Mining area vehicle anti-collision method and device and electronic equipment

Abstract

The invention provides a mining area vehicle anti-collision method and device and electronic equipment, and relates to the technical field of mining area vehicle automation. The method comprises the following steps: acquiring first state information and obstacle information of a vehicle in a target mining area in real time; the obstacle information is state related information of dynamic obstacles and static obstacles around the vehicle in the target mining area; the dynamic obstacles at least comprise other mining area vehicles around the target mining area vehicle; generating a future trajectory collision result and a future distance according to the first state information and the obstacle information; and controlling the target mining area vehicle to execute a corresponding anti-collision strategy according to the future trajectory collision result and the future distance. According to the method, through future trajectory collision detection of the target mining area vehicle and the dynamic obstacle and distance prediction of the target mining area vehicle and the static obstacle, the anti-collision effect of the mining area vehicle in the strip mining area production environment is improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of mine vehicle automation, and particularly relates to a mine vehicle anti-collision method, device and electronic equipment. BACKGROUND

[0002] With the development of vehicle intelligent driving technology, intelligent driving vehicles are gradually applied to open-pit mine production environments. The open-pit mine production environment has the characteristics of unstructured road, limited sight distance, complex traffic flow and the like, and the road conditions are extremely complex.

[0003] Mine vehicles are easily disturbed by bad weather and dust in such an environment, which affects the perception range of intelligent driving, and there are a large number of blind areas in intersection or curve scenes, which easily causes vehicle collision problems. Therefore, the anti-collision effect of mine vehicles needs to be further optimized. SUMMARY

[0004] The technical problem to be solved by the present application is to solve the above-mentioned deficiencies existing in the prior art, and to provide a mine vehicle anti-collision method, device and electronic equipment. The use of the mine vehicle anti-collision method can improve the anti-collision effect of mine vehicles.

[0005] In a first aspect, an embodiment of the present application provides a mine vehicle anti-collision method, comprising:

[0006] real-time acquisition of first state information of a target mine vehicle and obstacle information; the obstacle information is state-related information of dynamic obstacles and static obstacles around the target mine vehicle; the dynamic obstacles at least include other mine vehicles around the target mine vehicle;

[0007] generation of future trajectory collision results and future distances according to the first state information and the obstacle information; the future trajectory collision results are collision results between the future trajectory of the target mine vehicle and the future trajectory of the dynamic obstacles; the future distances are the predicted minimum distances between the target mine vehicle and the static obstacles;

[0008] control of the target mine vehicle to execute corresponding anti-collision strategies according to the future trajectory collision results and the future distances.

[0009] In some embodiments of the first aspect, the dynamic obstacles further include non-vehicle obstacles;

[0010] The real-time acquisition of the first state information of the target mine vehicle and the obstacle information comprises:

[0011] real-time acquisition of vehicle motion data collected by each sensor in the target mine vehicle to generate the first state information;

[0012] real-time acquisition of motion data of non-vehicle obstacles and position information of static obstacles collected by each sensor.

[0013] acquire state information of other mining area vehicles around the target mining area vehicle through vehicle-to-vehicle V2V communication, and generate second state information;

[0014] generate obstacle information according to the motion data, the position information, and the second state information of the non-vehicle obstacle.

[0015] In some embodiments of the first aspect, the future trajectory collision result and the future distance are generated according to the first state information and the obstacle information, including:

[0016] convert the first state information and the obstacle information to a local coordinate system with the mining area differential base station as the origin;

[0017] generate the future trajectory collision result according to the converted first state information, the motion data of the non-vehicle obstacle, and the second state information;

[0018] generate the future distance according to the converted first state information and the position information.

[0019] In some embodiments of the first aspect, the future trajectory collision result is generated according to the converted first state information, the motion data of the non-vehicle obstacle, and the second state information, including:

[0020] generate a first future trajectory corresponding to the target mining area vehicle according to the converted first state information;

[0021] generate a second future trajectory corresponding to the dynamic obstacle according to the converted motion data of the non-vehicle obstacle and the second state information;

[0022] perform collision detection on the first future trajectory and the second future trajectory based on the Separating Axis Theorem SAT to generate the future trajectory collision result.

[0023] In some embodiments of the first aspect, the second future trajectory corresponding to the dynamic obstacle is generated according to the converted motion data of the non-vehicle obstacle and the second state information, including:

[0024] perform nonlinear trajectory prediction on the converted second state information of the first adjacent vehicle using a Constant Velocity Constant Angular Velocity CVCYR model to generate a corresponding first predicted trajectory; the first adjacent vehicle is another mining area vehicle that is tracked for more than a preset number of frames and is adjacent to the target mining area vehicle;

[0025] perform linear trajectory prediction on the converted motion data of the non-vehicle obstacle and the converted second state information of the second adjacent vehicle using a Kalman Filter KF model to generate a corresponding second predicted trajectory; the second adjacent vehicle is another mining area vehicle that is tracked for less than or equal to a preset number of frames and is adjacent to the target mining area vehicle;

[0026] generate a second future trajectory according to the first predicted trajectory and the second predicted trajectory.

[0027] In some embodiments of the first aspect, the future distance is generated according to the converted first state information and the position information, comprising:

[0028] The future distance between the target mining area vehicle and the static obstacle is determined according to the first state information and the position information by using a Separated Axis Theorem (SAT).

[0029] In some embodiments of the first aspect, the future trajectory collision result comprises a collision time.

[0030] The target mining area vehicle is controlled to execute a corresponding anti-collision strategy according to the future trajectory collision result and the future distance, comprising:

[0031] A collision risk level is determined according to the collision time and a corresponding preset time threshold, the future distance and a corresponding preset distance threshold.

[0032] The target mining area vehicle is controlled to execute a corresponding anti-collision strategy according to the collision risk level.

