An unmanned system, method and access device for a luggage-pulling vehicle

Abstract

The present application relates to the technical field of luggage towing vehicle, and discloses an unmanned system, a method and an access device for a luggage towing vehicle, comprising a decision unit, the decision unit comprises a decision planning module, which is used for global path planning and local behavior decision based on task instructions, environment perception data, positioning information and towing state information, and generates a comprehensive control sequence, the comprehensive control sequence comprises vehicle driving control instructions and hitching control instructions. The unmanned system for the luggage towing vehicle adopts a first path planning constraint when the vehicle is in a non-towing state, adopts a second path planning constraint when the vehicle is in a towing state, and constrains the turning radius threshold and the curvature change continuity in the towing state, so that the generated global path or local trajectory is more in line with the characteristics of the towing marshalling operation, thereby reducing the drag hop deviation, reducing the impact at the towing connection, and improving the stability and safety during the towing turning process.

Description

Technical Field

[0001] This invention relates to the technical field of baggage towing vehicles, and more particularly to an unmanned driving system, method, and access device for baggage towing vehicles. Background Technology

[0002] In enclosed or semi-enclosed environments such as airports, ports, and logistics parks, baggage towing vehicles are typically used to tow baggage or cargo trailers, completing operations such as vehicle retrieval, coupling, transportation, parking, and uncoupling within the work area. Currently, most baggage towing operations still rely on manual driving, which presents problems such as high labor costs, high dependence on driver condition for efficiency, limited continuous operation capacity, and high safety risks. Furthermore, in scenarios with multiple concurrent tasks and frequently changing work areas, manual dispatching and driving struggle to achieve efficient coordination with upper-level dispatch systems, impacting overall operational efficiency.

[0003] With the development of autonomous driving technology, some autonomous driving technologies can already be applied to ordinary vehicle traffic scenarios. However, baggage towing vehicles have significantly different operating characteristics compared to ordinary road vehicles. First, baggage towing vehicles not only need to achieve autonomous driving, but also need to perform hook-up and unhooking operations with baggage trailers at specific work positions. Therefore, the control process involves not only the vehicle's own driving control, but also the coordinated control related to the hook-up action. Second, the kinematic characteristics of the vehicle differ significantly between non-towing and towing states. Especially in towing state, when the vehicle turns, it also needs to consider the following characteristics of the towed object. If the path planning method under single-vehicle conditions is still used, it is easy to cause excessively sharp turns, increased trailer sway, or decreased operational stability. Third, in locations such as parking positions, hook-up areas, or loading / unloading areas, vehicles are usually required to have high parking accuracy to meet the needs of subsequent hook-up, loading / unloading, or unhooking operations. Fourth, the operating area usually also contains ground support vehicles, personnel, fixed facilities, and other dynamic targets, making the environment complex and increasing the difficulty of perception, decision-making, and control of the autonomous driving system. Summary of the Invention

[0004] In view of the problem that existing autonomous driving solutions are difficult to adapt to the operation scenarios of baggage towing vehicles, this invention is proposed.

[0005] Therefore, the purpose of this invention is to provide an unmanned driving system, method, and access device for baggage towing vehicles, which is to enable the vehicle to plan routes and actions autonomously according to different environments and tasks.

[0006] To address the aforementioned technical problem of unmanned driving systems being unsuitable for baggage towing vehicles, this invention provides the following technical solution: an unmanned driving system for baggage towing vehicles, comprising a decision-making unit, wherein the decision-making unit includes a decision-planning module for performing global path planning and local behavior decisions based on task instructions, environmental perception data, positioning information, and traction status information, generating a comprehensive control sequence, wherein the comprehensive control sequence includes vehicle driving control instructions and coupling control instructions; wherein the decision-planning module includes an input interface for receiving task instructions, environmental perception data, positioning information, and traction status information; and an output interface for outputting the comprehensive control sequence to an execution side, wherein the execution side includes a vehicle execution mechanism and a coupling execution mechanism; the decision-planning module is configured to generate a global path or local trajectory according to a first path planning constraint when the vehicle is in a non-traction state; and to generate a global path or local trajectory according to a second path planning constraint when the vehicle is in a traction state; wherein the second path planning constraint includes increasing the turning radius threshold relative to the first path planning constraint and applying a curvature change continuity constraint to the local trajectory.

