Vehicle dumping position parking control method and device, vehicle and storage medium

Abstract

The invention provides a dumping position parking control method and device of a vehicle, the vehicle and a storage medium. According to the method, decoding and permission verification are carried out on dumping position planning data to obtain corresponding target dumping position information; based on the current position attitude information and the target dumping position information of the vehicle, a path planning algorithm is called to generate a dumping operation track, and navigation is performed based on the dumping operation track to complete dumping operation; when the vehicle is driven to the target dumping position and backs up, the information of the dumping position behind the vehicle and the vehicle operation data are collected in real time through a collecting unit; and inputting the dumping position information and the vehicle operation data into a pre-constructed terminal point judgment model, identifying whether the vehicle reaches a dumping terminal point or not through the terminal point judgment model, if so, triggering vehicle braking and parking control, and controlling a lifting device to complete unloading operation. The problem that an existing fixed coordinate navigation vehicle is difficult to accurately stop at a real dumping position during dumping, and consequently dumping operation is affected is solved.

Description

Technical Field

[0001] This application relates to the field of vehicle control technology in mining areas, specifically a method, device, vehicle, and storage medium for controlling the parking of vehicles at their dumping positions. Background Technology

[0002] In open-pit mines or large-scale earthmoving projects, dump trucks need to accurately unload materials into designated dumping areas. Traditional dumping operations mainly rely on the driver's visual judgment or simple sensor positioning to complete the reversing operation. However, because the earthen wall constantly changes as material accumulates during dumping operations, fixed-coordinate navigation strategies are difficult to adapt to the dynamic position of the actual dumping wall on site. This can easily lead to dump trucks failing to accurately stop at the optimal unloading position, thus affecting operational efficiency and safety. Summary of the Invention

[0003] The main objective of this application is to provide a method, device, vehicle, and storage medium for controlling the parking position of a vehicle during dumping, in order to solve the problem that existing fixed-coordinate navigation vehicles are unable to accurately park at the actual dumping position during dumping, thus affecting the dumping operation.

[0004] In a first aspect, embodiments of this application provide a method for controlling the parking position of a vehicle at its dumping site, wherein the vehicle is equipped with data acquisition units distributed at the dumping port of the cargo compartment and on the rear wheels; the method for controlling the parking position at its dumping site includes:

[0005] Receive dumping site planning data from the mine's vehicle scheduling system, decode and verify the dumping site planning data, and obtain target dumping site information that matches the vehicle's own driving.

[0006] Based on the vehicle's current position and orientation information and the target dumping location information, a path planning algorithm is invoked to generate a dumping operation trajectory, and navigation is performed based on the dumping operation trajectory to complete the dumping operation.

[0007] When the vehicle drives to the target dumping site and reverses, the data acquisition unit collects dumping site information and vehicle operation data behind the vehicle in real time.

[0008] The dumping position information and the vehicle operation data are input into a pre-built endpoint discrimination model. The endpoint discrimination model identifies whether the vehicle has reached the dumping endpoint. If it has, the vehicle braking and parking control are triggered, and the lifting device is controlled to complete the unloading operation.

[0009] In one feasible implementation, receiving dumping site planning data from the mine's vehicle scheduling system, and decoding the dumping site planning data to obtain target dumping site information matching the vehicle's own driving conditions includes:

[0010] Establish a TLS-encrypted communication link and use the communication link as a medium to perform two-way authentication between the vehicle-side certificate and the dispatcher-side certificate;

[0011] After completing two-way authentication, the system receives messages of dumping site planning data transmitted by the vehicle scheduling system using the MQTT protocol and encapsulated in Protobuf format based on the communication link.

[0012] The fields in the message are parsed and matched with the vehicle's own pose. Based on the matching result, the data is encapsulated according to a preset data structure to obtain the coordinates, orientation angle, number, and priority of the dumping position.

[0013] In one feasible implementation, the step of generating a dumping operation trajectory by invoking a path planning algorithm based on the vehicle's current position and attitude information and the target dumping location information, and then navigating based on the dumping operation trajectory to complete the dumping operation, includes:

[0014] The current pose information of the vehicle is obtained by using the RTK-GPS and IMU fusion positioning method. The current pose information includes the vehicle's current position coordinates, driving direction and the work route number.

[0015] Using the current pose information and the coordinates of the dumping site as constraints, the improved RRT* algorithm is invoked to generate a collision-free dumping operation trajectory in a 3D mesh map. The dumping operation trajectory includes the transportation operation route and the dumping site reversing trajectory.

[0016] The vehicle's navigation planning algorithm is invoked to drive sequentially based on the transportation operation route and the reversing trajectory of the dumping site, thereby completing the transportation and dumping operations.

[0017] In one feasible implementation, after generating a collision-free dumping operation trajectory in a 3D mesh map by calling the improved RRT* algorithm with the current pose information and the coordinates of the dumping location as constraints, the method further includes:

[0018] The reversing trajectory of the dumping site is smoothed multiple times using B-spline curve smoothing, and the smoothed reversing trajectory is constrained and corrected based on the maximum curvature limit of the vehicle.

[0019] In one feasible implementation, the invocation of the vehicle's navigation planning algorithm, which sequentially drives based on the transportation operation route and the reversing trajectory at the dumping site to complete the transportation and dumping operations, includes:

[0020] Based on the corrected reversing trajectory of the dumping position, the vehicle dynamics model is used to calculate the steering angle, acceleration and braking force control quantities of the vehicle when reversing in the dumping position.

[0021] Based on the navigation guidance of the transportation operation route, the vehicle moves towards the target dumping site, and after moving to a preset distance from the target dumping site, the vehicle reverses to perform the dumping operation based on the steering angle, the acceleration, and the braking force control amount.

[0022] In one feasible implementation, the step of obtaining the vehicle's current pose information using the RTK-GPS and IMU fusion positioning method includes:

[0023] The vehicle's heading angle is coarsely solved by receiving dual-antenna RTK-GPS carrier phase observations and using a differential de-fuzzy algorithm to obtain the baseline vector.

[0024] The vehicle's IMU is integrated within a preset time window to obtain the displacement increment, velocity increment, and attitude increment.

[0025] The coarse solution of the heading angle and the incremental data are input into an extended Kalman filter to update the 15-dimensional state vector, thereby obtaining the vehicle's current pose information.

[0026] In one feasible implementation, the real-time acquisition of soil dumping position information and vehicle operation data behind the vehicle through the acquisition unit includes:

[0027] Images of the dumping area behind the vehicle are captured by the image unit on the dumping port of the truck, and information on the cliff edge of the retaining wall and the excavated soil is identified.