[0033] In some embodiments of the first aspect, the target mining area vehicle is controlled to execute a corresponding anti-collision strategy according to the collision risk level, comprising:

[0034] If the collision risk level is a high level, the target mining area vehicle is controlled to slow down and send a cooperative obstacle avoidance indication to other mining area vehicles in the surrounding of the target mining area vehicle through vehicle-to-vehicle (V2V) communication.

[0035] If the collision risk level is a medium level, a moving direction of the target mining area vehicle is adjusted to a direction in which no collision occurs.

[0036] Based on the same inventive concept, the second aspect, the embodiments of the present application further provide a mining area vehicle anti-collision device, comprising:

[0037] An acquisition module is configured to acquire first state information of a target mining area vehicle and obstacle information in real time; the obstacle information is state-related information of dynamic obstacles and static obstacles in the surrounding of the target mining area vehicle; the dynamic obstacles at least include other mining area vehicles in the surrounding of the target mining area vehicle.

[0038] A generation module is configured to generate a future trajectory collision result and a future distance according to the first state information and the obstacle information; the future trajectory collision result is a collision result between a future trajectory of the target mining area vehicle and a future trajectory of the dynamic obstacle; the future distance is a predicted minimum distance between the target mining area vehicle and the static obstacle.

[0039] The control module is configured to control the target mining area vehicle to execute a corresponding anti-collision strategy according to the future trajectory collision result and the future distance.

[0040] In some embodiments of the second aspect, the dynamic obstacle further includes a non-vehicle obstacle.

[0041] The acquisition module is specifically configured to:

[0042] The acquisition module is specifically configured to:

[0043] In some embodiments of the second aspect, the generation module is specifically configured to:

[0044] The generation module is specifically configured to:

[0045] In some embodiments of the second aspect, when generating the future trajectory collision result according to the converted first state information, the motion data of the non-vehicle obstacle, and the second state information, the generation module is specifically configured to:

[0046] The generation module is specifically configured to:

[0047] In some embodiments of the second aspect, when generating the second future trajectory corresponding to the dynamic obstacle according to the converted motion data of the non-vehicle obstacle and the second state information, the generation module is specifically configured to:

[0048] The constant velocity constant angular velocity CVCYR model is used to perform nonlinear trajectory prediction on the second state information converted by the first adjacent vehicle to generate a corresponding first predicted trajectory; the first adjacent vehicle is another mining area vehicle adjacent to the target mining area vehicle and tracked for more than a preset number of frames; the linear Kalman filter KF model is used to perform linear trajectory prediction on the motion data converted by the non-vehicle obstacle and the second state information converted by the second adjacent vehicle to generate a corresponding second predicted trajectory; the second adjacent vehicle is another mining area vehicle adjacent to the target mining area vehicle and tracked for less than or equal to a preset number of frames; and the second future trajectory is generated according to the first predicted trajectory and the second predicted trajectory.

[0049] In some embodiments of the second aspect, the generating module, when generating the future distance according to the converted first state information and the position information, is specifically configured to:

[0050] The separation axis theorem SAT is used to determine the future distance between the target mining area vehicle and the static obstacle according to the first state information and the position information.

[0051] In some embodiments of the second aspect, the future trajectory collision result includes a collision time.

[0052] The control module is specifically configured to:

[0053] The collision risk level is determined according to the collision time and a corresponding preset time threshold, the future distance and a corresponding preset distance threshold; and the target mining area vehicle is controlled to execute a corresponding anti-collision strategy according to the collision risk level.

[0054] In some embodiments of the second aspect, when the control module controls the target mining area vehicle to execute the corresponding anti-collision strategy according to the collision risk level, the control module is specifically configured to:

[0055] If the collision risk level is a high level, the target mining area vehicle is controlled to slow down and send a cooperative obstacle avoidance indication to other mining area vehicles in the surrounding of the target mining area vehicle through vehicle-to-vehicle V2V communication; and if the collision risk level is a medium level, the moving direction of the target mining area vehicle is adjusted to a direction that will not cause a collision.

[0056] Based on the same inventive concept, in a third aspect, the embodiments of the present application also provide an electronic device, which comprises a memory and a processor;

[0057] The memory stores computer execution instructions;

[0058] The processor executes the computer execution instructions stored in the memory to implement the mining area vehicle anti-collision method according to any one of the first aspect.

[0059] Based on the same inventive concept, in a fourth aspect, the embodiments of the present application also provide a computer-readable storage medium, the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are executed by a processor to implement the mine vehicle anti-collision method according to any one of the first aspect.

[0060] According to the mine vehicle anti-collision method, device and electronic equipment provided by the embodiments of the present application, the first state information and the obstacle information of the target mine vehicle are acquired in real time. The future trajectory collision result and the future distance are generated according to the first state information and the obstacle information. At the same time, the corresponding anti-collision strategy is controlled to be executed by the target mine vehicle according to the future trajectory collision result and the future distance. Thus, through the future trajectory collision detection of the target mine vehicle and the dynamic obstacle (including other vehicles) and the distance prediction of the target mine vehicle and the static obstacle, the collision protection of the dynamic obstacle and the static obstacle in the open-pit mine production environment is taken into account, and thus the anti-collision effect of the mine vehicle in the open-pit mine production environment is improved. BRIEF DESCRIPTION OF DRAWINGS

[0061] Figure 1 Fig. 1 shows a flowchart of a mine vehicle anti-collision method provided by an embodiment of the present application;

[0062] Figure 2 Fig. 2 shows another flowchart of a mine vehicle anti-collision method provided by an embodiment of the present application;

[0063] Figure 3 Fig. 3 shows a whole flowchart of a mine vehicle anti-collision method provided by an embodiment of the present application;

[0064] Figure 4 Fig. 4 shows a trajectory prediction flowchart provided by an embodiment of the present application;

[0065] Figure 5 Fig. 5 shows a collision detection schematic diagram provided by an embodiment of the present application. DETAILED DESCRIPTION

[0066] In order for those skilled in the art to better understand the technical solutions of the present application, the present application will be further described in detail below with reference to the drawings and embodiments.