[0007] As a preferred embodiment of the unmanned driving system for baggage towing vehicles described in this invention, the decision-making unit further includes a communication module for receiving the task instructions issued by the cloud dispatch center; and a local task management module for caching the task instructions and providing the cached task instructions to the input interface when the communication module detects a communication interruption, so as to continue driving the decision planning module to generate the integrated control sequence.

[0008] As a preferred embodiment of the unmanned driving system for baggage towing vehicles described in this invention, the towing status information is determined based on the status feedback signal of the coupling actuator, the coupling completion signal, and / or the existence information of the towed object.

[0009] As a preferred embodiment of the unmanned driving system for baggage towing vehicles described in this invention, the decision-making unit further includes an environmental perception module, which is used to identify ground markings in the work area and output ground marking information to the input interface; the decision planning module generates guidance and control content for the vehicle to approach the target area based on the ground marking information.

[0010] As a preferred embodiment of the unmanned driving system for baggage towing vehicles described in this invention, the decision-making unit further includes a positioning and mapping module for outputting positioning information based on a high-precision map and differential positioning information; the decision planning module performs preliminary docking control based on the positioning information and performs fine-tuning of the docking position based on the ground marking information to generate the comprehensive control sequence containing position fine-tuning control instructions.

[0011] Another object of the present invention is to provide an unmanned driving control method for a baggage towing vehicle, comprising: Receive mission instructions, environmental perception data, positioning information, and traction status information; Global path planning and local behavior decisions are performed based on the task instructions, the environmental perception data, the positioning information, and the traction status information. When the vehicle is in a non-traction state, a global path or local trajectory is generated according to the first path planning constraint; when the vehicle is in a traction state, a global path or local trajectory is generated according to the second path planning constraint. Generate a comprehensive control sequence, which includes vehicle driving control commands and coupling control commands; The integrated control sequence is output to the execution side to control the vehicle actuator and the coupling actuator to perform corresponding actions.

[0012] As a preferred embodiment of the unmanned driving control method for baggage towing vehicles described in this invention, the receiving of task instructions includes: receiving task instructions issued by a cloud dispatch center; and, in the event of a communication interruption, reading pre-cached task instructions and continuing to execute the corresponding task path.

[0013] As a preferred embodiment of the unmanned driving control method for baggage towing vehicles described in this invention, the towing status information is determined based on the status feedback signal of the coupling actuator, the coupling completion signal, and / or the existence information of the towed object; the second path planning constraint includes increasing the turning radius threshold relative to the first path planning constraint and / or applying a curvature change continuity constraint to the local trajectory.

[0014] As a preferred embodiment of the unmanned driving control method for baggage towing vehicles described in this invention, the environmental perception data includes ground marking information; it also includes preliminary parking control based on the positioning information, and fine-tuning the vehicle parking process based on the ground marking information, so that the vehicle parks at the target parking position or hook-up area.

[0015] As a preferred embodiment of the unmanned driving control method for baggage towing vehicles described in this invention, the integrated control sequence includes outputting a hook-up control command when the vehicle reaches a preset hook-up position and outputting a hook-up control command when the vehicle reaches a preset unhooking position.

[0016] Another object of the present invention is to provide an unmanned access device for a baggage towing vehicle: comprising: an unmanned controller; a dual CAN access unit having a first CAN interface and a second CAN interface, the first CAN interface being used for electrical connection with CAN1-H and CAN1-L in the vehicle electrical system, and the second CAN interface being used for electrical connection with CAN2-H and CAN2-L in the vehicle electrical system; and a discrete signal acquisition unit for being electrically connected to the discrete signal terminals of the vehicle electrical system and for acquiring vehicle status signals and manual operation signals, wherein the vehicle status signals and manual operation signals include at least an ON signal, an ST signal, and a handbrake signal. The system includes: a braking signal, an emergency stop signal, and forward / reverse operation signals; an execution access unit for electrical connection with the execution component of the coupled actuator; and a control switching unit connected to the dual CAN access unit, the discrete signal acquisition unit, and the execution access unit. The autonomous driving controller is configured to output vehicle driving control commands via the dual CAN access unit and coupling control commands via the execution access unit under the automatic control state permitted by the control switching unit. The control switching unit is configured to prohibit or block the automatic output of the vehicle driving control commands and / or the coupling control commands when a manual takeover signal and / or abnormal signal is detected.

[0017] As a preferred embodiment of the unmanned access device for baggage towing vehicles described in this invention, the control switching unit is electrically connected to the emergency stop signal and is used to prohibit or block the unmanned controller from continuing to automatically output the vehicle driving control command and / or the coupling control command when the emergency stop signal is triggered, so as to put the vehicle into a safe control state.