[0028] Read the suspension displacement sensor signal, combine it with the vehicle load information to perform threshold compensation, and obtain the suspension compression amount;

[0029] Obtain the torque curve output by the wheel-side drive system, denoise the torque curve, and calculate the torque increment using a prediction model;

[0030] The triaxial acceleration values ​​of the rear axle were collected and analyzed using a sliding window difference method to obtain the acceleration abrupt change.

[0031] The ratio of the front and rear wheel speed difference to the vehicle speed is calculated to obtain the wheel slip ratio.

[0032] In one feasible implementation, the step of inputting the dumping position information and the vehicle operation data into a pre-constructed endpoint discrimination model, and using the endpoint discrimination model to identify whether the vehicle has reached the dumping endpoint, if so, triggering vehicle braking and parking control, and controlling the lifting device to complete the unloading operation, includes:

[0033] The cliff edge information, suspension compression, torque increment, acceleration mutation, wheel slip ratio, and vehicle speed are concatenated into a six-dimensional feature vector and then batch normalized.

[0034] The normalized feature vector is input into the endpoint discrimination model to calculate the collision probability that the vehicle has touched the dump wall;

[0035] Based on the collision probability, it is determined whether the soil discharge endpoint has been reached. If it has, the vehicle braking and parking control are triggered, and the lifting device is controlled to complete the unloading operation.

[0036] Secondly, embodiments of this application provide a vehicle dumping position parking control device, wherein the vehicle is equipped with data acquisition units distributed at the dumping port of the cargo compartment and on the rear wheels; the dumping position parking control device includes:

[0037] The receiving module is used to receive the dumping site planning data issued by the vehicle overall scheduling system in the mine, decode the dumping site planning data and verify the permissions to obtain the target dumping site information that matches the vehicle's own driving.

[0038] The planning module is used to generate a dumping operation trajectory by calling a path planning algorithm based on the vehicle's current position and the target dumping location information, and to complete the dumping operation by navigating based on the dumping operation trajectory.

[0039] The data acquisition module is used to collect information on the dumping site behind the vehicle and the vehicle's operation data in real time when the vehicle is driving to the target dumping site and reversing.

[0040] The docking module is used to input the dumping position information and the vehicle operation data into a pre-built endpoint discrimination model. The endpoint discrimination model identifies whether the vehicle has reached the dumping endpoint. If it has, the vehicle braking and parking control are triggered, and the lifting device is controlled to complete the unloading operation.

[0041] Thirdly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions cause the processor to implement the above-described vehicle dumping and parking control method.

[0042] The embodiments of this application bring the following beneficial effects:

[0043] The aforementioned vehicle dumping position parking control method, device, vehicle, and storage medium obtain matching target dumping position information by receiving and verifying dumping position planning data, and combine path planning navigation to ensure the accuracy of the driving trajectory; when reversing, the acquisition unit collects dumping position information and vehicle operation data in real time, and accurately identifies the dumping endpoint through the endpoint discrimination model and triggers braking and unloading, effectively solving the problem that it is difficult for fixed coordinate navigation vehicles to accurately park at the real dumping position, and improving the accuracy and efficiency of dumping operations.

[0044] Other features and advantages of this application will be set forth in the following description and will be apparent in part from the description or may be learned by practicing the application. The objectives and other advantages of this application are realized and obtained through the structures particularly pointed out in the description, claims and drawings.

[0045] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0046] To more clearly illustrate the specific embodiments of this application or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0047] Figure 1 A flowchart illustrating an embodiment of the vehicle dumping position parking control method provided in this application;

[0048] Figure 2 Flowchart of another embodiment of the vehicle dumping position parking control method provided in this application;

[0049] Figure 3 A schematic diagram of a vehicle dumping position control device provided in an embodiment of this application;

[0050] Figure 4 This is a schematic diagram of a vehicle provided in an embodiment of this application. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0052] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms “comprising” or “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0053] This application provides a control method for precise parking of vehicles in a spoil heap. In scenarios where the spoil heap wall may change dynamically, the method can guide the vehicle to reverse based on the initially planned spoil heap position and dynamically determine whether to continue reversing based on the changes in the vehicle's state as it approaches the spoil heap wall, thereby achieving higher precision parking control.

[0054] For ease of understanding, the specific process of the embodiments of this application is described below. The embodiments of this application are applied to in-vehicle terminals or vehicle autonomous driving systems. Please refer to [link / reference]. Figure 1 One embodiment of the vehicle dumping position parking control method in this application specifically includes the following steps:

[0055] S110: Receive the dumping site planning data issued by the mine's vehicle scheduling system, decode and verify the dumping site planning data, and obtain the target dumping site information that matches the vehicle's own driving.

[0056] It should be noted that multiple vehicles operate simultaneously throughout the mine. To avoid collisions between vehicles during soil dumping operations, the information of all vehicles required for operation is input into a vehicle scheduling system. Based on the map information of the soil dumping area, loading area, and the area between them, as well as vehicle information rules, the system schedules and plans all vehicles, obtaining the vehicle scheduling planning information and the soil dumping position information corresponding to each vehicle in the soil dumping area.

[0057] Of course, the information on the corresponding dumping position of a vehicle in the dumping area can be a fixed setting or a real-time dynamic setting. That is to say, when planning vehicle scheduling, the transportation information of the vehicle can be matched with the required earthwork information of each dumping position in the dumping area to make a binding scheduling setting. After the binding setting is made, the route is selected and planned based on the map information; or, after the vehicle is a certain distance away from the dumping area, it receives the dumping position planning data from the vehicle overall scheduling system in real time, matches and selects the target dumping position information based on the occupancy status of the dumping position and the traffic conditions of the surrounding dumping positions, and notifies the corresponding vehicle to perform reversing operations.

[0058] In this embodiment, after receiving the spoil disposal site planning data, the system parses the data using a communication protocol to extract relevant information about all spoil disposal sites within the spoil disposal area, such as the occupancy status, location information, and the location of retaining walls within the spoil disposal sites. The permission verification here should be understood as the vehicle scheduling system defining the operational location range of the spoil disposal area for the vehicle. Based on the relationship between the operational location range and the location information and occupancy status of the spoil disposal sites, it determines whether the vehicle is permitted to enter a spoil disposal site. Based on the verification result, at least one spoil disposal site is selected as the target spoil disposal site for that vehicle.

[0059] S120. Based on the vehicle's current position and orientation information and the target dumping location information, call the path planning algorithm to generate the dumping operation trajectory, and use the dumping operation trajectory for navigation to complete the dumping operation.