[0067] The features and exemplary embodiments of various aspects of the present application will be described in detail below. In order to make the purposes, technical solutions and advantages of the present application more clear, the present application will be further described in detail below with reference to the drawings and specific embodiments. It should be understood that the specific embodiments described herein are only configured to explain the present application and are not configured to limit the present application. The present application can be implemented without some of these specific details for those skilled in the art. The following description of the embodiments is only to provide a better understanding of the present application by showing examples of the present application.

[0068] It is to be understood that the terminology used herein such as first and second, and the like, is only intended to distinguish one entity or operation from another entity or operation, and is not intended to imply that such entities or operations are necessarily related or ordered in time. Also, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0069] As described in the background, the mine vehicles are susceptible to bad weather and dust interference in the open-pit mine production environment, affecting the perception range of intelligent driving, and there are a large number of blind areas in scenes such as intersections or curves, which are prone to vehicle collision problems, and therefore, the anti-collision effect of the mine vehicle needs to be further optimized.

[0070] Embodiment 1

[0071] The mine vehicle anti-collision method provided by the embodiments of the present application is applied to an electronic device. The electronic device can be a computer, or a device in the computer for implementing the mine vehicle anti-collision method. The electronic device can be a vehicle-mounted terminal or a vehicle-mounted unit, which is not limited in the embodiments of the present application. The following takes an example that the mine vehicle anti-collision method is executed by the electronic device.

[0072] As shown in Figure 1 The mine vehicle anti-collision method provided by the embodiments of the present application can include steps S101 to S103.

[0073] S101, real-time acquisition of first state information of a target mine vehicle and obstacle information. The obstacle information is state-related information of dynamic obstacles and static obstacles around the target mine vehicle. The dynamic obstacles at least include other mine vehicles around the target mine vehicle.

[0074] For example, if the electronic device is arranged on the target mine vehicle, the target mine vehicle is the vehicle to which the electronic device is deployed. If the electronic device is arranged in the control center, the target mine vehicle is the vehicle to be controlled and dispatched. In actual application, the electronic device can be deployed on the target mine vehicle, so that each mine vehicle can independently execute the anti-collision strategy.

[0075] Exemplarily, the first state information of the target mine area vehicle includes information such as latitude and longitude coordinates, speed, acceleration, heading angle, yaw rate, steering wheel angle, and turn signal state. The first state information can be used to predict the future trajectory of the target mine area vehicle, which is generated based on the first state information obtained in real time.

[0076] The obstacle information is state-related information of obstacles around the target mine area vehicle. The dynamic obstacles include other mine area vehicles and non-vehicle obstacles such as pedestrians and other mobile machines around the target mine area vehicle. The static obstacles include equipment, soil piles, and placed objects.

[0077] The state-related information of the dynamic obstacles can include motion data such as speed, acceleration, position coordinates, and distance from the target mine area vehicle. The state-related information of the other mine area vehicles can include vehicle identification, latitude and longitude coordinates, speed, acceleration, heading angle, yaw rate, steering wheel angle, and turn signal state.

[0078] The state-related information of the static obstacles can include position information such as distance from the target mine area vehicle and position coordinates.

[0079] Exemplarily, the way to obtain the first state information and the obstacle information can be through sensor data collection combined with vehicle-to-vehicle (V2V) communication data transmission. Through V2V communication, vehicle state information is exchanged in real time, and sensor-collected perception data is fused, providing a basis for mine area vehicle collision avoidance.

[0080] S102, generating future trajectory collision results and future distances according to the first state information and the obstacle information. The future trajectory collision results are the collision results between the future trajectory of the target mine area vehicle and the future trajectory of the dynamic obstacles. The future distances are the predicted minimum distances between the target mine area vehicle and the static obstacles.

[0081] Exemplarily, the future trajectory of the target mine area vehicle can be predicted according to the first state information, the future trajectory of the dynamic obstacles and the future distance change between the static obstacles and the target mine area vehicle can be predicted according to the obstacle information, thereby generating the future trajectory collision results between the target mine area vehicle and the dynamic obstacles and the future distances between the target mine area vehicle and the static obstacles.

[0082] S103, controlling the target mine area vehicle to execute corresponding collision avoidance strategies according to the future trajectory collision results and the future distances.

[0083] Exemplarily, the future trajectory collision results include whether to collide, collision time, collision position, etc., and the future distances represent the shortest distances between the target mine area vehicle and the static obstacles in a future period of time.

[0084] The future collision of the target mining area vehicle can be determined according to the future trajectory collision result and the future distance, and a corresponding anti-collision strategy is generated and executed based on the future collision.

[0085] For example, if the collision time in the future trajectory collision result is short and / or the future distance is short, the target mining area vehicle can be controlled to slow down and issue a collision avoidance instruction to other mining area vehicles, thereby avoiding collision with other vehicles or obstacles.

[0086] If the collision time in the future trajectory collision result is moderate and / or the future distance is moderate, the target mining area vehicle can be controlled to adjust the driving direction, thereby avoiding collision with other vehicles or obstacles.

[0087] According to the mining area vehicle anti-collision method provided by the embodiment of the present application, the first state information of the target mining area vehicle and the obstacle information are obtained in real time. The future trajectory collision result and the future distance are generated according to the first state information and the obstacle information. At the same time, the target mining area vehicle is controlled to execute the corresponding anti-collision strategy according to the future trajectory collision result and the future distance. Thus, through the future trajectory collision detection of the target mining area vehicle and the dynamic obstacle (including other vehicles) and the distance prediction of the target mining area vehicle and the static obstacle, the collision protection of the dynamic obstacle and the static obstacle in the open-pit mining area production environment is considered, thereby improving the anti-collision effect of the mining area vehicle in the open-pit mining area production environment.