[0018] As a preferred embodiment of the unmanned access device for baggage towing vehicles described in this invention, the dual CAN access unit is used to send the vehicle driving control command to the control node related to vehicle driving control via the CAN1 bus and / or CAN2 bus.

[0019] As a preferred embodiment of the unmanned access device for baggage towing vehicles described in this invention, the control switching unit is connected to the discrete signal acquisition unit and is configured to switch to manual control state when the braking signal, the handbrake signal and / or the forward / reverse operation signal are detected to meet the manual control conditions. At the same time, the unmanned controller is prohibited or blocked from automatically outputting the vehicle driving control command through the dual CAN access unit, and the current state of the coupling actuator is maintained.

[0020] As a preferred embodiment of the unmanned access device for baggage towing vehicles described in this invention, the discrete signal acquisition unit is further configured to acquire feedback signals related to the accelerator pedal; the execution access unit is further configured to acquire control signals and / or status feedback signals of the coupling actuator; the unmanned controller is configured to perform pedal status diagnosis based on the feedback signals, and perform coupling action anomaly diagnosis based on the control signals and / or status feedback signals, and output an anomaly signal to the control switching unit when an anomaly is diagnosed, so as to trigger manual takeover state and / or safety control state.

[0021] The beneficial effects of this invention are as follows: Addressing the scenario where baggage towing vehicles need to autonomously drive while simultaneously engaging, towing, and unhooking target baggage trailers, this invention integrates vehicle driving control commands and engagement control commands into a single comprehensive control sequence. A decision-making and planning module, based on task commands, environmental perception data, positioning information, and towing status information, makes collaborative decisions, enabling seamless transitions between different operational stages, such as approaching before engagement, towing after engagement, precise parking at the target area, and unhooking. Simultaneously, this invention employs a first path planning constraint when the vehicle is in a non-towing state and a second path planning constraint when the vehicle is in a towing state. Furthermore, it constrains the turning radius threshold and the continuity of curvature changes during towing, making the generated global path or local trajectory more consistent with the characteristics of towing formation operation. This reduces trailer sway, minimizes impact at the towing connection point, and improves stability and safety during towing turns.

[0022] This invention further enhances vehicle positioning accuracy and operational reliability by incorporating an environmental perception module, a ground marker recognition unit, and a positioning and mapping module. These modules enable vehicles to perform basic positioning and initial docking control based on high-precision maps and differential positioning information. Furthermore, they allow for precise correction of docking positions near the coupling area, parking position, or uncoupling area based on ground marker information. The unmanned driving access device provided by this invention achieves secure access to the original vehicle's electrical system through dual CAN access units. Combined with a discrete signal acquisition unit, an execution access unit, and a control switching unit, it not only enables manual takeover switching and anomaly safety protection for the vehicle's driving control link but also simultaneously acquires control signals and / or status feedback signals from the coupling actuator. It establishes an independent anomaly diagnosis and safety redundancy switching mechanism for coupling actions. Upon detecting a manual takeover signal, an emergency stop signal, or an abnormal coupling action, it can prohibit or shield automatic output and maintain the current state of the coupling actuator, thereby improving control safety and engineering feasibility during coupling operations. This further demonstrates the specialization and completeness of this invention compared to general unmanned driving solutions. Attached Figure Description

[0023] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of the decision-making unit of the unmanned driving system for baggage towing vehicles according to the present invention.

[0025] Figure 2 This is a schematic diagram of the unmanned driving system for baggage towing vehicles according to the present invention.

[0026] Figure 3 This is an electrical schematic diagram of the unmanned access device for baggage towing vehicles according to the present invention.

[0027] Figure 4 This is a schematic diagram showing the vehicle wiring relationship between the unmanned access device for baggage towing vehicles and the vehicle electrical system according to the present invention.

[0028] Figure 5 This is a schematic diagram showing the partial connection relationship between the unmanned driving controller, execution access unit, control switching unit, and vehicle control unit of the present invention.

[0029] Figure 6 This is a schematic diagram showing the partial connection relationship between the dual CAN access unit and the vehicle CAN communication harness of the present invention.

[0030] Figure 7 This is a schematic diagram showing the partial connection relationship between the discrete signal acquisition unit and the vehicle discrete signal harness of the present invention.