[0060] It should be noted that this path planning algorithm refers to a calculation method that generates a feasible path based on the spatial relationship between the vehicle's current position and the target position, such as an improved random sampling algorithm. Its purpose is to avoid the limitations of fixed coordinate navigation and adapt to dynamic path adjustments in complex terrain.

[0061] The vehicle obtains its current position and pose information through positioning methods such as GPS / RTK and inertial navigation. Then, it combines the target dumping site information to plan the path and generate a driving trajectory for the vehicle to reverse to the target dumping site. The vehicle then performs automatic reversing operation along the planned trajectory, gradually approaching the target dumping site until it reaches the retaining wall of the target dumping site.

[0062] S130. When the vehicle travels to the target dumping position and reverses, the dumping position information and vehicle operation data behind the vehicle are collected in real time through the data acquisition unit.

[0063] In this embodiment, a data acquisition unit is installed at the rear of the vehicle (i.e., on the side of the rear of the vehicle in the cargo compartment). This data acquisition unit can be a camera, radar, etc. When the vehicle reverses and is positioned within the target dumping area, the data acquisition unit is triggered to work and capture image data of the rear, which can be pictures, videos, etc. The image data is mainly image data of the soil accumulation in the dumping area behind the vehicle.

[0064] Specifically, the acquisition unit can also be an image acquisition device and a suspension displacement sensor, used to simultaneously acquire data on the edge morphology of the earth wall and the suspension status of the vehicle.

[0065] Furthermore, sensors are installed at the rear wheels of the vehicle to collect its operational data, such as whether the vehicle is blocked by a retaining wall, and data from the vehicle's own speed sensors, including its speed. Based on this data, it can be determined whether the vehicle has reached the planned dumping position. Once reached, the vehicle continues to reverse based on the dumping position information until the operational data generated during reversing reaches a threshold, confirming that the vehicle has reached the true target dumping position. This reversing control method avoids errors between planning and actual conditions, preventing inaccurate dumping and obstruction of the route.

[0066] S140. Input the dumping position information and vehicle operation data into the pre-built endpoint discrimination model. The endpoint discrimination model identifies whether the vehicle has reached the dumping endpoint. If it has, the vehicle braking and parking control are triggered, and the lifting device is controlled to complete the unloading operation.

[0067] In this embodiment, the endpoint discrimination model is actually trained using a neural network, which can achieve cliff recognition and state recognition of the arrival at the retaining wall. For example, a multi-dimensional feature classifier trained by machine learning algorithms is used to comprehensively determine whether the vehicle has reached the optimal unloading position, which can be implemented by using a six-dimensional feature vector input and probability calculation.

[0068] Here, the dumping position information is input into the endpoint discrimination model. The model identifies whether the cliff position has been reached based on the dumping position information input in multiple consecutive frames. It can be understood that the endpoint discrimination model identifies whether there is a cliff in each frame of dumping position information by a combination of neighborhood feature comparison and flat ground recognition, and marks it. The proportion of cliff features in the dumping position information of multiple consecutive frames is statistically analyzed to determine whether the vehicle has reached the edge of the cliff.

[0069] Furthermore, the vehicle's operating data specifically includes suspension displacement, wheel-side torque, rear axle acceleration, and front and rear wheel speed signals. After these data are input into the endpoint discrimination model, the model combines the corresponding endpoint constraints to determine whether the position of the retaining wall has been reached.

[0070] If the system determines that it has reached the edge of a cliff and the location of a retaining wall, it will trigger the braking and parking controls and control the lifting device to complete the unloading operation.

[0071] In practical applications, the priority for retaining walls along cliff edges is set to the highest priority. If either condition is met upon reaching the cliff edge, braking and parking controls are triggered, and the lifting device is controlled to complete the unloading operation. Otherwise, the vehicle continues reversing until the cliff edge condition is met.

[0072] Overall, the vehicle first receives dumping location information from the dispatch system via an encrypted communication link. After decoding and verification, the target location coordinates and priority parameters are extracted. Subsequently, based on real-time positioning data, a path planning algorithm is invoked to generate a work route including the transportation path and reversing trajectory. The navigation control module guides the vehicle to the target area. When the vehicle enters the reversing phase, sensors installed at the dumping port and rear wheels simultaneously collect data such as images of the earthen wall edge, suspension compression, and wheel speed difference. After normalization processing, this data is input into the endpoint discrimination model for probability calculation. When the model output probability exceeds a preset threshold, the braking system is immediately triggered, and the lifting device is controlled to perform the unloading operation, ensuring that the vehicle completes the operation at the optimal position.

[0073] By implementing the above technical solution, the dynamic information of the dumping area is collected in real time and combined with vehicle operation data. The endpoint discrimination model is used to accurately determine the unloading endpoint position and trigger braking and unloading control. This solves the problem of docking deviation caused by the dynamic changes of the dumping wall in traditional methods. It has the advantages of improving docking accuracy and operation efficiency and reducing safety hazards.

[0074] like Figure 2 The image shows another embodiment of the vehicle dumping position parking control method provided in this application. This embodiment takes into account the dynamic nature of terrain changes during dumping operations, requiring the integration of multi-source data fusion and intelligent discrimination models to analyze the vehicle's posture and surrounding environment information in real time during reversing, thereby achieving precise parking control. The method specifically includes the following steps:

[0075] S210: Receive the dumping site planning data issued by the mine's vehicle scheduling system, decode and verify the dumping site planning data, and obtain the target dumping site information that matches the vehicle's own driving.

[0076] It should be noted that the dumping site planning data is obtained by the vehicle scheduling system based on the dynamic allocation of mining area operations and the dynamic detection of the position of each vehicle in the dumping area to plan the unloading location information.

[0077] Specifically, the received dumping site planning data can be transmitted and decoded using a protocol-encrypted method. This ensures both the data's compatibility with the vehicle and that the vehicle receives a legitimate and valid target location instruction.

[0078] In this embodiment, a TLS-encrypted communication link is established, and the communication link is used as a medium to perform two-way authentication between the vehicle-side certificate and the dispatcher certificate. After the two-way authentication is completed, the message of the dumping site planning data transmitted by the vehicle overall dispatching system using the MQTT protocol and encapsulated in Protobuf format is received based on the communication link. The field content in the message is parsed, and the parsed field content is matched with the vehicle's own pose. Based on the matching result, the data is encapsulated according to a preset data structure to obtain the coordinates, orientation angle, number, and priority of the dumping site.

[0079] It's understandable that Protobuf format encapsulation refers to using a protocol buffer binary serialization method, which can be specifically achieved by defining a .proto file to generate a codec. The default data structure can be implemented using JSON or XML format.