[0088] Embodiment 2

[0089] As shown in Figure 2 The mining area vehicle anti-collision method provided by the embodiment of the present application is based on the mining area vehicle anti-collision method provided by the embodiment 1 of the present application, and further description can be included in steps S201 to S205.

[0090] S201, the first state information of the target mining area vehicle and the obstacle information are obtained in real time.

[0091] In some embodiments, the dynamic obstacle further includes a non-vehicle obstacle.

[0092] S201 can be specifically as follows:

[0093] The vehicle motion data collected by each sensor in the target mining area vehicle is obtained in real time to generate the first state information.

[0094] The motion data of the non-vehicle obstacle collected by each sensor and the position information of the static obstacle are obtained in real time.

[0095] The state information of other mining area vehicles around the target mining area vehicle is obtained through vehicle-to-vehicle V2V communication to generate the second state information.

[0096] The obstacle information is generated according to the motion data, position information and second state information of the non-vehicle obstacle.

[0097] For example, the sensor can be a speed sensor, a laser radar, a camera, etc. The non-vehicle obstacle includes a pedestrian, another mobile machine, etc. The motion data includes speed, acceleration, position coordinates, distance from the target mine vehicle, etc. The vehicle motion data includes latitude and longitude coordinates, speed, acceleration, heading angle, yaw rate, steering wheel angle, turn signal state, etc.

[0098] For example, the second state information is generated by receiving the state information broadcast by other mine vehicles in real time through vehicle-to-vehicle (V2V) communication. The second state information can include vehicle identification, latitude and longitude coordinates, speed, acceleration, heading angle, yaw rate, steering wheel angle, turn signal state, etc.

[0099] Through V2V communication, the lead perception of obstacles outside the line-of-sight blind area is realized, and collision accidents caused by dust raising, shielding and curves are effectively avoided, thereby providing a core technical guarantee for mine safety production.

[0100] S202, converting the first state information and the obstacle information to a local East-North-Up coordinate system with a mine differential base station as the origin.

[0101] For example, the mine differential base station is a global navigation satellite system receiving station fixed at a known accurate position in the mine. In the local East-North-Up coordinate system (ENU), the vehicle moves "forward" ≈ the north coordinate increases, and the vehicle moves "right" ≈ the east coordinate increases, which is consistent with the direction of vehicle travel, has high intuitiveness, and is very suitable for closed, high-precision and strong-collaboration application scenarios such as mines.

[0102] By converting the first state information and the obstacle information to the local East-North-Up coordinate system with the mine differential base station as the origin, all information such as distance, position, speed, etc. can be unified to the same high-precision reference system, thereby providing a basis for collision avoidance and collaborative control.

[0103] S203, generating a future trajectory collision result according to the converted first state information, the motion data of the non-vehicle obstacle and the second state information.

[0104] For example, the converted first state information, the motion data of the non-vehicle obstacle and the second state information are in the same high-precision reference system, so that the future trajectories of the target mine vehicle, the non-vehicle obstacle and other mine vehicles can be directly calculated, thereby generating the corresponding future trajectory collision result.

[0105] In some embodiments, S203 is specifically configured to:

[0106] The first future trajectory of the target mining area vehicle is generated based on the converted first state information.

[0107] The second future trajectory corresponding to the dynamic obstacle is generated based on the motion data and second state information of the converted non-vehicle obstacles.

[0108] Based on the Separating Axis Theorem (SAT), collision detection is performed on the first and second future trajectories to generate collision results for the future trajectories.

[0109] For example, the constant velocity and constant angular velocity (CVCYR) model can be used to generate the first future trajectory of the vehicle in the target mining area based on the first state information.

[0110] The converted motion data of non-vehicle obstacles can generate a corresponding second future trajectory, and the converted second state information can also generate a corresponding second future trajectory. In practical applications, if non-vehicle obstacles and other mining vehicles coexist, a corresponding second future trajectory can be generated for each dynamic obstacle (i.e., non-vehicle obstacles and other mining vehicles). During collision detection, the first future trajectory and each second future trajectory must be subjected to collision detection.

[0111] For example, the core idea of ​​the Separating Axis Theorem (SAT) is that if two convex objects (polygons / polyhedra) do not collide, then there must exist a straight line (i.e., the separating axis) such that the projections of the two objects onto this line are separated.

[0112] When performing collision detection on the first and second future trajectories based on the Separation Axis Theorem (SAT), a vehicle envelope rectangle (Oriented Bounding Box, OBB) that varies over time is generated for the predicted trajectory points of vehicles and dynamic obstacles in the target mining area. This allows for the determination of whether the vehicle envelope rectangles overlap on all possible separation axes.

[0113] For a potential split axis :

[0114] Project all vertices of polygon A onto Above, the scalar range is obtained. .

[0115] Project all vertices of polygon B onto Above, the scalar range is obtained. .

[0116] if Or then the axis is a separating axis, and the two polygons do not collide.

[0117] If no such separating axis is found among all candidate axes (the normals of all edges of both polygons), then the two polygons are determined to collide.

[0118] The way of SAT provides an accurate and efficient collision detection means for mining vehicles.

[0119] In some embodiments, the process of generating the second future trajectory corresponding to the dynamic obstacle according to the converted motion data and the second state information of the non-vehicle obstacle can be specifically as follows:

[0120] The constant velocity and constant angular velocity CVCYR model is used to perform nonlinear trajectory prediction on the converted second state information of the first adjacent vehicle to generate a corresponding first predicted trajectory. The first adjacent vehicle is another mining vehicle that is tracked for more than a preset number of frames and is adjacent to the target mining vehicle.

