Control method and device of autonomous vehicle, medium and autonomous vehicle

Abstract

This application provides a control method, device, medium, and autonomous vehicle for an autonomous vehicle. The method includes: obtaining the task type issued by a task scheduling platform, and controlling the autonomous vehicle to enter a driving state from a standby state if a task exists in the vehicle's task list; after the autonomous vehicle arrives at the target work point, controlling the autonomous vehicle to enter a waiting state according to the task type; obtaining environmental trigger information; and controlling the autonomous vehicle to enter a work execution state to complete collaborative work with a forklift when the environmental trigger information meets preset work conditions. This solves the problem that current port autonomous driving systems rely on manual intervention or additional independent modules, leading to system response delays.

Description

Technical Field

[0001] This application relates to the field of autonomous vehicle control technology, and more specifically, to an autonomous vehicle control method, an autonomous vehicle control device, a computer-readable storage medium, a processor, and an autonomous vehicle. Background Technology

[0002] Current port autonomous driving systems mostly adopt a "task-driven and path planning" control mode. Their control systems typically execute a single type of task based on a preset fixed process, such as loading and unloading operations in a fixed yard or transfer operations on the quay surface.

[0003] However, existing technologies have the following significant technical shortcomings when dealing with the complex, dynamic, and high-density multi-scenario operation environment of ports: Existing systems typically employ a unified state machine or finite state machine (FSM) structure, which cannot dynamically adjust the control process according to the task type (such as single-stage loading and unloading, two-stage mobile yard operations, container inspection, weighing, etc.). For example, for two-stage mobile yard loading and unloading tasks that require "first positioning, then docking," traditional systems must rely on manual intervention or additional independent modules for control, resulting in system redundancy and slow response. Summary of the Invention

[0004] The main objective of this application is to provide a control method, control device, computer-readable storage medium, processor, and autonomous vehicle for an autonomous vehicle, so as to at least solve the problem that current port autonomous driving systems must rely on manual intervention or additional independent modules for control, resulting in sluggish system response.

[0005] To achieve the above objectives, according to one aspect of this application, a control method for an autonomous vehicle is provided, comprising: acquiring a task type issued by a task scheduling platform, and, if a task exists in the vehicle task list, controlling the autonomous vehicle to enter a driving state from a standby state; after the autonomous vehicle arrives at a target work point, controlling the autonomous vehicle to enter a waiting state according to the task type; acquiring environmental trigger information, the environmental trigger information including cargo box approach / departure, traffic light status, barrier status, or forklift readiness status; and, if the environmental trigger information meets preset work conditions, controlling the autonomous vehicle to enter a work execution state to complete collaborative work with the forklift, wherein the state machine of the autonomous vehicle includes the standby state, the driving state, the waiting state, the work execution state, and the following positioning state.

[0006] Optionally, before controlling the autonomous vehicle to enter the operation execution state, the method further includes: when the task type is a loading and unloading task, determining that the preset operation condition indicates that the distance between the cargo box and the trailer is less than a preset distance; when the task type is a container inspection task, determining that the preset operation condition indicates that the signal light displays a preset color; and when the task type is a weighing task, determining that the preset operation condition indicates that the state of the weighbridge barrier changes from a raised state to a closed state.

[0007] Optionally, after acquiring the environmental triggering information, the method further includes: extracting the traffic light region using image segmentation and object detection techniques to obtain image features; and using a deep learning model to identify the color of the traffic light in the image features.

[0008] Optionally, after acquiring environmental triggering information, the method further includes: acquiring point cloud data collected by lidar, the point cloud data representing the three-dimensional position and shape information of the barrier, the barrier being used to allow or block the autonomous vehicle; and processing the point cloud data using a point cloud processing algorithm to determine the spatial position of the barrier and whether the barrier is in a raised state.

