Control method of platoon of autonomous vehicles, loading control system, electronic device

Abstract

The application provides a control method for a platoon of autonomous vehicles, a loading control system and an electronic device. The control method comprises: using a deep learning target detection network to identify first data output by an image acquisition unit, and extracting feature information of a target object when the target object is included in the identification result; comparing the feature information with a pre-constructed reflective vest feature library to obtain a comparison result; determining that the target object is a worker wearing a reflective vest when the confidence of the comparison result is higher than a preset threshold; combining second data output by a millimeter wave radar detection unit to obtain real-time spatial coordinates of the worker wearing the reflective vest, and determining a distance, a relative speed and a collision time between the worker wearing the reflective vest and the autonomous vehicle; and establishing a risk assessment model to calculate a risk level and generate a target control strategy. Through the application, the technical problem of high collision risk caused by the inability to accurately identify workers wearing reflective vests is solved.

Description

Technical Field

[0001] This application relates to the field of autonomous driving, and in particular to control methods, loading control systems, and electronic devices for platooning autonomous vehicles. Background Technology

[0002] Currently, the transportation of bulk commodities (coal, fly ash, slag, mineral powder, etc.) within the site generally adopts a platooning vehicle round-trip transfer mode, with core scenarios covering cargo transportation and loading operations between logistics parks, coal plants, power plants, and ash and slag yards.

[0003] The current loading process still relies on a combination of manual labor and semi-automatic equipment. When convoyed vehicles enter the site for loading, manual guidance is required for parking, loading, and weighing, resulting in a high degree of human intervention throughout the process. While all staff wear reflective vests, existing vehicles lack specialized reflective vest recognition capabilities and cannot employ specific avoidance algorithms for staff, leading to a high risk of scrapes, collisions, and other safety accidents. Furthermore, the weighing, loading, and departure processes are disconnected, and the lack of linkage between weighing data and the loading process results in low loading efficiency.

[0004] There is no effective solution to the above-mentioned technical problems in the existing technology. Summary of the Invention

[0005] The control method, loading control system, and electronic equipment for platooned autonomous vehicles provided in this application are intended to solve one or more of the aforementioned technical problems.

[0006] In a first aspect, embodiments of this application provide a control method for platooned autonomous vehicles. The autonomous vehicles are equipped with a perception system, which includes at least an image acquisition unit and a millimeter-wave radar detection unit. The control method includes: using a deep learning object detection network to identify first data output by the image acquisition unit, and extracting feature information of the target object when the identification result includes a target object; using a convolutional neural network model to compare the feature information with a pre-built reflective vest feature library to obtain a comparison result; determining that the target object is a worker wearing reflective vest when the confidence level of the comparison result is higher than a preset threshold; obtaining the real-time spatial coordinates of the worker wearing reflective vest by combining second data output by the millimeter-wave radar detection unit, and determining the distance, relative speed, and collision time between the worker wearing reflective vest and the autonomous vehicle based on the real-time spatial coordinates; establishing a risk assessment model based on the distance, relative speed, and collision time to calculate a risk level; generating a target control strategy based on the risk level, and controlling the autonomous vehicles to perform safe operations based on the target control strategy.

[0007] Secondly, this application also provides a loading control system for platooned autonomous vehicles, including a platooning driving and positioning module for merging autonomous vehicles to achieve positioning by fusing multi-source positioning information and controlling the autonomous vehicles to drive to the corresponding loading point based on the positioning results. The multi-source positioning information includes real-time differential positioning (RTK) from BeiDou or GPS, simultaneous localization and mapping (SLAM) positioning from LiDAR, and wheel speed-based mileage estimation positioning. A weighing control module is also provided for vehicle identification, weight acquisition, and net weight calculation after the autonomous vehicles enter the weighing area. Finally, a target recognition and avoidance module is used to identify and avoid personnel wearing reflective vests using the aforementioned control methods. The system includes: a vehicle identification and avoidance mechanism; a loading berth docking control module, which, after identifying the characteristics of a loading berth, controls the autonomous vehicle to move to the characteristic position of the loading berth, corrects the lateral and longitudinal errors of the autonomous vehicle, determines that docking is complete when the lateral and longitudinal errors meet preset conditions, and sends a start command to the loading equipment; a quota loading shutdown control module, which, based on real-time data from the weighing control module, outputs a shutdown signal to the loading equipment to stop loading when the load weight reaches a preset loading quota threshold; and a platoon departure scheduling module, which, after verifying that the weighing data is correct, outputs a scheduling command to control the autonomous vehicle to leave the loading point.

[0008] Thirdly, embodiments of this application provide an electronic device, including a memory, a processor, and a computer program stored in the memory, wherein the processor, when executing the computer program, implements the method described in any of the above-mentioned embodiments.

[0009] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the method described in any of the above-mentioned embodiments.

[0010] Fifthly, embodiments of this application provide a computer program product, including computer instructions, which, when executed by a processor, implement the method described in any of the above-mentioned embodiments.

[0011] In this embodiment, a deep learning object detection network is used to identify the first data output by the image acquisition unit. When the identification result includes a target object, the feature information of the target object is extracted. The feature information is compared with a pre-built reflective vest feature library to obtain a comparison result. When the confidence level of the comparison result is higher than a preset threshold, the target object is determined to be a worker wearing reflective vest. The real-time spatial coordinates of the worker wearing reflective vest are obtained by combining the second data output by the millimeter-wave radar detection unit. Based on the real-time spatial coordinates, the distance, relative speed, and collision time between the worker wearing reflective vest and the autonomous vehicle are determined. A risk assessment model is established based on the distance, relative speed, and collision time to calculate the risk level. A target control strategy is generated based on the risk level, and the autonomous vehicle is controlled to perform safe operations based on the target control strategy. In other words, after identifying the target object and its feature information through a deep learning object detection network, and combining it with a pre-built reflective vest feature library, it can accurately and reliably identify workers wearing reflective vests. Based on the real-time spatial coordinates of the workers wearing reflective vests, a risk assessment is performed, and autonomous vehicles are dynamically controlled to avoid collisions based on the assessment results. This solves the technical problems in related technologies, such as the inability to accurately identify on-site workers wearing reflective vests, the inability to specifically avoid collisions with workers, and the high risk of personnel collisions. It effectively reduces the risk of collisions between vehicles and workers and improves the safety and continuity of autonomous loading operations of platooned autonomous vehicles.