[0121] The linear Kalman filter KF model is used to perform linear trajectory prediction on the converted motion data of the non-vehicle obstacle and the converted second state information of the second adjacent vehicle to generate a corresponding second predicted trajectory. The second adjacent vehicle is another mining vehicle that is tracked for less than or equal to a preset number of frames and is adjacent to the target mining vehicle.

[0122] The second future trajectory is generated according to the first predicted trajectory and the second predicted trajectory.

[0123] For example, the preset number of frames can be set according to actual application, for example, set to 2 frames. The second future trajectory is composed of the first predicted trajectory and the second predicted trajectory.

[0124] For example, for a vehicle obstacle (i.e., another mining vehicle) that is stably tracked for more than two frames, the CVCYR model is used for nonlinear trajectory prediction, and the CVCYR model has better prediction effect for vehicles that turn or change lanes. The state recursive equation is as follows: let represent the state vector at time step , where is the position coordinate, is the heading angle, is the speed. The predicted state at time step is given by the following formula:

[0125]

[0126] wherein: It is the predicted time step (e.g., 0.06 seconds). It is the yaw rate, based on the historical heading angle sequence. The estimate is as follows:

[0127]

[0128] Where: weight It can be used to emphasize recent measurements.

[0129] For non-vehicle obstacles such as pedestrians or transient vehicles, since their movement can be equated to straight-line travel, the linear Kalman filter (KF) model performs better in prediction. It assumes that the eastward and northward velocity components are constant in the northeast-northeast coordinate system. The state vector is... The state transition model and prediction are given by the following equation:

[0130]

[0131]

[0132] in: It is time State estimates, It is the state transition matrix. It estimates the covariance matrix, W. t It is process noise, and its covariance matrix is The measurement update uses the standard Kalman filter equation.

[0133] S204. Generate the future distance based on the converted first state information and location information.

[0134] For example, a first future trajectory is generated based on the first state information, and the distance relationship between the static obstacle and the first future trajectory is determined based on the position information, thereby generating the future distance.

[0135] In some implementations, S204 is specifically used for:

[0136] The Separation Axis Theorem (SAT) is used to determine the future distance between vehicles and static obstacles in the target mining area based on the first state information and position information.

[0137] For example, the purpose of using SAT principle is not to determine instantaneous collision, but to calculate the minimum distance between the envelope rectangle of the target mine area vehicle and the envelope rectangle of the static obstacle when the target mine area vehicle travels along the first future trajectory. The calculation involves finding the minimum translation distance that separates the two objects along the path. For each pose on the target mine area vehicle path, a SAT check is conceptually performed, but the minimum overlap or maximum separation on each axis is calculated to determine the closest distance, thereby generating the future distance between the target mine area vehicle and the static obstacle.

[0138] S205 controls the target mine area vehicle to perform a corresponding anti-collision strategy according to the future trajectory collision result and the future distance.

[0139] In some embodiments, the future trajectory collision result includes a collision time.

[0140] S205 is specifically as follows:

[0141] A collision risk level is determined according to the collision time and a corresponding preset time threshold, the future distance and a corresponding preset distance threshold.

[0142] The target mine area vehicle is controlled to perform a corresponding anti-collision strategy according to the collision risk level.

[0143] For example, the preset time threshold can be set to multiple, for example, divided into three thresholds of high, medium and low. If the collision time is greater than or equal to the high time threshold, it represents a relatively low collision risk, and the moving direction can be slowly adjusted to achieve obstacle avoidance. If the collision time is between the medium time threshold and the high time threshold, it represents a relatively low risk, and a prompt can be issued and the moving direction is adjusted for obstacle avoidance. If the collision time is less than the low time threshold, it represents a relatively high collision risk, and immediate adjustment is required, such as directly reducing speed, adjusting the moving direction, and sending anti-collision coordination instructions to other mine area vehicles.

[0144] Similarly, the preset distance threshold can also be set to multiple, for example, divided into three thresholds of high, medium and low.

[0145] The collision risk level can be determined based on the relationship between the collision time and the corresponding preset time threshold, and the relationship between the future distance and the corresponding preset distance threshold. For example, if the collision time is less than the low time threshold and / or the future distance is less than the low distance threshold, it can be determined that the collision risk level is very high, and immediate speed reduction, moving direction adjustment and anti-collision coordination instructions to other mine area vehicles are required.

[0146] For example, the future trajectory collision result and the future distance can be displayed in the form of a list. For example, a collision information list (whether each obstacle collides, collision time (TTC, Time To Collision), collision point) and future distance of the static obstacle.

[0147] Risk_dynamic: The minimum TTC value of all dynamic obstacles is taken as the main basis for the current dynamic risk. .

[0148] Risk_static: The minimum distance d_static of static obstacles is the risk value itself. The closer the distance, the higher the risk.

[0149] At the same time, a set of configurable threshold parameters (preset time threshold, preset distance threshold) are maintained, which can be calibrated according to the vehicle load, road conditions and safety strategy, which is not limited in this embodiment.

[0150] In some embodiments, the process of controlling the target mine vehicle to execute the corresponding anti-collision strategy according to the collision risk level can be as follows:

[0151] If the collision risk level is high, the target mine vehicle is controlled to slow down, and a cooperative obstacle avoidance instruction is sent to other mine vehicles in the surrounding of the target mine vehicle through vehicle-to-vehicle (V2V) communication.

[0152] If the collision risk level is medium, the moving direction of the target mine vehicle is adjusted to a direction that will not cause a collision.

[0153] For example, when the collision risk level is high, the target mine vehicle executes a speed reduction and sends a V2V instruction to the surrounding other mine vehicles to cooperatively avoid obstacles, and when the collision risk level is medium, the steering is fine-tuned to avoid static obstacles, and at the same time, an audible and visual alarm in the cab can also be triggered.