[0009] Optionally, before controlling the autonomous vehicle to enter the operation execution state, the method further includes: determining the orientation difference between the forklift and the target operation point according to Δθ=|θmachine-θtarget|, determining the orientation error of the forklift, the forklift being used to unload goods from the autonomous vehicle; wherein, Δθ is the orientation difference between the forklift and the target operation point, θmachine is the orientation angle of the forklift, and θtarget is the difference between the actual orientation angle of the target operation point and the expected value.

[0010] Optionally, after determining the orientation difference between the forklift and the target work point, the method further includes: determining the Euclidean distance between the forklift and the target work point to obtain the target Euclidean distance; and determining whether the forklift is ready based on the target Euclidean distance and the orientation difference between the forklift and the target work point.

[0011] According to another aspect of this application, a control device for an autonomous vehicle is provided, comprising: a first acquisition unit, configured to acquire the task type issued by a task scheduling platform, and control the autonomous vehicle to enter a driving state from a standby state when a task exists in the vehicle task list; a first processing unit, configured to control the autonomous vehicle to enter a waiting state according to the task type after the autonomous vehicle arrives at the target work point; a second acquisition unit, configured to acquire environmental trigger information, the environmental trigger information including cargo box approach / departure, traffic light status, barrier status, or forklift ready status; and a second processing unit, configured to control the autonomous vehicle to enter a work execution state to complete collaborative work when the environmental trigger information meets preset work conditions, wherein the state machine of the autonomous vehicle includes the standby state, the driving state, the waiting state, the work execution state, and the following positioning state.

[0012] According to another aspect of this application, a computer-readable storage medium is provided, the computer-readable storage medium including a stored program, wherein, when the program is executed, it controls the device on which the computer-readable storage medium is located to perform any of the methods described.

[0013] According to another aspect of this application, a processor is provided for running a program, wherein the program, when running, performs any of the methods described.

[0014] According to another aspect of this application, an autonomous vehicle is provided, comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including methods for performing any one of the methods described.

[0015] Applying the technical solution of this application, the system automatically transitions from Standby to Move state upon the existence of the task list, without manual initiation. Upon arrival at the target point, the system automatically switches to Ready state based on the task type (e.g., loading / unloading, container inspection, weighing) and binds corresponding perception conditions (e.g., crane approaching, red light on, barrier not raised), achieving automatic matching of task, environment, and state. It outputs semantic information on the status of containers, signal lights, barriers, and forklift holding containers in real time, replacing the inefficient traditional method of relying on manual observation or individual sensor threshold judgments. The introduction of Follow state enables autonomous vehicles to follow the forklift to the docking point in mobile yard scenarios without manual guidance or additional path planning modules, completely eliminating communication delays or manual confirmation delays with external devices. This solves the problem of current port autonomous driving systems relying on manual intervention or additional independent modules, leading to system response delays. Attached Figure Description

[0016] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0017] Figure 1 A flowchart illustrating a control method for an autonomous vehicle according to an embodiment of this application is shown.

[0018] Figure 2 A structural block diagram of a control device for an autonomous vehicle provided according to an embodiment of this application is shown. Detailed Implementation

[0019] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0020] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0021] It should be noted that the terms "first," "second," etc., 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 for the embodiments of this application described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover 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.

[0022] As described in the background section, existing technologies have the following significant technical shortcomings when dealing with the complex, dynamic, and high-density multi-scenario operating environment of ports: Existing systems typically employ a unified state machine or finite state machine (FSM) structure, which cannot dynamically adjust the control process according to the task type (such as single-stage loading and unloading, two-stage mobile yard operations, container inspection, weighing, etc.). For example, for two-stage mobile yard loading and unloading tasks that require "first positioning, then docking," traditional systems must rely on manual intervention or additional independent modules for control, resulting in system redundancy and sluggish response. To address the problem that current port autonomous driving systems must rely on manual intervention or additional independent modules for control, leading to sluggish system response, embodiments of this application provide a control method for an autonomous vehicle, a control device for an autonomous vehicle, a computer-readable storage medium, a processor, and an autonomous vehicle.