[0012] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application, it can be implemented according to the contents of the specification. In order to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below. Attached Figure Description

[0013] In the accompanying drawings, unless otherwise specified, the same reference numerals throughout the drawings denote the same or similar parts or elements. These drawings are not necessarily drawn to scale. It should be understood that these drawings depict only some embodiments according to this application and should not be construed as limiting the scope of this application.

[0014] Figure 1 A schematic diagram of the structure of an autonomous driving system provided in an embodiment of this application is shown;

[0015] Figure 2 A flowchart of a control method for platooned autonomous vehicles provided in an embodiment of this application is shown;

[0016] Figure 3 This paper shows a structural block diagram of a control device for platooning autonomous vehicles provided in an embodiment of this application;

[0017] Figure 4 This invention provides a block diagram of a control device for another platooning autonomous vehicle according to an embodiment of the present application.

[0018] Figure 5 This paper shows a structural block diagram of a cargo loading control system for platooned autonomous vehicles provided in an embodiment of this application;

[0019] Figure 6 A schematic diagram of the framework of a loading control system for platooned autonomous vehicles provided in an embodiment of this application is shown.

[0020] Figure 7 A schematic flowchart of a loading control system for platooned autonomous vehicles provided in an embodiment of this application is shown.

[0021] Figure 8 A block diagram of an electronic device used to implement embodiments of this application is shown. Detailed Implementation

[0022] In the following description, only certain exemplary embodiments are briefly described. As those skilled in the art will recognize, the described embodiments can be modified in various ways without departing from the concept or scope of this application. Therefore, the drawings and description are considered to be exemplary in nature and not restrictive.

[0023] To facilitate understanding of the technical solutions of the embodiments of this application, the relevant technologies of the embodiments of this application are described below. The following relevant technologies are optional solutions and can be combined with the technical solutions of the embodiments of this application in any way, and all of them fall within the protection scope of the embodiments of this application.

[0024] The autonomous driving systems in autonomous vehicles typically include, for example: Figure 1The modules shown include: perception module 12, positioning module 14, communication module 16, planning module 18, control module 20, safety monitoring module 22, fault diagnosis module 24, and formation management module 26. The perception module 12 is used to collect information about the vehicle's surrounding environment, including roads, obstacles, vehicle distances, and traffic signals. Typically, this perception module 12 includes at least an image acquisition unit and a millimeter-wave radar detection unit. The image acquisition unit has multiple industrial cameras installed at the front, sides, and rear of the vehicles in the formation, forming a 360° full-coverage visual perception area. The cameras continuously collect images of the vehicle's surrounding environment and transmit them to the onboard computing platform in real time. For environments with high dust levels, such as coal mines and ash yards, the cameras are equipped with automatic dust covers, automatic cleaning devices, and image enhancement algorithms to improve recognition capabilities in complex environments. The millimeter-wave radar continuously detects the target's distance, speed, and direction of movement. By fusing millimeter-wave radar and visual recognition results, false recognition and missed recognition caused by dust obstruction, backlighting, low light at night, and partial obstruction can be avoided. Specifically, a Kalman filter algorithm can be used to fuse visual and radar data to obtain the real-time spatial coordinates of the target. The positioning module 14 is used to determine the absolute and relative positions of the vehicle; the communication module 16 is used to realize vehicle-to-vehicle communication and vehicle-to-infrastructure communication, which can receive commands from the lead vehicle, provide feedback on its own vehicle status to meet the needs of vehicle platooning, and also support information exchange communication for independent vehicle driving; the planning module 18 is used to generate driving trajectory and action plan based on perception and positioning data; the control module 20 is used to convert the planning results into specific operations such as steering, acceleration, and braking; the safety monitoring module 22 is used to monitor the operating parameters of each module in real time; the fault diagnosis module 24 is used to analyze the cause of the fault and determine the fault type; the platooning management module 26 is used to maintain the safe distance and position of the vehicles in the platoon and coordinate the vehicles to re-enter the platooning autonomous driving mode after the fault is recovered.

[0025] When using platooned autonomous vehicles for transportation to and from logistics parks, coal plants, power plants, ash and slag yards, existing technologies generally suffer from drawbacks such as high reliance on manual loading, significant safety hazards, poor process coordination, and insufficient versatility.

[0026] In view of the above problems, the technical solution of this application and how the technical solution of this application solves the aforementioned technical problems will be described in detail below with specific embodiments. The listed specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0027] like Figure 2 The diagram shown is a flowchart of a control method for platooned autonomous vehicles according to an embodiment of this application. The method may include:

[0028] Step S202: Use a deep learning object detection network to identify the first data output by the image acquisition unit, and extract the feature information of the target object when the identification result includes the target object.

[0029] Optionally, in this embodiment, the deep learning object detection network can be an end-to-end detection model such as the YOLO series (You Only Look Once), Region-based Convolutional Neural Network (Faster R-CNN), or Single Shot MultiBox Detector (SSD), used to locate and classify target objects in an image. For example, within the human target area, color space conversion is performed (Red Green Blue (RGB) → Hue Saturation Value (HSV) or Lab), detecting areas of high-saturation fluorescent colors (such as yellow and orange). Combined with high-brightness reflective area detection (brightness threshold segmentation, local contrast calculation), the vest reflective stripes are identified. Morphological feature analysis (stripe distribution position, width ratio, etc.) further confirms that the bright area is a reflective vest feature.

[0030] Optionally, in the embodiments of this application, the above-mentioned feature information includes, but is not limited to: fluorescent yellow area, fluorescent orange area, high reflectivity stripe area, reflective stripe geometric distribution features, and human body contour features.

[0031] Step S204: Compare the feature information with the pre-built reflective clothing feature library to obtain the comparison result.

