the possible collision vehicle. Based on the collision scenario, the target vehicle is controlled to avoid colliding with the potential collision vehicle.

5. The method according to claim 4, characterized in that, The collision scenario is a longitudinal, same-direction conflict scenario; the longitudinal, same-direction conflict scenario includes a following vehicle scenario. The step of controlling the target vehicle to avoid a collision with the potential collision vehicle based on the collision scenario includes: The target vehicle is controlled to maintain a safe distance from the vehicle that may collide, in order to avoid a collision.

6. The method according to claim 4, characterized in that, The collision scenario is a head-on conflict scenario; the head-on conflict scenario includes a passing scenario. The step of controlling the target vehicle to avoid a collision with the potential collision vehicle based on the collision scenario includes: The vehicles that need to give way are determined according to the first preset priority rule; If the vehicle that needs to yield is the target vehicle, then the target vehicle is controlled to perform a yielding operation before reaching the collision point in order to avoid a collision.

7. The method according to claim 4, characterized in that, The collision scenario is a lateral intersection conflict scenario; the lateral intersection conflict scenario includes intersection scenarios and merging scenarios in the same direction. The step of controlling the target vehicle to avoid a collision with the potential collision vehicle based on the collision scenario includes: Determine whether the target vehicle needs to yield or accelerate to overtake based on the second preset priority rule; If it is determined that the target vehicle needs to yield or overtake, the target vehicle is controlled to perform the corresponding yielding operation or speed-up overtaking operation before reaching the collision point, so as to avoid a collision.

8. The method according to claim 1, characterized in that, If a collision point exists between the first expansion path and any of the second expansion paths, before controlling the target vehicle to avoid a collision with a potentially colliding vehicle, the method further includes: Determine whether there is a spatial intersection between the first expansion path and all second expansion paths; If a spatial intersection is determined, then based on the first state information, the second state information, and the global traffic information map, it is determined whether the spatial intersection also has a temporal intersection. If it is determined that there is a time overlap, then a collision point is determined to exist between the first expansion path and any of the second expansion paths.

9. A collision avoidance device for unmanned mining operations, characterized in that, include: The acquisition module is used to acquire the global traffic information map, the first state information of the target vehicle, and the second state information of the surrounding vehicles in real time. The global traffic information map is used to represent the location and status of all vehicles in the mining area; the surrounding vehicles are the vehicles around the target vehicle. The generation module is used to generate a first expansion path corresponding to the target vehicle and a second expansion path corresponding to other vehicles based on the first state information, the second state information, and the global traffic information map; both the first expansion path and the second expansion path are safe driving areas generated after the corresponding paths are expanded. The control module is configured to control the target vehicle to avoid a collision with a potential collision vehicle if a collision point exists between the first expansion path and any of the second expansion paths; the potential collision vehicle is another vehicle corresponding to the second expansion path where a collision point exists.

10. An electronic device, characterized in that, include: Memory and processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the mine unmanned driving collision avoidance method as described in any one of claims 1 to 8.

Images