[0080] Specifically, when a vehicle receives dumping site planning data, an encrypted channel is first established via the TLS protocol. The vehicle and dispatching system exchange digital certificates to complete two-way identity verification, ensuring the security of the communication link. The vehicle scheduling system encodes the dumping site planning data in Protobuf format and pushes it to the vehicle via the MQTT protocol. After receiving the message, the vehicle extracts the field content using a Protobuf decoder, performs spatial matching between the dumping site coordinates and the vehicle's real-time positioning information, and filters out the target dumping sites associated with the current work route. After matching, the dumping site number, coordinates, direction angle, and work priority are encapsulated according to a preset structure to form structured data that can be directly called by the path planning module.

[0081] This system employs two-way certificate authentication and TLS encryption to build a secure channel, effectively preventing man-in-the-middle attacks and data tampering. Through Protobuf binary encoding and structured encapsulation, it improves transmission efficiency while ensuring data integrity, avoiding scheduling errors caused by format corruption.

[0082] S220. The current position and pose information of the vehicle is obtained by using the RTK-GPS and IMU fusion positioning method. The current position and pose information includes the current position coordinates of the vehicle, the driving direction and the operation route number.

[0083] The method of obtaining the vehicle's current pose information by using RTK-GPS and IMU fusion positioning specifically involves fusing real-time dynamic differential positioning technology with data from the inertial measurement unit to obtain the vehicle's current pose information. For example, by combining dual-antenna carrier phase observations with IMU three-axis specific force integrals, centimeter-level positioning accuracy and heading angle are fused and output to obtain the vehicle's current pose information.

[0084] Specifically, by receiving dual-antenna RTK-GPS carrier phase observations, a differential de-fuzzy algorithm is used to obtain the baseline vector, resulting in a coarse solution for the vehicle's heading angle. The vehicle's IMU is integrated within a preset time window to obtain the displacement increment, velocity increment, and attitude increment. The coarse solution for the heading angle and the incremental data are input into an extended Kalman filter to update the 15-dimensional state vector, thus obtaining the vehicle's current pose information.

[0085] Understandably, a dual-antenna RTK-GPS system generates carrier phase observations by receiving satellite signals. After eliminating common errors through double-difference processing, it obtains a preliminary estimate of the heading angle by solving the baseline vector. Simultaneously, the IMU continuously collects the vehicle's three-axis acceleration and angular velocity data. After integration within a time window, it outputs the vehicle's motion increment over a short period. These two data sources are input into an extended Kalman filter, which dynamically adjusts the state vector, including position, velocity, attitude, and sensor error parameters, through prediction and update steps, ultimately outputting fused, high-precision pose information. For example, in the complex terrain of an open-pit mine, when GPS signals fluctuate due to obstruction, the IMU's integrated data can temporarily maintain positioning continuity, while the Kalman filter dynamically adjusts the weights of both data through the noise covariance matrix to ensure the stability of the positioning results.

[0086] In practical applications, the driving direction includes heading angle, velocity, and acceleration; obtaining the vehicle's current pose information is specifically implemented in three steps:

[0087] Step 1, RTK-GPS positioning;

[0088] A dual-antenna RTK system is used to receive carrier phase observations between the base station and the mobile station. Integer ambiguities are then resolved using a differential deambiguity algorithm to obtain the vehicle's current coordinates in three-dimensional space, specifically including lateral, longitudinal, and altitude coordinates. Simultaneously, the vehicle's initial heading angle is calculated based on the direction vector between the two antennas. This stage also requires the synchronous recording of timestamps and error weighting parameters from the GNSS observation data to support subsequent time alignment and accuracy weighting with IMU data.

[0089] Step 2, IMU dynamic sampling;

[0090] Within a preset time window, the inertial measurement unit (IMU) continuously acquires the vehicle's three-axis acceleration and three-axis angular velocity data, and integrates these data to obtain the vehicle's velocity increment and attitude change increment during that time period. The linear acceleration is used to estimate the velocity change, while the angular velocity is used to estimate the changes in the vehicle's roll, pitch, and yaw angles.

[0091] Step 3, Extended Kalman Filter (EKF) fusion;

[0092] The fusion here actually involves constructing a 15-dimensional state vector, including position, velocity, attitude, and sensor readings, as the system state estimation object. This state vector includes: position coordinates (lateral, longitudinal, and altitude), velocity components (three directions), attitude angles (roll, pitch, and yaw), and three-axis acceleration and angular velocity readings.

[0093] The inertial motion model of the vehicle is used to update the state vector over time; the position coordinates and heading angle obtained by RTK-GPS are used as observations and input into the Kalman filter for state update; the speed and attitude changes estimated by IMU are combined to minimize the error and fuse them to output high-precision current vehicle pose information, including the current lateral, longitudinal, and altitude positions and yaw angle.

[0094] Step 4: Identify the work route number.

[0095] The fused horizontal and vertical coordinates of the vehicle are projected onto the center lines of each work path on the digital operation map of the mine. For each path, the minimum Euclidean distance from the current vehicle's location to all points on the path's center line is calculated, and the path number with the smallest distance below a preset threshold is used as the work route number of the current vehicle. This path number will serve as an important input parameter for subsequent path planning algorithms to generate targeted soil disposal trajectories.

[0096] S230. Using the current pose information and the coordinates of the dumping site as constraints, call the improved RRT* algorithm to generate a collision-free dumping operation trajectory in the three-dimensional mesh map. The dumping operation trajectory includes the transportation operation route and the dumping site reversing trajectory.

[0097] The improved RRT* algorithm is specifically an algorithm obtained by introducing a dynamic step size adjustment mechanism and obstacle expansion layer constraints on the basis of the fast expanding random tree algorithm. For example, it can automatically optimize the path node generation strategy according to the terrain undulation and obstacle distribution in a 3D grid map to avoid collisions between vehicles and slopes or material piles.

[0098] S240: Call the vehicle's navigation planning algorithm and drive sequentially based on the transportation operation route and the reversing trajectory of the dumping site to complete the transportation and dumping operations;

[0099] In this embodiment, after generating a collision-free dumping operation trajectory in a three-dimensional mesh map by calling the improved RRT* algorithm with the current pose information and the coordinates of the dumping position as constraints, the method further includes: performing multiple smoothing processes on the dumping position reversing trajectory using B-spline curve smoothing, and constraining and correcting the smoothed dumping position reversing trajectory based on the vehicle's maximum curvature limit.