[0032] Optionally, in this embodiment, the reflective vest feature library records the color feature values ​​(including the numerical range of fluorescent yellow and fluorescent orange in the HSV color gamut), reflectance threshold, stripe shape ratio, and corresponding multi-source sensor data of the reflective vest under different lighting conditions. When a human target is detected, the deep learning target detection network extracts color features, reflectance features, and stripe shape features in the target area and matches them with the reflective vest feature library; if the matching result meets all preset conditions, the category label "worker wearing reflective vest" is output; otherwise, "ordinary person" is output.

[0033] Step S206: When the confidence level of the comparison result is higher than the preset threshold, the target object is determined to be a worker wearing reflective clothing.

[0034] Step S208: Combine the second data output by the millimeter-wave radar detection unit to obtain the real-time spatial coordinates of the worker wearing reflective clothing, and determine the distance, relative speed and collision time between the worker wearing reflective clothing and the autonomous vehicle based on the real-time spatial coordinates.

[0035] Optionally, in this embodiment, the results of millimeter-wave radar and visual recognition are fused and verified to avoid misidentification and missed identification caused by dust obstruction, backlight environment, low light at night, and partial obstruction. Then, the Kalman filter algorithm can be used to fuse visual data and radar data to obtain real-time spatial coordinates.

[0036] Step S210: Establish a risk assessment model based on the distance, relative speed, and collision time to calculate the risk level.

[0037] Step S212: Generate a target control strategy based on the risk level, and control the autonomous vehicle to perform safe operations based on the target control strategy.

[0038] Optionally, in this embodiment, the risk levels include, but are not limited to, Level 1 risk, Level 2 risk, Level 3 risk, and Level 4 risk. Level 1 risk corresponds to a relatively far distance between the person and the vehicle, allowing normal passage. Level 2 risk corresponds to the person entering the warning area, requiring the vehicle to actively slow down. Level 3 risk corresponds to a potential collision risk, requiring the vehicle to detour and avoid the person. Level 4 risk corresponds to a possible collision, requiring the vehicle to brake suddenly and stop.

[0039] Through the above steps S202~S212, a deep learning object detection network is used to identify the first data output by the image acquisition unit, and when the identification result includes a target object, the feature information of the target object is extracted; the feature information is compared with a pre-built reflective vest feature library to obtain a comparison result; when the confidence level of the comparison result is higher than a preset threshold, the target object is determined to be a worker wearing reflective vest; combined with the second data output by the millimeter-wave radar detection unit, the real-time spatial coordinates of the worker wearing reflective vest are obtained, and the distance, relative speed, and collision time between the worker wearing reflective vest and the autonomous vehicle are determined based on the real-time spatial coordinates; a risk assessment model is established based on the distance, relative speed, and collision time to calculate the risk level; a target control strategy is generated based on the risk level, and the autonomous vehicle is controlled to perform safe operations based on the target control strategy. In other words, after identifying the target object and its feature information through a deep learning object detection network, and combining it with a pre-built reflective vest feature library, it can accurately and reliably identify workers wearing reflective vests. Based on the real-time spatial coordinates of the workers wearing reflective vests, a risk assessment is performed, and autonomous vehicles are dynamically controlled to avoid collisions based on the assessment results. This solves the technical problems in related technologies, such as the inability to accurately identify on-site workers wearing reflective vests, the inability to specifically avoid collisions with workers, and the high risk of personnel collisions. It effectively reduces the risk of collisions between vehicles and workers and improves the safety and continuity of autonomous loading operations of platooned autonomous vehicles.

[0040] In one possible implementation, the above method further includes: S11, assigning an identifier to the worker wearing reflective clothing and obtaining the corresponding movement trajectory to obtain the historical movement trajectory; S12, when the worker wearing reflective clothing is obscured, predicting the current movement position of the worker wearing reflective clothing based on the historical movement trajectory.

[0041] Optionally, the above-mentioned movement trajectory includes, but is not limited to: current position, movement speed, and movement direction.

[0042] The above steps S11~S12 can continuously track the movement trajectory of workers wearing reflective vests. When the workers wearing reflective vests are temporarily obscured by vehicles, equipment or cargo piles, their movement position can still be predicted based on historical trajectories, thus achieving continuous tracking.

[0043] Optionally, in this embodiment of the application, the above-mentioned establishment of a risk assessment model based on the distance, the relative speed, and the collision time to calculate the risk level may include: S21, multiplying the distance by a preset distance weighting coefficient to obtain a first calculation item; S22, multiplying the relative speed by a preset speed weighting coefficient to obtain a second calculation item; multiplying the collision time by a preset time weighting coefficient to obtain a third calculation item; S23, weighting the first calculation item, the second calculation item, and the third calculation item to obtain the risk level.

[0044] For example, the risk level can be calculated using the following formula:

[0045] R = α × D + β × V + γ × T (Formula 1)

[0046] Where R is the risk level; D is the distance factor; V is the relative speed factor; T is the collision time prediction factor; and α, β, and γ are weighting coefficients.

[0047] By performing weighted fusion calculations on the first calculation item, the second calculation item, and the third calculation item through the above steps S21~S23, the risk level can be calculated by comprehensively integrating multi-dimensional perception information, distinguishing the impact weight of different risk factors on the safety of autonomous vehicles, avoiding misjudgment or omission by a single detection indicator, and reducing the risk of collisions in platooned autonomous driving.

[0048] When the target control strategy corresponding to the above-mentioned risk level is to control the autonomous vehicle to detour, the autonomous vehicle is controlled to perform safe operations based on the control strategy, which may include: S31, determining whether the current environment of the autonomous vehicle meets the detour conditions; S32, if the detour conditions are met, calculating the obstacle avoidance path and controlling the autonomous vehicle to drive according to the obstacle avoidance path; S33, if the detour conditions are not met, controlling the autonomous vehicle to decelerate or stop.

[0049] Optionally, in this embodiment of the application, the method further includes: after the worker wearing reflective clothing leaves the danger zone, controlling the autonomous vehicle to automatically resume the original planned path.