[0023] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0024] This embodiment provides a control method for an autonomous vehicle. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Also, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0025] Figure 1 This is a flowchart of a control method for an autonomous vehicle according to an embodiment of this application. Figure 1 As shown, the method includes the following steps:

[0026] Step S101: Obtain the task type issued by the task scheduling platform, and if there is a task in the vehicle task list, control the autonomous vehicle to enter the driving state from the standby state.

[0027] The Order Dispatch Platform (ODP) is the central task coordination and instruction generation module in the port automation operation system. It is used to dynamically generate and issue the operation task sequence of automated vehicles (ADVs) based on the port operation plan, equipment status, operation priority and real-time resource constraints.

[0028] The vehicle task list (TaskList) is a queue of job tasks that are cached locally and updated in real time by autonomous vehicles. It is dynamically distributed by the task scheduling platform (ODP) through the communication link to guide the vehicle to execute all tasks in sequence within the current job cycle.

[0029] The vehicle task list uses a structured data format (such as JSON or Protocol Buffer), and each task item contains fields:

[0030] task_id: A unique identifier for the task;

[0031] task_type: Task type (e.g., Loading, Unloading, Inspection, Weighing, TemporaryYard);

[0032] target_location: Coordinates of the target work point (X, Y);

[0033] priority: task priority;

[0034] is_two_phase: Whether it is a two-phase task (such as a mobile yard operation).

[0035] Ready State is a passive cooperative waiting state that an autonomous vehicle enters after arriving at the target work point, in order to wait for external work equipment or environmental trigger signals.

[0036] Step S102: After the autonomous vehicle arrives at the target work point, the autonomous vehicle is controlled to enter the waiting state according to the task type.

[0037] If the task type is loading / unloading: the system will initiate multimodal perception analysis of the cargo box and the crane boom. By fusing camera images and LiDAR point cloud data, the spatial relative relationships between the cargo box and the trailer, and between the crane boom and the trailer, are constructed. The system continuously monitors whether the GoodsStatus meets the trigger conditions of "cargo box approaching trailer" (loading) or "crane boom approaching trailer" (unloading). Once the perception module outputs the corresponding status as True, the state machine is immediately triggered to transition to the Work state, and collaborative operation is initiated.

[0038] If the task type is Inspection: the system focuses on the continuous recognition of the BulbStatus. Through high-precision image segmentation and color recognition models, it analyzes the red and green status of the designated BulbStatus in the work area in real time. Only when the sensing system stably outputs a "Red" signal is the inspection personnel considered to be in place and a work permit issued. The state machine then transitions from Ready to Work, and the inspection process begins.

[0039] If the task type is Weighing: The system uses LiDAR point cloud analysis and visual inspection to jointly determine the gate status of the weighbridge. When the point cloud algorithm identifies that the gate structure is not raised and the visual model confirms that the gate is in the "Close" state, the system determines that the vehicle has correctly parked in the weighing area and meets the weighing requirements, and then triggers the state transition to Work to start the weighing process.

[0040] If the task type is Temporary Yard loading / unloading: The system enters a two-stage positioning and waiting process. In the first stage, after the vehicle arrives at the first end point, it enters the Ready state. At this time, it no longer waits for the container or crane, but focuses on monitoring the forklift's ready status (IsConstructionReady). This status is determined by three sub-conditions: whether the forklift is holding a container (judged by the perception model); the angle between the forklift's orientation and the target work point is ≤30°; and the forklift is stationary and within 10 meters of the target point. Only when all three conditions are met will IsConstructionReady=True be output, triggering the state machine to enter the Follow state, guiding the vehicle to automatically follow the forklift to the second docking point. After arrival, it enters the next stage of the Ready state, waiting for the container / crane to approach, and finally enters the Work state.

[0041] Step S103: Obtain environmental trigger information, including cargo container approaching / leaving, signal light status, railing status, or forklift ready status.