[0100] The vehicle receives carrier phase observations via dual-antenna RTK-GPS and combines them with acceleration and angular velocity data collected by the IMU within a preset time window. After fusion processing using an extended Kalman filter, the current position coordinates and heading angle are output. Based on this pose information, the path planning algorithm uses the soil dumping site coordinates as the target point and calls an improved RRT* algorithm on a 3D grid map to generate a transportation operation route. This improved RRT* algorithm dynamically adjusts the direction of random tree expansion to avoid obstacle areas marked on the map, thus obtaining the transportation operation route.

[0101] For planning the reversing trajectory at the dumping site, a sequence of reversing path points with the center of the vehicle's rear axle as the reference is generated at the dumping site coordinates, and the path abrupt changes are eliminated by smoothing with B-spline curves to obtain the dumping site reversing trajectory.

[0102] When the navigation completes the transportation and dumping operations, it guides the vehicle to the vicinity of the target dumping site based on the generated transportation route, and then switches to the reversing trajectory control mode to drive the vehicle to complete the precise parking along the planned path.

[0103] Furthermore, after generating the initial reversing trajectory using the improved RRT* algorithm, the generated path may contain broken lines or abrupt turning points. Therefore, the trajectory needs to be iteratively optimized multiple times using B-spline curves. For example, the first smoothing process can eliminate sawtooth fluctuations in the path, and the second smoothing process can adjust the control points to make the curvature change continuous. After smoothing, the curvature values ​​of each point on the trajectory need to be traversed. If the curvature of a certain segment of the trajectory is found to exceed the maximum value allowed by the vehicle steering system, a new control point is inserted in that segment of the trajectory, and a corrected trajectory that conforms to the curvature constraint is generated by recalculating the B-spline basis function coefficients.

[0104] In another embodiment, the navigation planning algorithm of the vehicle is invoked to drive sequentially based on the transportation operation route and the reversing trajectory of the dumping site to complete the transportation and dumping operations. This includes: inputting the corrected reversing trajectory of the dumping site into the vehicle dynamics model to calculate the steering angle, acceleration, and braking force control quantities of the vehicle when reversing at the dumping site; guiding the vehicle to move towards the target dumping site based on the transportation operation route navigation, and after moving to a preset distance from the target dumping site, controlling the vehicle to reverse and perform the dumping operation based on the steering angle, acceleration, and braking force control quantities.

[0105] It should be noted that the vehicle dynamics model refers to a mathematical model used to simulate the mechanical characteristics of a vehicle during motion. Specifically, it can be implemented using multibody dynamics modeling methods combined with a tire-ground contact model to predict the dynamic response of a vehicle under different trajectories.

[0106] Specifically, after generating the smoothed reversing trajectory for the dumping site, the trajectory data is input into the vehicle dynamics model for kinematic simulation. The optimal combination of steering angle, acceleration, and braking force parameters is obtained by solving the vehicle's motion differential equations. The vehicle is guided towards the target dumping site based on the transportation route. When the distance between the vehicle and the target dumping site reaches a preset threshold, a reversing control mode switch is triggered. At this time, the vehicle, based on the steering angle, acceleration, and braking force control values ​​output by the dynamics model, executes the reversing operation through the coordinated action of the vehicle's steer-by-wire system and electric braking system, allowing the vehicle to precisely stop at the dumping endpoint along the corrected reversing trajectory.

[0107] S250: When the vehicle travels to the target dumping position and reverses, the data acquisition unit collects the dumping position information and vehicle operation data behind the vehicle in real time.

[0108] In this step, images of the dumping area behind the vehicle are acquired through the image unit on the dumping port of the truck bed, and the cliff edge information of the retaining wall and the excavated soil is identified; the suspension displacement sensor signal is read, and threshold compensation is performed in combination with the vehicle load information to obtain the suspension compression; the torque curve output by the wheel-side drive system is acquired, the torque curve is denoised, and the torque increment is calculated through a prediction model; the three-axis acceleration values ​​of the rear axle are acquired and analyzed using a sliding window differential method to obtain the acceleration mutation; the ratio of the front and rear wheel speed difference to the vehicle speed is calculated to obtain the wheel slip ratio.

[0109] Understandably, the process of reading the suspension displacement sensor signal, combining it with vehicle load information for threshold compensation to obtain the suspension compression amount involves estimating the real-time load mass m using a vehicle weighing sensor, and calculating the suspension compression threshold according to a preset linear relationship; performing a first-order low-pass filter on the original suspension displacement sequence to obtain a smooth displacement curve; performing time difference analysis on the smooth displacement curve to calculate the compression rate; comparing the compression rate with the threshold, and setting a suspension anomaly flag if the threshold is exceeded.

[0110] The process involves acquiring the torque curve output by the wheel-side drive system, denoising the torque curve, and calculating the torque increment using a prediction model. Specifically, this is achieved by performing Savitzky-Golay filtering on the raw wheel-side torque data to remove high-frequency noise; performing Z-score normalization on the filtered torque curve; inputting the normalized torque curve and historical torque sequences into an LSTM neural network prediction model to obtain a dynamic threshold; and comparing the current torque increment with the threshold, setting a torque anomaly flag if the threshold is exceeded.

[0111] S260. Input the soil dumping position information and the vehicle operation data into the pre-built endpoint discrimination model. The endpoint discrimination model identifies whether the vehicle has reached the soil dumping endpoint. If it has, the vehicle braking and parking control are triggered, and the lifting device is controlled to complete the unloading operation.

[0112] In this step, the cliff edge information, suspension compression, torque increment, acceleration mutation, wheel slip ratio, and vehicle speed are concatenated into a six-dimensional feature vector and then batch normalized. The normalized feature vector is input into the endpoint discrimination model to calculate the collision probability that the vehicle has touched the dump wall. Based on the collision probability, it is determined whether the dumping endpoint has been reached. If it has, the vehicle braking and parking control are triggered, and the lifting device is controlled to complete the unloading operation.

[0113] In other words, during the reversing process, the acquisition unit continuously collects images of the dumping area and extracts the coordinates of the cliff edge of the retaining wall using computer vision algorithms. Suspension displacement sensors monitor the extension and retraction of the suspension hydraulic cylinders in real time, and dynamically compensate the measured values ​​based on the current load data to obtain accurate suspension compression. The torque signal from the wheel-side drive system, after being low-pass filtered to remove high-frequency noise, is input into a long short-term memory network to predict the torque increment within the next three seconds. Rear axle acceleration data is used to calculate the difference value through a sliding window to capture sudden acceleration changes caused by the vehicle contacting the earth wall. The ratio of the front and rear wheel speed difference to the current vehicle speed is calculated in real time to obtain the wheel slip ratio. These six types of data are standardized and concatenated into a feature vector, which is then input into a trained endpoint discrimination model. When the model outputs a probability value exceeding a preset threshold, it determines that the vehicle has reached the dumping endpoint, triggering the braking system to decelerate and park, while simultaneously activating the lifting mechanism to unload the material.