[0154] Secondly, the obstacle data and trajectory are updated in real time, and if the risk is removed, the normal driving state is gradually restored.

[0155] For example, when controlling the vehicle, the control instruction can be converted into a signal executable by the vehicle's drive-by-wire system (drive, brake, steering) or human-machine interaction interface.

[0156] The mine vehicle anti-collision method of the present embodiment can fundamentally improve safety: through V2V communication, it realizes the advanced perception of obstacles outside the line-of-sight blind area, effectively avoiding collision accidents caused by dust, obstruction, curves, etc., and provides core technical support for mine safety production.

[0157] It can adapt to complex mine scenarios: the differentiated trajectory prediction model (CVCYR / KF) closely combines the motion characteristics of mine vehicles and personnel, and the collision detection algorithm (SAT) can handle rectangular envelopes of any orientation, with high precision and strong reliability.

[0158] Safety and efficiency can be balanced: the hierarchical control strategy based on TTC and distance avoids unnecessary emergency stop, triggers maximum braking only when it is really dangerous, ensures safety and minimizes interference with transportation efficiency.

[0159] The huge economic losses caused by equipment damage and production interruption due to collision can be greatly reduced, and the reliability of automated operation is improved, laying a solid foundation for intelligent upgrading of mines.

[0160] In order to better understand the mine vehicle anti-collision method provided by the embodiments of the present application, an exemplary description will be made in conjunction with a specific application implementation.

[0161] The embodiment realizes anti-collision in the mine truck transportation scenario, takes the main mine truck (target mine vehicle) that undertakes the mine transportation task as the core, matches the V2V communication unit (supports mine over-the-horizon communication), high-precision positioning module (RTK (Real-Time Kinematic, Real-Time Kinematic) and IMU (Inertial Measurement Unit, Inertial Measurement Unit) combination, outputs position, heading angle, speed and other data) and environment perception sensor (radar, camera, identifies obstacle type and position), and road side is supplemented with V2V relay equipment, solves the communication blind area caused by mine dust and terrain obstruction.

[0162] As shown in Figure 3 , the implementation process is developed in a closed loop of "information interaction - trajectory prediction - collision detection - hierarchical control - dynamic adjustment":

[0163] Firstly, the main mine truck broadcasts its own state in real time through V2V communication, and receives the motion data of surrounding dynamic obstacles (such as other mine trucks (neighbor vehicles) and personnel) and the position information of static obstacles (such as material piles and pits). At the same time, the motion data of the main mine truck and the surrounding obstacle information are collected to realize self-positioning and perception data acquisition.

[0164] The data fusion and coordinate unification are carried out in the ENU coordinate system to integrate them into a structured data set.

[0165] As shown in Figure 4 , for different types of dynamic obstacles, the obstacle state history is obtained, and the obstacle type is judged. According to the obstacle type, a differentiated trajectory prediction model (nonlinear model CVCYR for vehicle type and stable tracking of more than or equal to two frames, linear model KF for non-vehicle type and vehicle type that cannot be stably tracked for more than or equal to two frames) is used for trajectory prediction, and the predicted trajectory is output. The static obstacle is based on the actual contour to build a collision area containing safety redundancy, and the main mine truck synchronously predicts its future trajectory and establishes a collision area.

[0166] The predicted trajectory and collision detection are as shown in the figure. Figure 5 As shown in the figure, the purple and blue boxes are the vehicle body, the yellow line represents the distance between the vehicle and an obstacle, the red straight line represents the trajectory direction, the orange represents an obstacle, and there is a collision with the trajectory. Collision detection is performed by the Separated Axis Theorem (SAT), and the time to collision (TTC) and the collision distance with the obstacle are calculated. The decision and control module distinguishes between high, medium and low risk levels based on the time to collision (TTC) and the collision distance with the obstacle.

[0167] Control instructions are generated for different risk levels and executed through the drive-by-wire system. In high-risk situations, the host truck performs deceleration and cooperates with the surrounding obstacles through V2V instructions to avoid obstacles. In medium-risk situations, steering is adjusted to avoid static obstacles, and an audible and visual alarm is triggered in the cab. The system periodically repeats the above process, updates the obstacle data and trajectory in real time, and gradually restores normal driving conditions if the risk is removed.

[0168] This approach is suitable for unstructured road environments in mining areas, can effectively identify dynamic and static obstacles, break through the limitations of perception through V2V communication, balance safety and transportation efficiency through hierarchical control, and avoid truck collision accidents.

[0169] The method of the embodiment builds a real-time state information sharing platform between vehicles through a V2V communication network, breaking the physical limitations of single-vehicle perception. At the same time, a differentiated trajectory prediction algorithm is designed, which uses a constant velocity and constant angle velocity (CVCYR) nonlinear model for vehicles and non-vehicle dynamic obstacles, and a constant speed assumption linear model for high-precision trajectory prediction. Secondly, a comprehensive risk assessment mechanism is proposed that combines dynamic obstacle trajectory collision detection and static obstacle envelope collision detection, based on collision time (TTC) and collision distance, a set of hierarchical warning and control strategies are designed to achieve adaptive response to potential collision risks.

[0170] It can be understood that the above-mentioned various method embodiments mentioned in the present application can be combined with each other to form combined embodiments without violating the principle logic. Due to the limited space, the present disclosure will not be repeated. Those skilled in the art can understand that in the above-mentioned method of the specific implementation, the specific execution order of each step should be determined by its function and possible internal logic.

[0171] Embodiment 3

[0172] The mine vehicle anti-collision device provided by the embodiment of the present application is located in an electronic device, for example, a vehicle-mounted terminal. The mine vehicle anti-collision device can include:

[0173] The acquisition module is configured to acquire first state information of the target mining area vehicle and obstacle information in real time. The obstacle information is state-related information of dynamic obstacles and static obstacles around the target mining area vehicle. The dynamic obstacles at least include other mining area vehicles around the target mining area vehicle.