[0042] Step S104: When the above-mentioned environmental trigger information meets the preset operation conditions, control the above-mentioned autonomous vehicle to enter the operation execution state in order to complete the collaborative operation with the forklift. The state machine of the above-mentioned autonomous vehicle includes the above-mentioned standby state, the above-mentioned driving state, the above-mentioned waiting in position state, the above-mentioned operation execution state and the following positioning state.

[0043] In the multi-scenario operation state machine control method for port autonomous vehicles based on task-driven and perception-triggered operation proposed in this invention, the Follow state is a key intermediate state designed specifically for two-stage tasks (such as container loading and unloading tasks in mobile yards). It is used to enable the vehicle to dynamically and accurately follow and position mobile operation equipment (such as forklifts), thereby completing the high-precision operation docking in the second stage.

[0044] Core behavior in Follow state:

[0045] Target Dynamic Update: The vehicle no longer uses fixed coordinates as the target, but instead uses the end of the forklift boom (or its preset docking point) as the dynamic target point. This dynamic target point is calculated in real time from the forklift's RTK positioning data (Xmachine, Ymachine, θmachine) and the boom extension model (e.g., boom end coordinates = forklift coordinates + boom offset vector).

[0046] Path replanning: The planning module (Planner) generates a tracking path in real time, guiding vehicles to follow the forklift machine at low speed, with high precision and a smooth trajectory. The path takes into account vehicle dynamic constraints (maximum turning angle, acceleration), safety distance (≥0.5m), and boom swing margin.

[0047] Multi-sensor fusion positioning: It integrates the vehicle's own RTK positioning, forklift RTK positioning, LiDAR point cloud matching, and visual recognition of the end marker point of the boom to achieve centimeter-level relative positioning accuracy.

[0048] RTK (Real-Time Kinematic Positioning) is a high-precision positioning technology based on the Global Navigation Satellite System, which can provide centimeter-level positioning accuracy under real-time conditions.

[0049] Attitude alignment control: The vehicle continuously adjusts its orientation (θadv) during the following process to keep the trailer centerline parallel to the boom's movement plane, ensuring that the spreader can smoothly hook up or unload the cargo box.

[0050] State switching trigger: When the relative distance between the vehicle and the dynamic target point is ≤0.1m (accuracy threshold) and the relative orientation difference Δθ is ≤5°, the docking is determined to be completed and the system automatically switches to the Work (job execution) state.

[0051] The system automatically transitions from Standby to Move mode upon the existence of a task list, without manual intervention. Upon arrival at the target point, the system automatically switches to Ready mode based on the task type (e.g., loading / unloading, container inspection, weighing) and binds corresponding sensing conditions (e.g., crane approaching, red light on, barrier not raised), achieving automatic matching of task, environment, and status. It outputs semantic information on the status of containers, signal lights, barriers, and forklift holding containers in real time, replacing the inefficient traditional method that relies on manual observation or individual sensor threshold judgments. The introduction of Follow mode enables autonomous vehicles to follow the forklift to the docking point in mobile yard scenarios without manual guidance or additional path planning modules, completely eliminating communication delays or manual confirmation delays with external devices. This solves the problem of current port autonomous driving systems relying on manual intervention or additional independent modules, resulting in system response delays.

[0052] In one embodiment of this application, before controlling the autonomous vehicle to enter the operation execution state, the method further includes: when the task type is a loading and unloading task, determining that the preset operation condition indicates that the distance between the cargo box and the trailer is less than a preset distance; when the task type is a container inspection task, determining that the preset operation condition indicates that the signal light displays a preset color; and when the task type is a weighing task, determining that the preset operation condition indicates that the state of the weighbridge barrier changes from a raised state to a closed state.