[0114] Specifically, by comparing the collision probability P with the preset threshold P0, if P≥P0, it is determined that the end point of the soil discharge has been reached and a stop trigger sign is generated; during the effective period of the stop trigger sign, hydraulic braking, parking and lifting control commands are issued and the unloading completion status is monitored.

[0115] Understandably, by comparing the collision probability P with a preset threshold P0, if P ≥ P0, it is determined that the end point of the dumping has been reached, and a parking trigger sign is generated, including:

[0116] The system performs a logical AND operation on the parking trigger flag and the parking conditions calculated based on the rule engine to form the final parking judgment value and preset the threshold P0 in a timely manner; it determines whether the real-time speed of the vehicle is less than or equal to 0.5 km / h; it determines whether the millimeter-wave radar ranging d is less than the set minimum safe distance; and it generates a parking execution command only when the final parking judgment value, speed condition and distance condition are all satisfied.

[0117] During the valid period of the stop trigger sign, hydraulic braking, parking, and lifting control commands are issued, and the unloading completion status is monitored, including:

[0118] Send a braking command to the vehicle's hydraulic braking system to decelerate the vehicle to zero; after the vehicle stops, control the parking brake to keep the vehicle stationary; while in the parking state, control the lifting valve to open to a preset angle to tilt the cargo box to complete unloading; after unloading is completed, close the lifting valve and record the unloading timestamp, and update the operation status.

[0119] In another embodiment, the method further includes:

[0120] The ToF depth camera at the rear of the vehicle was used to collect RGB-D point cloud data of the spoil heap.

[0121] Perform a semantic segmentation algorithm on the point cloud data and extract the point cloud from the surface of the spoil wall;

[0122] Perform point-to-plane iterative nearest-point registration between the point cloud on the surface of the spoil wall and the vehicle coordinate system;

[0123] The registered distance features of the spoil heap wall are fused with spoil heap location information and vehicle operation data to form a feature vector, which is used as the input for determining the spoil heap endpoint.

[0124] In this embodiment, after the complete unloading operation, the following is also included:

[0125] Record the vehicle's final parking position after it stops and encapsulate it as unloading result data;

[0126] The unloading result data is uploaded to the vehicle scheduling system via gRPC streaming communication.

[0127] The unloading result data is used to update the three-dimensional voxel model of the spoil wall constructed based on Bayesian theory;

[0128] After the three-dimensional voxel model of the spoil wall is updated and converged, the vehicle scheduling system generates the next spoil position and sends it to the vehicles, forming a closed-loop optimization.

[0129] In this embodiment, after determining that the retaining wall has been reached, the method further includes:

[0130] Based on the results of identifying the cliff edge, the vehicle dynamics constraints are linearized and the optimal control quantity is solved;

[0131] The optimal control quantity is encapsulated into a CAN bus control message and sent to the steering, drive, and braking actuators.

[0132] The actuator feedback is monitored in real time and the trajectory deviation is calculated. When the deviation exceeds the threshold, the control quantity is re-optimized.

[0133] In this embodiment, the process of controlling the vehicle based on the reversing trajectory of the dumping site also includes: local path replanning, specifically:

[0134] The distance to the spoil wall is measured in real time using a laser scanner at the rear of the vehicle, and the difference between the measured and predicted distances is calculated.

[0135] When the absolute value of the distance difference is greater than 1m, a local path replanning request is triggered;

[0136] Starting from the current vehicle position and ending at the dumping location, the D-Hybrid A* algorithm is run within a 10m local area to generate an updated path.

[0137] The updated path is smoothly spliced ​​with the original reverse trajectory of the dumping site and then output to the controller for vehicle control.

[0138] In summary, this embodiment, by collecting dynamic information of the dumping area in real time and combining it with vehicle operation data, uses the endpoint discrimination model to accurately determine the unloading endpoint position and trigger braking and unloading control. This solves the docking deviation problem caused by the dynamic changes of the dumping wall in traditional methods, and has the advantages of improving docking accuracy and operation efficiency, and reducing safety hazards.

[0139] For the corresponding method embodiments described above, see [link to relevant documentation]. Figure 3 The diagram shows a schematic of a dump truck parking control device for a vehicle. This device is used in dump trucks or other transport vehicles in mining areas. The automated driving vehicle is equipped with data collection units distributed at the dump truck's dumping port and on the rear wheels. The parking control device includes:

[0140] The receiving module 310 is used to receive the dumping site planning data issued by the vehicle overall scheduling system in the mine, decode the dumping site planning data and verify the permissions to obtain the target dumping site information that matches the vehicle's own driving.

[0141] The planning module 320 is used to generate a soil dumping operation trajectory by calling a path planning algorithm based on the vehicle's current position attitude information and the target soil dumping location information, and to complete the soil dumping operation by navigating based on the soil dumping operation trajectory.

[0142] The acquisition module 330 is used to collect information on the dumping site behind the vehicle and vehicle operation data in real time when the vehicle drives to the target dumping site and reverses.

[0143] The docking module 340 is used to input the dumping position information and the vehicle operation data into a pre-built endpoint discrimination model. The endpoint discrimination model identifies whether the vehicle has reached the dumping endpoint. If it has, the vehicle braking and parking control are triggered, and the lifting device is controlled to complete the unloading operation.

[0144] In this embodiment, the receiving module 310 is specifically used for:

[0145] Establish a TLS-encrypted communication link and use the communication link as a medium to perform two-way authentication between the vehicle-side certificate and the dispatcher-side certificate;

[0146] After completing two-way authentication, the system receives messages of dumping site planning data transmitted by the vehicle scheduling system using the MQTT protocol and encapsulated in Protobuf format based on the communication link.

[0147] The fields in the message are parsed and matched with the vehicle's own pose. Based on the matching result, the data is encapsulated according to a preset data structure to obtain the coordinates, orientation angle, number, and priority of the dumping position.

[0148] In this embodiment, the planning module 320 is specifically used for:

[0149] The current pose information of the vehicle is obtained by using the RTK-GPS and IMU fusion positioning method. The current pose information includes the vehicle's current position coordinates, driving direction and the work route number.

[0150] Using the current pose information and the coordinates of the dumping site as constraints, the improved RRT* algorithm is invoked to generate a collision-free dumping operation trajectory in a 3D mesh map. The dumping operation trajectory includes the transportation operation route and the dumping site reversing trajectory.