[0050] For example, suppose an autonomous coal truck is traveling along a fixed loading route. Sensing cameras and millimeter-wave radar detect two workers wearing reflective vests clearing accumulated coal in the loading area lane. The risk level is assessed, and a detour control strategy is matched, initiating a safe operating procedure. The system checks if detour conditions are met, such as: no coal piles or parked trucks in the right-hand bypass lane; sufficient road width for the entire vehicle to pass; no other vehicles in the bypass lane; no obstacles within 50 meters in front and behind; site markings permitting temporary lane changes; and a minimum lateral safety distance of at least 4 meters from the workers after detour. If all conditions are met, the detour is deemed feasible. A right-biased, arc-shaped obstacle avoidance path is quickly generated and sent to the chassis actuator. The vehicle smoothly changes lanes slightly to the right, driving at low speed along the planned obstacle avoidance path past the workers' area, continuously monitoring their movements, maintaining a safe distance, and not interfering with on-site inspection operations. The radar continuously tracks the workers wearing reflective vests. Once both workers have left the danger zone and returned to the side safety walkway, the vehicle automatically corrects its trajectory, switches back to the initial coal loading route, resumes normal speed, and continues to the loader station to wait for coal loading. If the passageways on both sides around the workers wearing reflective vests are filled with raw coal and parked vehicles waiting to be loaded, leaving no space to maneuver, the vehicle is deemed unsuitable for detour. The automated coal truck gradually reduces its speed until it comes to a complete and smooth stop, remaining at a safe distance and waiting without moving forward until the workers have left the danger zone, at which point a new route plan is devised.

[0051] Through the above steps S31~S33, the vehicle adaptively selects to detour or avoid obstacles or slow down and stop according to the on-site environment. While ensuring driving safety, it reduces unnecessary parking and waiting, and balances the operational safety of autonomous vehicles with the efficiency of traffic flow within the site.

[0052] This application also proposes a platooning cooperative avoidance mechanism. Optionally, when the autonomous vehicle is the lead vehicle in the platooning autonomous driving, the above method further includes: S41, sending warning information to the following vehicles in the platoon through vehicle-to-vehicle (V2V) communication.

[0053] Optionally, the aforementioned warning information includes, but is not limited to: deceleration information, stopping information, and danger clearance information.

[0054] Optionally, formation control adopts a lead vehicle following mode, where the following distance is determined by the following formula:

[0055] (Formula 2)

[0056] in, V is the following distance; V is the vehicle speed; t is the safe following distance; S is the fixed safe following distance.

[0057] Through step S41 above, the following autonomous vehicles can enter a safe state in advance, avoiding rear-end collisions caused by sudden braking of the vehicle in front. Once the danger has passed, the lead vehicle sends a release signal, and the convoy automatically resumes normal operation.

[0058] This application embodiment, taking into account the actual operational characteristics of coal plants, power plants, ash and slag yards, also establishes a special scene recognition model. Specifically, the above method also includes: S51, when the operation scene of the autonomous vehicle is identified as the target operation scene based on the pre-built recognition model, the target control strategy is updated to obtain the updated control strategy, wherein the safety coefficient of the updated control strategy is higher than that of the target control strategy.

[0059] Optionally, the target operation scenarios mentioned above include, but are not limited to: multi-person gathering operation scenarios, nighttime inspection scenarios, loader collaborative operation scenarios, weighbridge area operation scenarios, and manual maintenance scenarios at loading docks.

[0060] The updated control strategies mentioned above could include automatically expanding the safety buffer zone, reducing vehicle speed, and increasing the safety redundancy level.

[0061] The special scene recognition mechanism proposed in step S51 above solves the technical problem that related technologies cannot adapt to different loading environments and layouts of coal plants, power plants, ash and slag yards, and logistics parks, requiring separate adaptation and resulting in high R&D and operating costs.

[0062] Corresponding to the application scenarios and methods provided in the embodiments of this application, the embodiments of this application also provide a control device for platooning autonomous vehicles, such as... Figure 3 The diagram shown is a structural block diagram of a control device for an automated platooning vehicle according to an embodiment of this application. The device may include:

[0063] The recognition module 32 is used to recognize the first data output by the image acquisition unit using a deep learning object detection network, and extract the feature information of the target object when the recognition result includes the target object;

[0064] The comparison module 34 is used to compare the feature information with the pre-built reflective clothing feature library to obtain the comparison result;

[0065] The determination module 36 is used to determine that the target object is a worker wearing reflective clothing when the confidence level of the comparison result is higher than a preset threshold.

[0066] The first calculation module 38 is used to combine the second data output by the millimeter-wave radar detection unit to obtain the real-time spatial coordinates of the worker wearing reflective clothing, and to determine the distance, relative speed and collision time between the worker wearing reflective clothing and the autonomous vehicle based on the real-time spatial coordinates.

[0067] The second calculation module 40 is used to establish a risk assessment model based on the distance, the relative speed and the collision time to calculate the risk level;

[0068] The control module 42 is used to generate a target control strategy based on the risk level, and control the autonomous vehicle to perform safe operations based on the target control strategy.

[0069] pass Figure 3 The device described uses a deep learning object detection network to identify the first data output by the image acquisition unit. When the identification result includes a target object, it extracts the feature information of the target object. It then compares this feature information with a pre-built reflective vest feature library to obtain a comparison result. When the confidence level of the comparison result is higher than a preset threshold, it determines that the target object is a worker wearing reflective vest. Combining this with the second data output by the millimeter-wave radar detection unit, it obtains the real-time spatial coordinates of the worker wearing reflective vest and determines the distance, relative speed, and collision time between the worker and the autonomous vehicle based on these coordinates. A risk assessment model is established based on the distance, relative speed, and collision time to calculate the risk level. A target control strategy is generated based on the risk level, and the autonomous vehicle is controlled to perform safe operations based on this strategy. In other words, after identifying the target object and its feature information through a deep learning object detection network, and combining it with a pre-built reflective vest feature library, it can accurately and reliably identify workers wearing reflective vests. Based on the real-time spatial coordinates of the workers wearing reflective vests, a risk assessment is performed, and autonomous vehicles are dynamically controlled to avoid collisions based on the assessment results. This solves the technical problems in related technologies, such as the inability to accurately identify on-site workers wearing reflective vests, the inability to specifically avoid collisions with workers, and the high risk of personnel collisions. It effectively reduces the risk of collisions between vehicles and workers and improves the safety and continuity of autonomous loading operations of platooned autonomous vehicles.