[0053] The preset distance needs to be set according to the size of the cargo container, and the preset color can be red. By using "the distance between the cargo container and the trailer is less than the preset value," "the signal light displays the preset color (such as red)," and "the weighbridge barrier changes from raised to closed" as hard trigger conditions for entering the operation execution state, the system avoids blindly starting operation actions without meeting the actual operation conditions, effectively preventing collisions, cargo damage, equipment damage, or personnel risks caused by misoperation. Especially in the high-density, multi-machine collaborative port environment, it greatly reduces the probability of accidents. Secondly, it enhances the system's robustness and adaptability to complex environments. Traditional control systems often rely on fixed time delays or simple position arrival judgments, which are easily affected by environmental interference (such as changes in lighting, obstruction, equipment delays) leading to false triggers. This solution introduces semantic condition judgment based on multimodal perception (vision, lidar, RTK), enabling the system to accurately identify the actual physical state of operation readiness, rather than relying solely on position or time, thereby maintaining stable and reliable control logic in dynamic and unstructured port scenarios.

[0054] In one embodiment of this application, after obtaining environmental triggering information, the above method further includes: extracting the traffic light region using image segmentation technology and object detection technology to obtain image features; and using a deep learning model to identify the color of the traffic light in the above image features.

[0055] The system continuously acquires images of traffic lights in the port's operational area using a vehicle-mounted high-definition camera. After preprocessing (such as noise reduction, contrast enhancement, and geometric correction), the images are input into an image segmentation model based on a fully convolutional network (FCN) or Mask R-CNN. This model performs pixel-level classification of the images, accurately locating the spatial position and contour boundaries of the traffic lights within the image, effectively eliminating background interference (such as streetlights, reflections, and other light sources), and achieving robust segmentation of the traffic light areas. Subsequently, the segmented traffic light areas are used as input to the object detection module, employing a lightweight deep learning detection model (such as YOLOv5s or EfficientDet) for color classification. During the training phase, the model has learned from a large number of labeled samples (red, green, yellow, etc.), extracting deep features such as color texture, brightness distribution, and shape structure through multi-layer convolution and attention mechanisms, ultimately outputting semantic labels for the traffic lights (such as Red or Green), achieving high-precision, low-latency color recognition.

[0056] Color recognition results serve as a key trigger signal, directly driving the state machine from "waiting in place" to "operation execution," ensuring that the vehicle only starts operation after confirming safety and compliance, greatly improving operational safety and process standardization.

[0057] In one embodiment of this application, after obtaining environmental triggering information, the method further includes: obtaining point cloud data collected by lidar, wherein the point cloud data represents the three-dimensional position and shape information of the barrier, and the barrier is used to allow or block the autonomous vehicle; and processing the point cloud data using a point cloud processing algorithm to determine the spatial position of the barrier and whether the barrier is in a raised state.

[0058] After acquiring point cloud data from LiDAR, the system first preprocesses the raw point cloud, including denoising, downsampling, and coordinate normalization, to improve data quality and reduce computational load. Then, voxel grid filtering is used to reduce the spatial dimensionality of the point cloud, preserving key structural features. Next, the RANSAC (Random Sample Consensus) algorithm is used to fit a planar or linear geometric model of the railing structure, separating the point cloud of the railing structure from other environmental obstacles (such as the ground, vehicles, and pillars). After acquiring the geometric model of the railing, the system analyzes its spatial orientation relative to a reference plane (such as the ground plane or a fixed base) to determine whether the railing is currently in a "raised" state: if the tilt angle exceeds a preset threshold (such as 15°), it is determined to be "raised"; if the tilt angle is close to horizontal (less than 5°), it is determined to be "lowered". To enhance robustness, the system further combines point cloud density and height distribution characteristics to verify whether the railing has detached from the ground contact area, eliminating false detections (such as vehicle obstruction or projection interference). Finally, based on the geometric analysis results and threshold determination, the dynamic state (Open / Close) of the barrier is output, which is used by the state machine to determine whether the passage trigger condition is met during the Ready phase.