[0151] The vehicle's navigation planning algorithm is invoked to drive sequentially based on the transportation operation route and the reversing trajectory of the dumping site, thereby completing the transportation and dumping operations.

[0152] In this embodiment, the planning module 320 is further used for:

[0153] The reversing trajectory of the dumping site is smoothed multiple times using B-spline curve smoothing, and the smoothed reversing trajectory is constrained and corrected based on the maximum curvature limit of the vehicle.

[0154] In this embodiment, the planning module 320 is specifically used for:

[0155] Based on the corrected reversing trajectory of the dumping position, the vehicle dynamics model is used to calculate the steering angle, acceleration and braking force control quantities of the vehicle when reversing in the dumping position.

[0156] Based on the navigation guidance of the transportation operation route, the vehicle moves towards the target dumping site, and after moving to a preset distance from the target dumping site, the vehicle reverses to perform the dumping operation based on the steering angle, the acceleration, and the braking force control amount.

[0157] In this embodiment, the planning module 320 is specifically used for:

[0158] The vehicle's heading angle is coarsely solved by receiving dual-antenna RTK-GPS carrier phase observations and using a differential de-fuzzy algorithm to obtain the baseline vector.

[0159] The vehicle's IMU is integrated within a preset time window to obtain the displacement increment, velocity increment, and attitude increment.

[0160] The coarse solution of the heading angle and the incremental data are input into an extended Kalman filter to update the 15-dimensional state vector, thereby obtaining the vehicle's current pose information.

[0161] In this embodiment, the acquisition module 330 is specifically used for:

[0162] Images of the dumping area behind the vehicle are captured by the image unit on the dumping port of the truck, and information on the cliff edge of the retaining wall and the excavated soil is identified.

[0163] Read the suspension displacement sensor signal, combine it with the vehicle load information to perform threshold compensation, and obtain the suspension compression amount;

[0164] Obtain the torque curve output by the wheel-side drive system, denoise the torque curve, and calculate the torque increment using a prediction model;

[0165] The triaxial acceleration values ​​of the rear axle were collected and analyzed using a sliding window difference method to obtain the acceleration abrupt change.

[0166] The ratio of the front and rear wheel speed difference to the vehicle speed is calculated to obtain the wheel slip ratio.

[0167] In this embodiment, the exam suspension module 340 is specifically used for:

[0168] The cliff edge information, suspension compression, torque increment, acceleration mutation, wheel slip ratio, and vehicle speed are concatenated into a six-dimensional feature vector and then batch normalized.

[0169] The normalized feature vector is input into the endpoint discrimination model to calculate the collision probability that the vehicle has touched the dump wall;

[0170] Based on the collision probability, it is determined whether the soil discharge endpoint has been reached. If it has, the vehicle braking and parking control are triggered, and the lifting device is controlled to complete the unloading operation.

[0171] In summary, by integrating dynamic path planning and multi-source data discrimination mechanisms, the system continuously senses environmental changes and adjusts control strategies during reversing, thus solving the positioning deviation problem caused by terrain dynamics.

[0172] Through the synergy of real-time data acquisition and intelligent discrimination models, vehicles can adaptively adjust their stopping points, reducing the need for manual intervention while improving operational efficiency and safety.

[0173] This application further proposes a computer-readable storage medium storing computer-executable instructions. When these computer-executable instructions are invoked and executed by a processor, they cause the processor to implement a method for controlling the dumping and parking of vehicles.

[0174] Computer-readable storage media refers to physical carriers capable of persistently storing data and supporting processor access. These can be implemented using solid-state drives, flash memory chips, or optical discs. Their function is to provide a stable and reliable instruction storage environment for algorithm execution. Computer-executable instructions are sets of code written in programming languages ​​that can be compiled and executed. These can be implemented using Python, C++, or Java. Their function is to translate the vehicle's dumping and parking control methods into operational procedures that can be recognized by the hardware. A processor is an integrated circuit that performs arithmetic and logical operations. It can be implemented using multi-core CPUs, GPUs, or TPU architectures. Its function is to accelerate feature extraction, matrix factorization, and model training processes through parallel computing.

[0175] This embodiment also provides a vehicle, including a processor and a memory. The memory stores machine-executable instructions that can be executed by the processor. The processor executes the machine-executable instructions to implement the vehicle's dumping and parking control method described in the above embodiment. The vehicle can be a server or an on-board terminal.

[0176] See Figure 4 As shown, the vehicle includes a processor 400 and a memory 401. The memory 401 stores machine-executable instructions that can be executed by the processor 400. The processor 400 executes the machine-executable instructions to implement the vehicle's dumping position parking control method of the above embodiment.

[0177] Further, Figure 4 The vehicle shown also includes a bus 402 and a communication interface 403, with the processor 400, communication interface 403 and memory 401 connected via the bus 402.

[0178] The memory 401 may include high-speed random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 403 (which can be wired or wireless), such as the Internet, wide area network, local area network, or metropolitan area network. The bus 402 may be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 4The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0179] The processor 400 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of the processor 400 or by instructions in software form. The processor 400 may be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor. The software module can reside in a mature storage medium in the field, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory 401. The processor 400 reads the information in memory 401 and, in conjunction with its hardware, completes the steps of the vehicle dumping and parking control method described in the aforementioned embodiment.

[0180] This embodiment also provides a computer-readable storage medium storing computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions cause the processor to implement the vehicle dumping and parking control method of the above embodiment.

[0181] The computer program product of the vehicle dumping position parking control method, device and storage medium provided in the embodiments of this application includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods described in the preceding method embodiments. For specific implementation, please refer to the method embodiments, which will not be repeated here.

[0182] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the device described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0183] Furthermore, in the description of the embodiments of this application, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0184] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0185] In the description of this application, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0186] Finally, it should be noted that the above embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The protection scope of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the technical scope disclosed in this application. Such modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be covered within the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims.

Claims

1. A method for controlling the parking of a vehicle at its designated dumping position, characterized in that, The vehicle is equipped with data collection units distributed at the dumping port of the cargo compartment and on the rear wheels; the dumping position docking control method includes: Receive dumping site planning data from the mine's vehicle scheduling system, decode and verify the dumping site planning data, and obtain target dumping site information that matches the vehicle's own driving. Based on the vehicle's current position and orientation information and the target dumping location information, a path planning algorithm is invoked to generate a dumping operation trajectory, and navigation is performed based on the dumping operation trajectory to complete the dumping operation. When the vehicle drives to the target dumping site and reverses, the data acquisition unit collects dumping site information and vehicle operation data behind the vehicle in real time. The dumping position information and the vehicle operation data are input into a pre-built endpoint discrimination model. The endpoint discrimination model identifies whether the vehicle has reached the dumping endpoint. If it has, the vehicle braking and parking control are triggered, and the lifting device is controlled to complete the unloading operation.