[0070] Optionally, the aforementioned feature information includes at least one of the following: fluorescent yellow area, fluorescent orange area, high reflectivity stripe area, geometric distribution features of reflective stripes, and human body contour features.

[0071] The device further includes a tracking module, which is used to assign an identifier to the worker wearing reflective clothing and obtain the corresponding movement trajectory to obtain the historical movement trajectory; when the worker wearing reflective clothing is obscured, the module predicts the current movement position of the worker wearing reflective clothing based on the historical movement trajectory.

[0072] The second calculation module 40 is also used to multiply the distance by a preset distance weighting coefficient to obtain a first calculation item; multiply the relative speed by a preset speed weighting coefficient to obtain a second calculation item; multiply the collision time by a preset time weighting coefficient to obtain a third calculation item; and weight the first calculation item, the second calculation item, and the third calculation item to obtain the risk level.

[0073] When the target control strategy corresponding to the risk level is to control the autonomous vehicle to detour, the control module 42 is also used to determine whether the current environment of the autonomous vehicle meets the detour conditions; when the detour conditions are met, the obstacle avoidance path is calculated and the autonomous vehicle is controlled to drive according to the obstacle avoidance path; when the detour conditions are not met, the autonomous vehicle is controlled to decelerate or stop.

[0074] When the autonomous vehicle is the lead vehicle in a platooned autonomous driving system, the above-mentioned device also includes a sending module for sending warning information to following vehicles in the platoon via vehicle-to-vehicle (V2V) communication.

[0075] In one possible implementation, the above-mentioned device further includes: an update module, used to update the target control strategy when the operation scenario of the autonomous vehicle is identified as the target operation scenario based on the pre-built recognition model, to obtain an updated control strategy, wherein the safety factor of the updated control strategy is higher than that of the target control strategy.

[0076] The functions of each module in each device in the embodiments of this application can be found in the corresponding description in the above method, and they have corresponding beneficial effects, which will not be repeated here.

[0077] The embodiments of this application will be illustrated below with specific examples.

[0078] like Figure 4As shown, this example addresses the safety risks associated with frequent worker movement and manual operation at bulk commodity yards such as coal plants, power plants, ash yards, and logistics parks. It achieves automatic identification, motion state analysis, and safe avoidance control for workers wearing reflective vests. The system includes: a perception and acquisition layer, a target recognition unit, a multi-source fusion unit, a target tracking unit, a risk assessment unit, an avoidance decision unit, an execution control unit, a formation-based collaborative avoidance unit, and a safety redundancy unit. The perception and acquisition layer includes an image acquisition unit and a millimeter-wave radar unit. The target recognition unit includes human target detection and reflective vest feature recognition. The multi-source fusion unit fuses visual and radar results. The target tracking unit includes target ID allocation, trajectory tracking, and trajectory prediction. The risk assessment unit establishes a risk assessment model based on the distance, relative speed, and direction of movement between workers and vehicles, calculates the risk level, and classifies it. The avoidance decision unit generates corresponding control strategies based on the risk level, including normal passage, deceleration, detour avoidance, and emergency braking / stopping. It also performs local path planning and generates control commands. The formation coordination and avoidance unit sends warning information to following vehicles via V2V. A safety redundancy unit operates as follows: If the camera malfunctions, millimeter-wave radar continues target detection. If the millimeter-wave radar malfunctions, the vision system continues target detection. If both vision and radar malfunction simultaneously, the vehicle enters a minimum speed operating mode and stops to wait if necessary. This ensures the system continues to operate safely even in the event of a single point of failure.

[0079] Through the above design, highly reliable identification and active avoidance of workers wearing reflective vests can be achieved in complex industrial environments, effectively reducing the risk of collisions between vehicles and workers, and improving the safety and continuity of autonomous loading operations of convoy vehicles.

[0080] Corresponding to the application scenarios and methods provided in the embodiments of this application, the embodiments of this application also provide a loading control system for platooned autonomous driving vehicles. The application scenarios of this system include, but are not limited to: Scenario 1: Coal transportation operations from coal plant to power plant, specifically including: the central dispatch module issues a "coal loading at coal plant → unloading at power plant" task; platooned vehicles travel to the coal plant according to the route; vehicles automatically drive into the weighbridge, collect tare weight and upload it; while traveling to the fixed loading point at the coal plant, the system identifies and automatically avoids workers wearing reflective vests; vehicles precisely stop below the coal bunker, triggering the loading equipment to start; during the loading process, the weight is monitored in real time, and the system automatically stops after reaching the preset limit; vehicles automatically re-weigh, calculate the net weight and upload it; the platoon travels to the power plant for unloading, completing the transportation and loading task. Scenario 2: Ash / slag transportation operations from ash / slag yard to logistics park. Specifically, the process includes: the central dispatch module issuing the task of "loading ash / slag from the ash / slag yard to the logistics park for distribution," the convoy of vehicles driving to the ash / slag yard; automatic weighing to collect tare weight, driving to the ash / slag yard discharge port, avoiding on-site inspection and operation personnel en route; automatic vehicle alignment, starting the discharge equipment to load goods, controlling the loading amount based on the weighing data to avoid overloading; a second weighing to confirm the net weight, the convoy of vehicles leaving the ash / slag yard, heading to the logistics park to complete the distribution of goods, and the data being archived synchronously.

[0081] like Figure 5 The diagram shown is a structural block diagram of a platooning autonomous vehicle loading control system according to an embodiment of this application. The system may include:

[0082] The platooning driving and positioning module 52 is used to locate the autonomous vehicle by fusing multi-source positioning information, and control the autonomous vehicle to drive to the corresponding loading point based on the positioning results. The multi-source positioning information includes real-time kinematic (RTK) positioning of Beidou or Global Positioning System (GPS), simultaneous localization and mapping (SLAM) positioning of lidar, and wheel speed odometer estimation positioning.