[0174] The generation module is configured to generate a future trajectory collision result and a future distance according to the first state information and the obstacle information. The future trajectory collision result is a collision result between a future trajectory of the target mining area vehicle and a future trajectory of the dynamic obstacles. The future distance is a predicted minimum distance between the target mining area vehicle and the static obstacles.

[0175] The control module is configured to control the target mining area vehicle to execute a corresponding anti-collision strategy according to the future trajectory collision result and the future distance.

[0176] In some embodiments, the dynamic obstacles further include non-vehicle obstacles.

[0177] The acquisition module is specifically configured to:

[0178] The acquisition module is specifically configured to:

[0179] In some embodiments, the generation module is specifically configured to:

[0180] The generation module is specifically configured to:

[0181] In some embodiments, when the generation module generates the future trajectory collision result according to the converted first state information, the motion data of the non-vehicle obstacles and the second state information, the generation module is specifically configured to:

[0182] The generation module is specifically configured to:

[0183] In some embodiments, the generating module, when generating the second future trajectory according to the converted motion data of the non-vehicle obstacle and the second state information, is specifically configured to:

[0184] The constant velocity and constant angular velocity CVCYR model is used to perform nonlinear trajectory prediction on the converted second state information of the first adjacent vehicle to generate a corresponding first predicted trajectory. The first adjacent vehicle is another mining area vehicle adjacent to the target mining area vehicle and tracked for more than a preset number of frames. The linear Kalman filter KF model is used to perform linear trajectory prediction on the converted motion data of the non-vehicle obstacle and the converted second state information of the second adjacent vehicle to generate a corresponding second predicted trajectory. The second adjacent vehicle is another mining area vehicle adjacent to the target mining area vehicle and tracked for less than or equal to a preset number of frames. The second future trajectory is generated according to the first predicted trajectory and the second predicted trajectory.

[0185] In some embodiments, the generating module, when generating the future distance according to the converted first state information and the position information, is specifically configured to:

[0186] The separation axis theorem SAT is used to determine the future distance between the target mining area vehicle and the static obstacle according to the first state information and the position information.

[0187] In some embodiments, the future trajectory collision result includes a collision time.

[0188] The control module is specifically configured to:

[0189] The collision risk level is determined according to the collision time and a corresponding preset time threshold, the future distance and a corresponding preset distance threshold. The target mining area vehicle is controlled to execute a corresponding anti-collision strategy according to the collision risk level.

[0190] In some embodiments, the control module, when controlling the target mining area vehicle to execute the corresponding anti-collision strategy according to the collision risk level, is specifically configured to:

[0191] If the collision risk level is a high level, the target mining area vehicle is controlled to slow down and send a cooperative obstacle avoidance indication to other mining area vehicles in the surrounding of the target mining area vehicle through vehicle-to-vehicle V2V communication. If the collision risk level is a medium level, the moving direction of the target mining area vehicle is adjusted to a direction in which no collision occurs.

[0192] The mining area vehicle anti-collision device provided by the embodiments of the present application has the beneficial effects and implementation manners of the mining area vehicle anti-collision methods provided by embodiments 1 and 2 of the present application. For specific descriptions of the mining area vehicle anti-collision methods, refer to the specific descriptions of the mining area vehicle anti-collision methods in embodiments 1 and 2 above. The present embodiment will not be described again here.

[0193] Embodiment 4

[0194] The electronic device is intended to be implemented in various forms of devices having data processing functions, such as a vehicle terminal and other suitable control devices. The components shown herein, their connections and relationships, and their functions, are merely examples and are not intended to limit the implementations described and / or claimed in this document.

[0195] The electronic device includes a processor and a memory. The various components are interconnected by different buses, and can be mounted on a common main board or otherwise as needed. The processor can process instructions executed within the electronic device.

[0196] The memory is a non-transitory computer readable storage medium provided by the present application. The memory stores instructions executable by at least one processor to cause the at least one processor to execute the mine vehicle anti-collision method provided by the present application. The non-transitory computer readable storage medium of the present application stores computer instructions for causing a computer to execute the mine vehicle anti-collision method provided by the present application.

[0197] The 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 vehicle anti-collision method in the embodiments of the present 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, that is, implements the mine vehicle anti-collision method in the above method embodiments.

[0198] The electronic device provided by the embodiments of the present application has the beneficial effects and implementation manners of the mine vehicle anti-collision method provided by embodiments 1 and 2 of the present application. For specific descriptions of the mine vehicle anti-collision method, please refer to the above embodiments 1 and 2. This embodiment will not be repeated here.

[0199] Embodiment 5

[0200] The embodiment also provides a computer readable storage medium, which stores a computer program. When the computer program is executed by a processor, the mine vehicle anti-collision method in embodiments 1 or 2 is implemented.

[0201] The computer readable storage medium provided by the embodiments of the present application has the beneficial effects and implementation manners of the mine vehicle anti-collision method provided by embodiments 1 and 2 of the present application. For specific descriptions of the mine vehicle anti-collision method, please refer to the above embodiments 1 and 2. This embodiment will not be repeated here.

[0202] As is well known to those of ordinary skill in the art, computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storage of 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), electrically erasable programmable read only memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by a computer. Further, it is well known to those of ordinary skill in the art that communication media typically embodies computer readable program instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media. In this document, the terms "computer storage medium" and "computer readable medium" are used to generally refer to media such as removable storage drives, memory, certain hardware storage components, or storage media of memory devices (e.g., random access memory, or RAM) such as computer storage media and signals. Accordingly, the disclosure is considered to be drawn, at least in part, to a computer readable medium containing instructions for execution by a processor such that the instructions, which define actions to be taken by the processor, can be read into a machine for execution by the machine. The machine could be of any of a variety of

[0203] Computer readable program instructions described herein can be downloaded to respective computing / processing devices from a computer readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network and / or a wireless network. The network can comprise copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers and / or edge servers. A network adapter card or network interface in each computing / processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium within the respective computing / processing device.