[0031] Explanation of reference numerals in the attached diagram: 100, Decision-making unit; 101, Decision planning module; 101a, Input interface; 101b, Output interface; 102, Execution side; 102a, Vehicle actuator; 102b, Connected actuator; 103, Communication module; 104, Local task management module; 105, Environmental perception module; 106, Positioning and mapping module; 200, Cloud dispatch center; 300, Unmanned driving controller; 301, Dual CAN access unit; 301a, First CAN interface; 301b, Second CAN interface; 302, Discrete signal acquisition unit; 303, Execution access unit; 304, Control switching unit. Detailed Implementation

[0032] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0033] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not prohibited or shielded by the specific embodiments disclosed below.

[0034] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0035] Secondly, the present invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of the present invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not according to the usual scale. Furthermore, the schematic diagrams are merely examples and should not limit the scope of protection of the present invention. In addition, actual fabrication should include the three-dimensional spatial dimensions of length, width, and depth.

[0036] Example 1

[0037] Reference Figure 1 The first embodiment of the present invention provides an unmanned driving system for a baggage towing vehicle. The system includes a decision unit 100, wherein the decision unit 100 includes a decision planning module 101, which is used to perform global path planning and local behavior decision-making based on task instructions, environmental perception data, positioning information and traction status information, and generate a comprehensive control sequence, which includes vehicle driving control instructions and coupling control instructions.

[0038] The decision planning module 101 includes an input interface 101a for receiving task instructions, environmental perception data, positioning information, and traction status information; and an output interface 101b for outputting a comprehensive control sequence to the execution side 102, which includes a vehicle execution mechanism 102a and a coupling execution mechanism 102b. The decision planning module 101 is configured to generate a global path or local trajectory according to a first path planning constraint when the vehicle is in a non-traction state; and to generate a global path or local trajectory according to a second path planning constraint when the vehicle is in a traction state.

[0039] The second path planning constraint includes increasing the turning radius threshold relative to the first path planning constraint and imposing a curvature change continuity constraint on the local trajectory.

[0040] The task instructions characterize the operational tasks and target requirements that the vehicle needs to complete, and may include information such as the target operation location, operation sequence, operational objectives, or operational status requirements. Environmental perception data characterizes the environmental conditions around the vehicle and its accessibility. Positioning information characterizes the vehicle's current position, attitude, and motion state. After processing the received information, the decision planning module 101 determines the vehicle's route from its current position to the target position and generates corresponding local behavioral decision results based on the current environmental conditions.

[0041] The integrated control sequence can be understood as a set of control commands with a sequential execution relationship generated by the decision planning module 101, rather than a single control variable. The integrated control sequence can be dynamically updated according to changes in vehicle status, position information, and environmental perception data. Vehicle driving control commands are used to control the vehicle to complete driving actions such as starting, accelerating, decelerating, braking, steering, stopping, and fine-tuning its position; coupling control commands are used to control the execution actions related to coupling operations. By unifying vehicle driving control and coupling operation control into the same integrated control sequence, the vehicle can achieve coordinated connection between driving and coupling actions during the execution of the operation task, thereby improving the continuity of operations and control consistency.

[0042] Input interface 101a is used to receive and transmit task commands, environmental perception data, and positioning information for use by decision planning module 101; output interface 101b is used to send the integrated control sequence generated by decision planning module 101 to execution side 102. Vehicle actuator 102a in execution side 102 is used to respond to vehicle driving control commands, and attachment actuator 102b is used to respond to attachment control commands.

[0043] For example, taking an airport baggage tractor towing a single baggage trailer as an example, the total mass of the trailer can be taken as 2.0t, the equivalent towing length of the trailer can be taken as 2.40m, and the planned reference speed can be taken as 3.0m / s. The above parameters are all example parameters or adjustable parameters used by the planning layer for trajectory generation and filtering. Those skilled in the art can calibrate and adjust them according to the trailer mass, operating speed, and vehicle size.

[0044] The decision-making and planning module 101 can determine the traction status based on the coupling completion signal from the coupling actuator 102b, the presence information of the towed object, the force information at the traction connection point, the trailer sway angle information, and the sway angle change rate information. When the coupling actuator 102b outputs a coupling completion signal, the confidence level of the towed object is greater than 0.70, and the horizontal resultant force at the traction connection point is greater than 50N and lasts for more than 300ms, the vehicle is determined to be in a traction state; when the coupling completion signal fails, or the horizontal resultant force at the traction connection point is less than 50N and lasts for more than 500ms, the vehicle is determined to be in a non-traction state. Furthermore, when the absolute value of the trailer sway angle is greater than 1.5° or the sway angle change rate is greater than 3° / s, it can be used as an auxiliary verification condition for the trailer to enter the actual following state. The above thresholds are all example values / adjustable parameters. By adopting the above determination method, frequent switching of the planning state can be avoided at the moment of coupling, on uneven ground, or under short-term vibration conditions.