[0083] The weighing control module 54 is used to complete vehicle identification, weight collection and net weight calculation after the autonomous vehicle enters the weighing area.

[0084] The target recognition and avoidance module 56 is used to identify and avoid the worker wearing reflective clothing using the above control method.

[0085] The loading berth docking control module 58 is used to control the autonomous vehicle to move to the loading berth feature position after recognizing the loading berth feature, and to correct the lateral and longitudinal errors of the autonomous vehicle. When the lateral and longitudinal errors meet the preset conditions, the docking is determined to be completed, and a start command is sent to the loading equipment.

[0086] Optionally, in this embodiment of the application, after the vehicle approaches the loading berth, the characteristic markings of the loading berth are identified by lidar, and QR code markings, reflective signs or fixed structural features are extracted by the visual positioning system.

[0087] Optionally, in this embodiment, the lateral and longitudinal errors of the autonomous vehicle can be corrected using a proportional-integral-derivative (PID) algorithm.

[0088] The quota loading stop control module 60 is used to output a stop signal to stop loading when the load weight reaches the preset loading quota threshold, based on the real-time data of the weighing control module.

[0089] The platoon departure scheduling module 62 is used to output scheduling instructions after the weighing data is verified to be correct, so as to control the autonomous vehicle to leave the loading point.

[0090] pass Figure 5 The system shown features fully automated loading, improving operational efficiency and safety while reducing costs. It resolves the technical issues of fragmented weighing, loading, and departure processes in related technologies, and the lack of linkage between weighing data and the loading process, resulting in low efficiency. This achieves a high degree of automation. Furthermore, the target recognition and avoidance module 56 addresses the technical problem of inaccurately identifying workers wearing reflective vests and the inability to specifically avoid them, leading to potential collisions. This effectively reduces the risk of collisions between vehicles and workers, improving the safety and continuity of autonomous loading operations for platooned automated vehicles.

[0091] Optionally, the system may also include a scheduling module, which is used to plan a driving route to the loading point after allocating the loading point according to the task type, and to receive the weighing data, loading amount and vehicle operating status information of the autonomous vehicle.

[0092] The unified scheduling mechanism of the aforementioned scheduling module solves the technical problem in related technologies where platooned vehicles often operate independently, lack unified scheduling, and are prone to causing congestion and excessively long waiting times.

[0093] Understandably, the aforementioned scheduling module is responsible for the unified management of loading tasks in logistics parks, coal plants, power plants, and ash and slag yards. It allocates loading points and plans the optimal driving routes according to task types (coal transportation, ash transportation, slag transportation, etc.); it synchronizes weighing data, loading volume, vehicle status, and other information to form a complete operation log, supporting data traceability and docking with the site management system.

[0094] Optionally, the scheduling module includes a task management submodule, a route planning submodule, a resource allocation submodule, and a data management submodule. The task management submodule receives transportation tasks issued by the terminal management system and establishes a task queue; the resource allocation submodule calculates the comprehensive priority based on the vehicle's current location, vehicle load status, target terminal loading capacity, and weighbridge queuing status, specifically using the following formula 3:

[0095] P = αD + βT + γQ (Formula 3)

[0096] Where D is the distance between the vehicle and the target loading point; T is the estimated loading waiting time; Q is the task urgency; and α, β, and γ are weighting coefficients. Loading points are automatically allocated according to priority. The route planning submodule adopts... The algorithm combines the station's electronic map to generate the optimal route, whereby... The algorithm is a graph search-based path planning algorithm. By introducing a heuristic function, it minimizes the search range while ensuring the optimal path is found, thereby improving search efficiency. It also performs dynamic replanning based on real-time traffic conditions. The data management submodule establishes a transportation database to uniformly store vehicle information, weighing data, loading records, and operational trajectories.

[0097] The embodiments of this application will be illustrated below with specific examples.

[0098] like Figure 6 As shown, the system provided in this example adopts a three-tier architecture of central dispatch center + vehicle-mounted execution unit + field-end equipment. The central dispatch center deploys: a dispatch module. The vehicle-mounted execution unit deploys: a platooning and positioning module, a reflective vest personnel identification and safety avoidance module, a vehicle-mounted weighing data acquisition unit within the automatic weighing module, and an automatic docking module for fixed loading points. The field-end equipment deploys: a weighbridge device within the automatic weighing module; a field-end IoT communication module; and a loading equipment control unit. The central dispatch center establishes real-time communication connections with the vehicle-mounted execution units and field-end equipment via a 5G network to achieve task scheduling, status monitoring, and data synchronization.

[0099] The automatic weighing module enables unattended automatic weighing before and after loading. Vehicles automatically enter the weighbridge upon arrival, and the tare weight is collected. After loading, the module automatically re-weighs the vehicle, collecting the gross weight and automatically calculating the net weight (net weight = gross weight - tare weight). Data is uploaded to the central dispatch module in real time, eliminating the risk of cheating and ensuring data traceability. The system includes a Radio Frequency Identification (RFID) unit, a weighbridge unit, and a weight calculation unit. Upon entering the weighbridge, the system automatically identifies the vehicle; obtains the tare weight data W1; after loading, it obtains the gross weight data W2; and automatically calculates the net weight, as shown in Formula 4.

[0100] W=W2-W1 (Formula 4)

[0101] Where W represents the net weight of the goods.

[0102] When the net weight reaches 98% of the target weight, the system sends a reduction loading instruction; when it reaches 100%, loading automatically stops.

[0103] The automatic docking module for fixed loading points is compatible with fixed loading points such as coal bunkers in coal plants, power plant loading ports, and ash and slag discharge ports. Upon vehicle arrival, it automatically and accurately aligns itself, triggering the loading equipment to start. Combined with real-time data from an automatic weighing module, loading automatically stops when the cargo weight reaches a preset threshold. It supports parameterized configuration of loading logic for different sites. As the vehicle approaches the loading area, it uses LiDAR to identify loading point features; a visual positioning system extracts QR code markers, reflective signs, or fixed structural features; a PID control algorithm performs lateral and longitudinal corrections; and the vehicle gradually approaches the loading point. When the lateral error is less than 5cm and the longitudinal error is less than 10cm, docking is considered complete; a start command is then sent to the loading equipment.