[0059] In one embodiment of this application, before controlling the aforementioned autonomous vehicle to enter the operation execution state, the method further includes: determining the orientation difference between the forklift and the target work point according to Δθ = |θmachine - θtarget|, and determining the orientation error of the forklift, wherein the forklift is used to unload goods from the aforementioned autonomous vehicle; wherein Δθ is the orientation difference between the forklift and the target work point, θmachine is the orientation angle of the forklift, and θtarget is the difference between the actual orientation angle of the target work point and the expected value. After determining the orientation difference between the forklift and the target work point, the method further includes: determining the Euclidean distance between the forklift and the target work point to obtain the target Euclidean distance; and determining whether the forklift is ready based on the target Euclidean distance and the orientation difference between the forklift and the target work point.

[0060] Threshold judgments are made on these two key parameters (target Euclidean distance and the orientation difference between the forklift and the target work point): if the Euclidean distance D_target is less than or equal to the preset 10-meter safe working range threshold, the forklift is considered to have entered the near-field effective range of the target work area; at the same time, if the angle difference Δθ is less than or equal to 30°, the working orientation of the forklift is determined to meet the task requirements, and its boom can be effectively aligned with the vehicle trailer or cargo box.

[0061] In determining whether a forklift is ready, a comprehensive assessment of its Euclidean distance and orientation difference with the target work point significantly improves the accuracy and safety of collaborative operations. The introduction of Euclidean distance ensures the forklift is spatially close to the target work area, avoiding response delays or false triggers due to excessive distance. The orientation difference Δθ constraint ensures the forklift's operating posture is aligned with the docking direction of the vehicle / trailer, preventing inaccurate spreader alignment, cargo collisions, or loading / unloading failures due to angular deviations. This combined assessment not only eliminates misjudgments based on single conditions such as "close location but incorrect posture" or "correct posture but excessive distance," but also enables the system to identify stable states where collaborative operations are truly feasible. This dual-dimensional constraint mechanism effectively enhances the robustness of multi-machine collaboration, reduces the need for manual intervention, and strengthens the automation and reliability of port operations. Especially in high-density, high-sequence port environments, it provides crucial assurance for achieving zero-error triggering of the "perception, decision-making, and execution" closed loop.

[0062] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0063] This application also provides a control device for an autonomous vehicle. It should be noted that the control device for an autonomous vehicle in this application can be used to execute the control method for an autonomous vehicle provided in this application. This device is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0064] The following describes the control device for autonomous vehicles provided in the embodiments of this application.

[0065] Figure 2 This is a schematic diagram of a control device for an autonomous vehicle according to an embodiment of this application. Figure 2As shown, the device includes: a first acquisition unit 21, used to acquire the task type issued by the task scheduling platform, and control the autonomous vehicle to enter the driving state from the standby state when there is a task in the vehicle task list; a first processing unit 22, used to control the autonomous vehicle to enter the waiting state according to the task type after the autonomous vehicle arrives at the target work point; a second acquisition unit 23, used to acquire environmental trigger information, including cargo box approach / departure, traffic light status, barrier status, or forklift ready status; and a second processing unit 24, used to control the autonomous vehicle to enter the work execution state to complete the collaborative work when the environmental trigger information meets the preset work conditions, wherein the state machine of the autonomous vehicle includes the standby state, the driving state, the waiting state, the work execution state, and the following positioning state.

[0066] In one embodiment of this application, the device includes: a first determining unit configured to determine, before controlling the autonomous vehicle to enter the operation execution state, that the preset operation condition indicates that the distance between the cargo box and the trailer is less than a preset distance when the task type is a loading and unloading task; a second determining unit configured to determine, when the task type is a container inspection task, that the preset operation condition indicates that the signal light displays a preset color; and a third determining unit configured to determine, when the task type is a weighing task, that the preset operation condition indicates that the state of the weighbridge barrier changes from a raised state to a closed state.