[0045] During path planning, when the vehicle is in a non-traction state, the decision planning module 101 uses single-vehicle path planning constraints to generate the trajectory. An example constraint could be: a maximum curvature of 0.20 / m and a maximum curvature change rate of 0.06 / m². Here, curvature characterizes the degree of trajectory bending, and the curvature change rate characterizes how quickly the degree of trajectory bending changes along the trajectory length. This constraint ensures the vehicle maintains high maneuverability when not attached to a trailer.

[0046] When the vehicle is in a towing state, the decision planning module 101 uses an equivalent kinematic model of a trailer for path generation. Under the simplified planning conditions of low speed, planar motion, and neglecting the influence of tire lateral deviation, the trailer sway angle variation can be expressed as: , In the formula, α represents the trailer sway angle, which is the angle between the heading of the tractor and the heading of the trailer; s represents the arc length of the tractor along the current planned trajectory; dα / ds represents the rate of change of the trailer sway angle α relative to the trajectory arc length s, used to characterize the degree of change of the sway angle per unit distance traveled by the vehicle; κ represents the curvature of the current planned trajectory of the tractor; sinα represents the sine value of the sway angle α; L t,eq Indicates the equivalent traction length of the trailer; sinα / L t,eq This represents the sway angle recovery term introduced by the trailer following characteristics. Therefore, this formula is used to characterize that, under traction conditions, the trailer sway angle change is not only related to the curvature of the tractor's trajectory, but also to the equivalent traction length of the trailer and the current sway angle magnitude. Thus, under traction conditions, path planning cannot be performed solely based on a single-vehicle model, but sway angle changes need to be incorporated into the trajectory constraints.

[0047] The second path planning constraints under traction conditions can be set as follows: maximum curvature of 0.10 / m, maximum curvature change rate of 0.03 / m², soft threshold for swing angle of 15°, hard threshold for swing angle of 20°, soft threshold for lateral force at traction connection point of 1.8kN, and hard threshold for lateral force at traction connection point of 3.0kN. Compared with the path planning constraints under non-traction conditions, the second path planning constraints, by reducing the allowable curvature and rate of curvature change, and introducing swing angle and force constraints, make the generated trajectory more suitable for traction train operation.

[0048] Furthermore, the decision planning module 101 can approximate the lateral force at the traction connection point corresponding to the candidate trajectory, and its example relationship can be expressed as: , In the formula, Fh,y represents the estimated lateral force at the traction connection point, in N; mt represents the total mass of the trailer, in kg; v represents the current planned reference speed, in m / s; |sinα| represents the absolute value of the sine of the swing angle α; L t,eq This represents the equivalent traction length of the trailer, in meters (m); the symbol "≈" indicates an approximate estimation relationship. This formula is used at the planning level to quickly compare the force variation trends corresponding to different candidate trajectories, rather than for precise verification of the mechanical structure strength.

[0049] In one example, when the total mass of the trailer is 2000 kg, the planned reference speed is 3.0 m / s, the equivalent towing length of the trailer is 2.40 m, and the trailer sway angle is 13.8°, the estimated lateral force at the towing connection point calculated by the formula is approximately 1.79 kN; when the trailer sway angle is 24.0°, the estimated lateral force at the towing connection point calculated by the formula is approximately 3.05 kN. Therefore, it can be seen that when the sway angle increases to around 24.0°, the lateral force at the towing connection point will approach or exceed the hard threshold of 3.0 kN, and the corresponding trajectory is not suitable for the current towing conditions.

[0050] In one implementation, the decision planning module 101 can filter candidate trajectories according to the following rules: when a candidate trajectory has a curvature greater than 0.10 / m, a curvature change rate greater than 0.03 / m², a trailer sway angle greater than 20°, an estimated lateral force at the traction connection point greater than 3.0kN, or a trailer sweep envelope exceeding the current passage boundary, the candidate trajectory is directly eliminated; when a candidate trajectory does not exceed the above hard thresholds, but has a trailer sway angle greater than 15° or an estimated lateral force at the traction connection point greater than 1.8kN, a penalty cost is imposed on the candidate trajectory, and the trajectory with a smaller sway angle peak, a lower force peak, and a more continuous curvature change is preferentially selected as the target trajectory.