[0204] It is to be understood that the above implementation is merely exemplary implementation adopted for illustrating the principles of the present application, and the present application is not limited thereto. Various modifications and improvements can be made by those of ordinary skill in the art without departing from the spirit and scope of the present application, and these modifications and improvements are also considered to be within the scope of protection of the present application.

Claims

1. A method of preventing a collision of a mine site vehicle, characterized by, The method comprises the following steps: real-time acquisition of first state information and obstacle information of a target mine vehicle; the obstacle information is the state-related information of dynamic obstacles and static obstacles around the target mine vehicle; the dynamic obstacles at least include other mine vehicles around the target mine vehicle; generation of future trajectory collision results and future distances according to the first state information and the obstacle information; the future trajectory collision results are the collision results between the future trajectory of the target mine vehicle and the future trajectory of the dynamic obstacles; the future distance is the predicted minimum distance between the target mine vehicle and the static obstacles; control of the target mine vehicle to execute corresponding anti-collision strategies according to the future trajectory collision results and the future distance.

2. The method of claim 1, wherein, The dynamic obstacles also include non-vehicle obstacles; The real-time acquisition of the first state information and the obstacle information of the target mine vehicle comprises the following steps: real-time acquisition of vehicle motion data collected by each sensor in the target mine vehicle to generate the first state information; real-time acquisition of motion data of the non-vehicle obstacles collected by each sensor and position information of the static obstacles; acquisition of state information of other mine vehicles around the target mine vehicle through vehicle-to-vehicle (V2V) communication to generate second state information; generation of the obstacle information according to the motion data of the non-vehicle obstacles, the position information, and the second state information.

3. The method of claim 2, wherein, The generation of the future trajectory collision results and the future distances according to the first state information and the obstacle information comprises the following steps: conversion of the first state information and the obstacle information to a local coordinate system of the northeast sky with a mine differential base station as the origin; generation of the future trajectory collision results according to the converted first state information, the motion data of the non-vehicle obstacles, and the second state information; generation of the future distance according to the converted first state information and the position information.

4. The method of claim 3, wherein, The generation of the future trajectory collision results according to the converted first state information, the motion data of the non-vehicle obstacles, and the second state information comprises the following steps: generation of a first future trajectory corresponding to the target mine vehicle according to the converted first state information; generation of a second future trajectory corresponding to the dynamic obstacles according to the converted motion data of the non-vehicle obstacles and the second state information; collision detection of the first future trajectory and the second future trajectory based on the separation axis theorem (SAT) to generate the future trajectory collision results.

5. The method of claim 4, wherein, The generation of the second future trajectory corresponding to the dynamic obstacles according to the converted motion data of the non-vehicle obstacles and the second state information comprises the following steps: nonlinear trajectory prediction of the converted second state information of a first adjacent vehicle using a constant velocity and constant angular velocity (CVCYR) model to generate a corresponding first predicted trajectory; the first adjacent vehicle is another mine vehicle that is tracked for more than a preset number of frames and is adjacent to the target mine vehicle; The linear Kalman filter KF model is used to perform linear trajectory prediction on the converted motion data of the non-vehicle obstacle and the converted second state information of the second adjacent vehicle, to generate a corresponding second predicted trajectory; the second adjacent vehicle is another mining area vehicle adjacent to the target mining area vehicle and tracked for less than or equal to a preset number of frames; The second future trajectory is generated according to the first predicted trajectory and the second predicted trajectory.

6. The method of claim 3, wherein, The future distance is generated according to the converted first state information and the position information, including: The separation axis theorem SAT is used to determine the future distance between the target mining area vehicle and the static obstacle according to the first state information and the position information.

7. The method of claim 1, wherein, The future trajectory collision result includes a collision time; The target mining area vehicle is controlled to execute a corresponding anti-collision strategy according to the future trajectory collision result and the future distance, including: A collision risk level is determined according to the collision time and a corresponding preset time threshold, the future distance and a corresponding preset distance threshold; The target mining area vehicle is controlled to execute a corresponding anti-collision strategy according to the collision risk level.

8. The method of claim 7, wherein, The target mining area vehicle is controlled to execute a corresponding anti-collision strategy according to the collision risk level, including: If the collision risk level is high, the target mining area vehicle is controlled to slow down, and a cooperative obstacle avoidance indication is sent to other mining area vehicles in the surrounding of the target mining area vehicle through vehicle-to-vehicle V2V communication; If the collision risk level is medium, the moving direction of the target mining area vehicle is adjusted to a direction that will not cause a collision.

9. A mine site vehicle anti-collision device, characterized by, Including: An acquisition module is configured to acquire first state information of a target mining area vehicle and obstacle information in real time; The obstacle information is state-related information of dynamic obstacles and static obstacles in the surrounding of the target mining area vehicle; the dynamic obstacles at least include other mining area vehicles in the surrounding of the target mining area vehicle; A generation module is configured to generate a future trajectory collision result and a future distance according to the first state information and the obstacle information; the future trajectory collision result is a collision result between a future trajectory of the target mining area vehicle and a future trajectory of the dynamic obstacle; The future distance is a predicted minimum distance between the target mining area vehicle and the static obstacle; A control module is configured to control the target mining area vehicle to execute a corresponding anti-collision strategy according to the future trajectory collision result and the future distance.

10. An electronic device, comprising: Including: A memory and a processor; The memory stores computer execution instructions; The processor executes the computer execution instructions stored in the memory to implement the mining area vehicle anti-collision method according to any one of claims 1 to 8.

Images