[0067] In one embodiment of this application, the above-mentioned apparatus includes: a third processing unit for extracting the traffic light region using image segmentation technology and target detection technology after acquiring environmental triggering information, thereby obtaining image features; and a fourth processing unit for identifying the color of the traffic light in the above-mentioned image features using a deep learning model.

[0068] In one embodiment of this application, the device includes: a third acquisition unit for acquiring point cloud data collected by a lidar after acquiring environmental triggering information, wherein the point cloud data represents the three-dimensional position and shape information of the barrier, and the barrier is used to allow or block the autonomous vehicle; and a fifth processing unit for processing the point cloud data using a point cloud processing algorithm to determine the spatial position of the barrier and whether the barrier is in a raised state.

[0069] In one embodiment of this application, the above-mentioned device includes: a fourth determining unit configured to determine the orientation difference between the forklift and the target work point according to Δθ=|θmachine-θtarget| before controlling the above-mentioned autonomous vehicle to enter the work execution state, and to determine the orientation error of the above-mentioned forklift, the forklift being used to unload the goods on the above-mentioned autonomous vehicle; wherein, Δθ is the orientation difference between the above-mentioned forklift and the above-mentioned target work point, θmachine is the orientation angle of the above-mentioned forklift, and θtarget is the difference between the actual orientation angle of the target work point and the expected value.

[0070] In one embodiment of this application, the apparatus includes: a fifth determining unit for determining the Euclidean distance between the forklift and the target work point after determining the orientation difference between the forklift and the target work point, thereby obtaining a target Euclidean distance; and a sixth determining unit for determining whether the forklift is ready based on the target Euclidean distance and the orientation difference between the forklift and the target work point.

[0071] The control device for the aforementioned autonomous vehicle includes a processor and a memory. The first acquisition unit, first processing unit, second acquisition unit, and second processing unit are all stored as program units in the memory. The processor executes these program units stored in the memory to achieve the corresponding functions. All of the above modules reside in the same processor; alternatively, the modules may be located in different processors in any combination.

[0072] The processor contains a kernel, which retrieves the corresponding program units from memory. One or more kernels can be configured, and adjusting kernel parameters can address the problem of sluggish system response in current port autonomous driving systems that rely on manual intervention or additional independent modules.

[0073] The memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0074] This invention provides a computer-readable storage medium including a stored program, wherein, when the program is executed, it controls the device containing the computer-readable storage medium to perform the control method for the autonomous vehicle.

[0075] This invention provides a processor for running a program, wherein the program executes the control method for the autonomous vehicle.

[0076] This invention provides a device including a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs at least the following steps: obtaining the task type issued by a task scheduling platform, and if a task exists in the vehicle task list, controlling an autonomous vehicle to enter a driving state from a standby state; after the autonomous vehicle arrives at the target work point, controlling the autonomous vehicle to enter a waiting state based on the task type; obtaining environmental trigger information, including cargo box approach / departure, traffic light status, barrier status, or forklift readiness status; and if the environmental trigger information meets preset work conditions, controlling the autonomous vehicle to enter a work execution state to complete collaborative work with the forklift. The state machine of the autonomous vehicle includes the standby state, the driving state, the waiting state, the work execution state, and the following positioning state. The device described herein can be a server, PC, PAD, mobile phone, etc.

[0077] This application also provides a computer program product, which, when executed on a data processing device, is suitable for executing an initialization program having at least the following method steps: obtaining the task type issued by a task scheduling platform, and controlling an autonomous vehicle to enter a driving state if a task exists in the vehicle task list; after the autonomous vehicle arrives at the target work point, controlling the autonomous vehicle to enter a waiting state according to the task type; obtaining environmental trigger information, including cargo box approach / departure, traffic light status, barrier status, or forklift readiness status; and controlling the autonomous vehicle to enter a work execution state to complete collaborative work with the forklift when the environmental trigger information meets preset work conditions, wherein the state machine of the autonomous vehicle includes the aforementioned waiting state, driving state, waiting state, work execution state, and following positioning state.