[0104] The on-site IoT communication module utilizes 5G and IoT communication technologies to achieve data interoperability between platooned vehicles, weighbridges, loading equipment, the station management system, and the central dispatch module. This ensures communication even in complex environments such as logistics parks, coal plants, power plants, and ash disposal sites (e.g., dust, obstructions). The communication module employs a 5G network and industrial IoT gateway architecture. The communication protocol uses Message Queuing Telemetry Transport (MQTT) and Transmission Control Protocol / Internet Protocol (TCP / IP) to establish data links between vehicles, weighbridges, loading equipment, and the dispatch center. Data includes: vehicle location; weighing data; loading status; equipment status; and task instructions. In case of communication failure, it automatically switches to local caching mode and automatically retransmits data after connection restoration. The signal is stable with no data latency.

[0105] It should be noted that the specific parameters of each module in the embodiments of this application (such as recognition accuracy, positioning accuracy, and communication delay) can be adjusted according to the environmental requirements of different stations, and all are within the scope of protection of this invention. The system establishes a station parameter configuration library. For coal plants, power plants, ash yards and logistics parks, the following parameters are configured respectively: (1) Positioning parameters include positioning accuracy threshold, parking error threshold and loading point identification feature parameters. (2) Safety parameters include speed limit, safety distance and personnel warning area range. (3) Loading parameters include target loading weight, loading speed and loading termination threshold. (4) Communication parameters include communication frequency, data upload cycle and cache capacity. After the vehicle enters the target station, it automatically reads the corresponding station configuration file and completes parameter switching without manual intervention. New station adaptation parameters and task type implementation schemes are added. The central dispatch module establishes a station template database. Each station corresponds to: station ID, map data, loading point coordinates, weighbridge location, communication configuration and loading rules. When adding a station, only the corresponding parameter template needs to be imported to complete the adaptation. When a new transportation task is added, the following are defined through the task configuration table: cargo type, loading terminal, unloading terminal, target weight, and transportation priority. The system automatically generates a new transportation process, enabling rapid expansion. The central dispatch module can be upgraded later to add new terminal adaptation parameters and task types, improving system scalability. In the future, it can be integrated into the business settlement platform, making waybill execution and settlement unified.

[0106] The complete process of autonomous loading by the above system, such as Figure 7As shown, the process includes: Task assignment: The scheduling module assigns specific tasks based on transportation needs (such as loading coal from a coal plant to a power plant, loading ash from an ash yard to a logistics park, etc.). Fleet entry: Fleeted vehicles travel to the corresponding stations (coal plant, power plant, ash yard) according to the routes planned by the scheduling module. Automatic weighing (tare weight): Vehicles automatically enter the weighbridge area, complete tare weight collection, and upload the data to the central scheduling module. Driving to loading point: During the journey, the reflective vest personnel identification and safety avoidance module monitors on-site personnel in real time and automatically avoids them to ensure driving safety. Automatic loading alignment: Vehicles precisely stop at the fixed loading point, triggering the loading equipment to start loading. Stopping when the limit is reached: Based on real-time data from the automatic weighing module, loading automatically stops when the cargo weight reaches the preset limit. Secondary weighing (gross weight): Vehicles automatically enter the weighbridge for a second weighing, collect the gross weight, calculate the net weight, and upload the data to the central scheduling module. Departure by convoy: After loading is completed and data is confirmed to be correct, the convoy vehicles proceed to the next depot (such as unloading at a power plant or distribution at a logistics park) according to instructions. Data archiving: The weight, time, vehicle information, and depot information for this operation are all uploaded to the central dispatch module for archiving, facilitating subsequent traceability.

[0107] This application's embodiments achieve a practical application that closely matches real-world business scenarios: fully adapting to the round-trip transportation and loading needs of bulk commodities between logistics parks, coal plants, power plants, and ash and slag yards, demonstrating strong practicality. It is safe and reliable: specifically designed to identify workers wearing reflective vests, enabling automatic avoidance and significantly reducing the incidence of collisions within the site. It is highly automated: the entire process of weighing, loading, and departure is unmanned, reducing reliance on manual labor, improving operational efficiency, and shortening turnaround time within the site. It is highly versatile: the modular design allows for flexible adjustment of module parameters according to the environment and loading layout of different sites, enabling one system to adapt to multiple sites. It is highly efficient in scheduling: the central scheduling module uniformly manages tasks and vehicles, avoiding congestion within the site and improving the coordination of convoy transportation.

[0108] The functions of each module in each system of the embodiments of this application can be found in the corresponding description in the above method, and they have corresponding beneficial effects, which will not be repeated here.

[0109] Figure 8 This is a block diagram of an electronic device used to implement embodiments of this application. Figure 8 As shown, the electronic device includes a memory 801 and a processor 802. The memory 801 stores a computer program that can run on the processor 802. When the processor 802 executes the computer program, it implements the method described in the above embodiments. The number of memories 801 and processors 802 can be one or more.

[0110] The electronic device also includes:

[0111] The communication interface 803 is used to communicate with external devices and exchange and transmit data.

[0112] If the memory 801, processor 802, and communication interface 803 are implemented independently, they can be interconnected via a bus to communicate with each other. This bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. This bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 8 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0113] Optionally, in a specific implementation, if the memory 801, processor 802, and communication interface 803 are integrated on a single chip, then the memory 801, processor 802, and communication interface 803 can communicate with each other through an internal interface.

[0114] This application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method provided in this application.

[0115] This application also provides a chip including a processor for calling and executing instructions stored in a memory, causing a communication device with the chip installed to perform the method provided in this application.

[0116] This application also provides a chip, including: an input interface, an output interface, a processor, and a memory. The input interface, output interface, processor, and memory are connected through an internal connection path. The processor is used to execute code in the memory. When the code is executed, the processor is used to execute the method provided in the application embodiment.

[0117] It should be understood that the aforementioned processor can be a Central Processing Unit (CPU), or other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. General-purpose processors can be microprocessors or any conventional processor. It is worth noting that the processor can be a processor supporting Advanced Reduced Instruction Set Machines (ARM) architecture.