2. The method for controlling the stopping position of the excavation site according to claim 1, characterized in that, The receiving mine's vehicle scheduling system sends out dumping site planning data, which is then decoded to obtain target dumping site information that matches the vehicle's own driving position. This includes: Establish a TLS-encrypted communication link and use the communication link as a medium to perform two-way authentication between the vehicle-side certificate and the dispatcher-side certificate; After completing two-way authentication, the system receives messages of dumping site planning data transmitted by the vehicle scheduling system using the MQTT protocol and encapsulated in Protobuf format based on the communication link. The fields in the message are parsed and matched with the vehicle's own pose. Based on the matching result, the data is encapsulated according to a preset data structure to obtain the coordinates, orientation angle, number, and priority of the dumping position.

3. The method for controlling the stopping position of the excavation site according to claim 2, characterized in that, The process of generating a dumping operation trajectory by invoking a path planning algorithm based on the vehicle's current position and the target dumping location information, and then navigating based on the dumping operation trajectory to complete the dumping operation, includes: The current pose information of the vehicle is obtained by using the RTK-GPS and IMU fusion positioning method. The current pose information includes the vehicle's current position coordinates, driving direction and the work route number. Using the current pose information and the coordinates of the dumping site as constraints, the improved RRT* algorithm is invoked to generate a collision-free dumping operation trajectory in a 3D mesh map. The dumping operation trajectory includes the transportation operation route and the dumping site reversing trajectory. The vehicle's navigation planning algorithm is invoked to drive sequentially based on the transportation operation route and the reversing trajectory of the dumping site, thereby completing the transportation and dumping operations.

4. The method for controlling the stopping position of the excavation site according to claim 3, characterized in that, After generating a collision-free dumping operation trajectory in a 3D mesh map by calling the improved RRT* algorithm with the current pose information and the coordinates of the dumping location as constraints, the method further includes: The reversing trajectory of the dumping site is smoothed multiple times using B-spline curve smoothing, and the smoothed reversing trajectory is constrained and corrected based on the maximum curvature limit of the vehicle.

5. The method for controlling the stopping position of the excavation site according to claim 4, characterized in that, The navigation planning algorithm for the invoked vehicle sequentially travels based on the transportation operation route and the reversing trajectory at the dumping site to complete the transportation and dumping operations, including: Based on the corrected reversing trajectory of the dumping position, the vehicle dynamics model is used to calculate the steering angle, acceleration and braking force control quantities of the vehicle when reversing in the dumping position. Based on the navigation guidance of the transportation operation route, the vehicle moves towards the target dumping site, and after moving to a preset distance from the target dumping site, the vehicle reverses to perform the dumping operation based on the steering angle, the acceleration, and the braking force control amount.

6. The method for controlling the stopping position of the excavation site according to claim 3, characterized in that, The method of obtaining the vehicle's current pose information using RTK-GPS and IMU fusion positioning includes: The vehicle's heading angle is coarsely solved by receiving dual-antenna RTK-GPS carrier phase observations and using a differential de-fuzzy algorithm to obtain the baseline vector. The vehicle's IMU is integrated within a preset time window to obtain the displacement increment, velocity increment, and attitude increment. The coarse solution of the heading angle and the incremental data are input into an extended Kalman filter to update the 15-dimensional state vector, thereby obtaining the vehicle's current pose information.

7. The method for controlling the dumping position according to any one of claims 1-6, characterized in that, The real-time acquisition of soil dumping position information and vehicle operation data behind the vehicle through the acquisition unit includes: Images of the dumping area behind the vehicle are captured by the image unit on the dumping port of the truck, and information on the cliff edge of the retaining wall and the excavated soil is identified. Read the suspension displacement sensor signal, combine it with the vehicle load information to perform threshold compensation, and obtain the suspension compression amount; Obtain the torque curve output by the wheel-side drive system, denoise the torque curve, and calculate the torque increment using a prediction model; The triaxial acceleration values ​​of the rear axle were collected and analyzed using a sliding window difference method to obtain the acceleration abrupt change. The ratio of the front and rear wheel speed difference to the vehicle speed is calculated to obtain the wheel slip ratio.

8. The method for controlling the stopping position of the excavation site according to claim 7, characterized in that, The process of inputting the dumping position information and vehicle operation data into a pre-built endpoint discrimination model, identifying whether the vehicle has reached the dumping endpoint through the endpoint discrimination model, and if so, triggering vehicle braking and parking control, and controlling the lifting device to complete the unloading operation, includes: The cliff edge information, suspension compression, torque increment, acceleration mutation, wheel slip ratio, and vehicle speed are concatenated into a six-dimensional feature vector and then batch normalized. The normalized feature vector is input into the endpoint discrimination model to calculate the collision probability that the vehicle has touched the dump wall; Based on the collision probability, it is determined whether the soil discharge endpoint has been reached. If it has, the vehicle braking and parking control are triggered, and the lifting device is controlled to complete the unloading operation.

9. A vehicle dumping position parking control device, characterized in that, The vehicle is equipped with data collection units distributed at the soil discharge port of the cargo compartment and on the rear wheels; the parking control device includes: The receiving module is used to receive the dumping site planning data issued by the mine's vehicle scheduling system, decode the dumping site planning data and verify the permissions to obtain the target dumping site information that matches the vehicle's own driving. The planning module is used to generate a dumping operation trajectory by calling a path planning algorithm based on the vehicle's current position and the target dumping location information, and to complete the dumping operation by navigating based on the dumping operation trajectory. The data acquisition module is used to collect information on the dumping site behind the vehicle and the vehicle's operation data in real time when the vehicle is driving to the target dumping site and reversing. The docking module is used to input the dumping position information and the vehicle operation data into a pre-built endpoint discrimination model. The endpoint discrimination model identifies whether the vehicle has reached the dumping endpoint. If it has, the vehicle braking and parking control are triggered, and the lifting device is controlled to complete the unloading operation.

10. A vehicle, characterized in that, The vehicle includes: a memory and at least one processor, wherein the memory stores instructions; The at least one processor invokes the instructions in the memory to cause the vehicle to perform the dumping position parking control method for the vehicle as described in any one of claims 1-8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when invoked and executed by a processor, cause the processor to implement the vehicle dumping and parking control method according to any one of claims 1-8.

Images