[0078] It is obvious to those skilled in the art that the modules or steps of the present invention described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. They can be implemented using computer-executable program code, and thus can be stored in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those described herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.

[0079] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0080] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0081] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0082] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0083] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0084] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, like read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0085] Computer-readable media include both permanent and non-permanent, removable and non-removable media that can store information by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0086] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0087] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0088] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A control method for an autonomous vehicle, characterized in that, include: Obtain the task type issued by the task scheduling platform, and if there is a task in the vehicle task list, control the autonomous vehicle to enter the driving state from the standby state; After the autonomous vehicle arrives at the target work point, it is controlled to enter a waiting state according to the task type. Acquire environmental trigger information, including cargo container approach / departure, signal light status, railing status, or forklift readiness status; When the environmental trigger information meets the preset operating conditions, the autonomous vehicle is controlled to enter the operation execution state to complete the collaborative operation with the forklift. The state machine of the autonomous vehicle includes the standby state, the driving state, the positioning and waiting state, the operation execution state, and the following and positioning state.

2. The method according to claim 1, characterized in that, Before controlling the autonomous vehicle to enter the operation execution state, the method further includes: When the task type is a container loading and unloading task, the preset operating condition indicates that the distance between the container and the trailer is less than a preset distance. When the task type is a box inspection task, the preset operation condition indicator light is determined to display a preset color. When the task type is a weighing task, the preset operating conditions indicate that the state of the weighbridge guardrail changes from the raised state to the closed state.

3. The method according to claim 1, characterized in that, After obtaining the environmental trigger information, the method further includes: Image features are obtained by extracting the traffic light region using image segmentation and object detection techniques. A deep learning model is used to identify the color of the traffic lights in the image features.

4. The method according to claim 1, characterized in that, After obtaining the environmental trigger information, the method further includes: The system acquires point cloud data collected by lidar, the point cloud data representing the three-dimensional position and shape information of the barrier, the barrier being used to allow or block the autonomous vehicle; The point cloud data is processed using a point cloud processing algorithm to determine the spatial position of the railing and whether the railing is in a raised state.

5. The method according to claim 1, characterized in that, Before controlling the autonomous vehicle to enter the operation execution state, the method further includes: The orientation difference between the forklift and the target work point is determined by Δθ=|θmachine-θtarget|, and the orientation error of the forklift is determined. The forklift is used to unload goods from the autonomous vehicle. Wherein, Δθ is the orientation difference between the forklift and the target work point, θmachine is the orientation angle of the forklift, and θtarget is the difference between the actual orientation angle of the target work point and the expected value.

6. The method according to claim 5, characterized in that, After determining the orientation difference between the forklift and the target work point, the method further includes: Determine the Euclidean distance between the forklift and the target work point to obtain the target Euclidean distance; The readiness of the forklift is determined based on the target Euclidean distance and the orientation difference between the forklift and the target work point.

7. A control device for an autonomous vehicle, characterized in that, include: The first acquisition unit is used to acquire the task type issued by the task scheduling platform, and control the autonomous vehicle to enter the driving state from the standby state if there is a task in the vehicle task list. The first processing unit is configured to control the autonomous vehicle to enter a waiting state according to the task type after the autonomous vehicle arrives at the target work point. The second acquisition unit is used to acquire environmental trigger information, including cargo box approaching / leaving, signal light status, railing status, or forklift ready status. The second processing unit is used to control the autonomous vehicle to enter the operation execution state to complete the collaborative operation when the environmental trigger information meets the preset operation conditions. The state machine of the autonomous vehicle includes the standby state, the driving state, the positioning and waiting state, the operation execution state, and the following and positioning state.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device on which the computer-readable storage medium is located to perform the method according to any one of claims 1 to 6.

9. A processor, characterized in that, The processor is used to run a program, wherein the program executes the method according to any one of claims 1 to 6 when it runs.

10. An autonomous vehicle, characterized in that, include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs comprising methods for performing any one of claims 1 to 6.

Images