[0118] Further, optionally, the aforementioned memory may include read-only memory and random access memory. The memory may be volatile memory or non-volatile memory, or may include both. Non-volatile memory may include read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory may include random access memory (RAM), which serves as an external cache. By way of example, but not limitation, many forms of RAM are available. Examples include Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDR SDRAM), Enhanced Synchronous DRAM (ESDRAM), Sync Link DRAM (SLDRAM), and Direct Rambus RAM (DR RAM).

[0119] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions according to this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another.

[0120] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of those different embodiments or examples.

[0121] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "a plurality of" means two or more, unless otherwise explicitly specified.

[0122] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process. Furthermore, the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functionality involved.

[0123] The logic and / or steps described in the flowchart or otherwise herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus or device (such as a computer-based system, a processor-included system or other system that can fetch and execute instructions from, an instruction execution system, apparatus or device).

[0124] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. All or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware, the program being stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiments.

[0125] Furthermore, the functional units in the various embodiments of this application can be integrated into a single processing module, or each unit can exist physically separately, or two or more units can be integrated into a single module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. This storage medium can be a read-only memory, a disk, or an optical disk, etc.

[0126] The above description is merely an exemplary embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various variations or substitutions within the technical scope described in this application, and these should all be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A control method for platooned autonomous vehicles, characterized in that, Applied to loading and unloading yards, the autonomous vehicle is equipped with a perception module, which includes at least an image acquisition unit and a millimeter-wave radar detection unit. The control method includes: The first data output by the image acquisition unit is identified using a deep learning object detection network, and when the identification result includes a target object, the feature information of the target object is extracted, wherein the feature information includes at least one of the following: fluorescent yellow area, fluorescent orange area, high reflectivity stripe area, reflective stripe geometric distribution features, and human body contour features; The feature information is compared with a pre-built reflective clothing feature library to obtain the comparison results; When the confidence level of the comparison result is higher than a preset threshold, the target object is determined to be a worker wearing reflective clothing. Combining the second data output by the millimeter-wave radar detection unit, the real-time spatial coordinates of the worker wearing reflective clothing are obtained, and the distance, relative speed, and collision time between the worker wearing reflective clothing and the autonomous vehicle are determined based on the real-time spatial coordinates. A risk assessment model is established based on the distance, relative speed, and collision time to calculate the risk level; A target control strategy is generated based on the risk level, and the autonomous vehicle is controlled to perform safe operations based on the target control strategy. The method further includes: assigning an identifier to the worker wearing reflective clothing and obtaining the corresponding movement trajectory to obtain a historical movement trajectory; when the worker wearing reflective clothing is obscured, predicting the current movement position of the worker wearing reflective clothing based on the historical movement trajectory; A risk assessment model is established based on the distance, relative speed, and collision time to calculate the risk level, including: determining a distance factor based on the distance and multiplying the distance factor by a preset distance weighting coefficient to obtain a first calculation item; determining a relative speed factor based on the relative speed and multiplying the relative speed factor by a preset speed weighting coefficient to obtain a second calculation item; determining a collision time factor based on the collision time and multiplying the collision time factor by a preset time weighting coefficient to obtain a third calculation item; and weighting the first calculation item, the second calculation item, and the third calculation item to obtain the risk level.

2. The control method according to claim 1, characterized in that, When the target control strategy corresponding to the risk level is to control the autonomous vehicle to detour, controlling the autonomous vehicle to perform safe operations based on the control strategy includes: Determine whether the current environment of the autonomous vehicle meets the detour conditions; When the detour conditions are met, an obstacle avoidance path is calculated, and the autonomous vehicle is controlled to travel along the obstacle avoidance path. If the detour conditions are not met, the autonomous vehicle is controlled to slow down or stop.

3. The control method according to claim 2, characterized in that, When the autonomous vehicle is the lead vehicle in a platooned autonomous driving system, the method further includes: Warning information is sent to following vehicles in the queue via vehicle-to-vehicle (V2V) communication.

4. The control method according to claim 1, characterized in that, The method further includes: When the autonomous vehicle's operating scenario is identified as the target operating scenario based on the pre-built recognition model, the target control strategy is updated to obtain an updated control strategy, wherein the safety coefficient of the updated control strategy is higher than that of the target control strategy.

5. A loading control system for platooning autonomous vehicles, characterized in that, include: The platooning and positioning module is used to locate autonomous vehicles by fusing multi-source positioning information and control the autonomous vehicles to drive to the corresponding loading point based on the positioning results. The multi-source positioning information includes real-time differential positioning (RTK) of Beidou or GPS, simultaneous positioning and mapping (SLAM) positioning of LiDAR, and wheel speed odometer estimation positioning. The weighing control module is used to complete vehicle identification, weight collection, and net weight calculation after the autonomous vehicle enters the weighing area. A target recognition and avoidance module is used to identify and avoid the worker wearing reflective clothing using the method described in any one of claims 1 to 4; The loading berth docking control module is used to control the autonomous vehicle to move to the loading berth feature position after recognizing the loading berth feature, and to correct the lateral and longitudinal errors of the autonomous vehicle. When the lateral and longitudinal errors meet the preset conditions, the docking is determined to be completed, and a start command is sent to the loading equipment. The quota loading shutdown control module is used to output a shutdown signal to the loading equipment to stop loading when the load weight reaches the preset loading quota threshold, based on the real-time data of the weighing control module. The platoon departure scheduling module is used to output scheduling instructions after the weighing data is verified to be correct, so as to control the autonomous vehicles to leave the loading point.

6. The loading control system according to claim 5, characterized in that, Also includes: The scheduling module is used to plan the driving route to the loading point after allocating the loading point according to the task type, and to receive the weighing data, loading amount and vehicle operation status information of the autonomous vehicle.

7. An electronic device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory, wherein the processor, when executing the computer program, implements the method of any one of claims 1 to 4.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method described in any one of claims 1 to 4.

9. A computer program product, characterized in that, Includes computer instructions that, when executed by a processor, implement the method of any one of claims 1 to 4.

Images