[0051] For example, in a 90° turn, if the candidate trajectory is generated according to the path planning constraints under non-traction conditions, the maximum sway angle can reach approximately 24.0°, and the peak lateral force at the traction connection point can reach approximately 3.05 kN. However, after determining that the vehicle has entered traction mode, the decision planning module 101 switches to the second path planning constraints to regenerate the trajectory, which can reduce the maximum sway angle to approximately 13.8° and the peak lateral force at the traction connection point to approximately 1.79 kN. This reduces trailer sway and impact at the traction connection point, improving operational stability during traction turns. The above values ​​are example values / adjustable parameters.

[0052] During operation, input interface 101a receives task instructions, environmental perception data, positioning information, and traction status information. Decision planning module 101 performs task analysis, path planning, and local behavior decisions based on this information. Depending on whether the vehicle is in a non-traction or traction state, it invokes corresponding path planning constraints to generate a global path or local trajectory. Subsequently, output interface 101b outputs the integrated control sequence to execution side 102 to control vehicle actuator 102a and coupling actuator 102b to perform corresponding actions. During vehicle operation, decision planning module 101 can also continuously adjust the integrated control sequence based on real-time changes in environmental perception data, positioning information, and traction status information, enabling the vehicle to complete autonomous driving and coupling-related operations according to task requirements.

[0053] Example 2

[0054] Reference Figure 2 This is the second embodiment of the present invention, which differs from the first embodiment in that it further includes a communication module 103 for receiving task instructions issued by the cloud scheduling center 200.

[0055] The local task management module 104 is used to cache task instructions and provide cached task instructions to the input interface 101a when the communication module 103 detects a communication interruption, so as to continue to drive the decision planning module 101 to generate a comprehensive control sequence.

[0056] The communication module 103 is used to establish a communication connection between the vehicle and the cloud dispatch center 200 to receive task instructions issued by the cloud dispatch center 200. By setting up the communication module 103, the vehicle can receive task arrangements from the upper-level dispatch system, thereby improving the unified dispatch capability of vehicle operations.

[0057] The local task management module 104 is used to locally cache, manage, and invoke task instructions received via the communication module 103. The task instructions cached by the local task management module 104 can be currently executing task instructions, or at least a portion of subsequent task instructions to be executed. Thus, when communication is normal, the local task management module 104 can provide task sources to the decision planning module 101 in conjunction with the input interface 101a; when communication is abnormal, the local task management module 104 can still continue to output cached task instructions as a local task source.

[0058] Furthermore, the communication module 103 detects a communication interruption, which refers to a state where the communication link between the vehicle and the cloud dispatch center 200 is interrupted, disconnected, abnormally lost, or unable to receive new task instructions normally. In this state, the local task management module 104 provides the cached task instructions to the input interface 101a, enabling the decision planning module 101 to continue path planning and behavior decisions based on the cached task instructions, environmental perception data, and positioning information, and generate a comprehensive control sequence. Thus, even in the event of a temporary anomaly in vehicle-cloud communication, the vehicle can still maintain the continuous execution capability of the current task stage, thereby improving the continuity and reliability of system operation.

[0059] During operation, the cloud dispatch center 200 first sends task instructions to the vehicle through the communication module 103, and the local task management module 104 caches the received task instructions. When communication is normal, the task instructions are input to the decision planning module 101 through the input interface 101a. The decision planning module 101 combines environmental perception data and positioning information to generate a comprehensive control sequence, which is then output to the execution side 102 for execution through the output interface 101b. If the communication module 103 detects a communication interruption during vehicle operation, the local task management module 104 provides the pre-cached task instructions to the input interface 101a to continue driving the decision planning module 101 to generate a comprehensive control sequence, enabling the vehicle to continue completing the current stage of the operation. After communication is restored, the communication module 103 can re-receive task instructions from the cloud dispatch center 200 to restore normal dispatch status.

[0060] The remaining structure is the same as that in Example 1.

[0061] Example 3

[0062] Reference Figure 2 This is the third embodiment of the present invention. The difference between this embodiment and the second embodiment is that the decision planning module 101 switches the path planning constraints based on the traction state information, and dynamically adjusts the second path planning constraints when the vehicle is in traction state, so as to generate a global path or local trajectory that is more suitable for traction train operation.

[0063] In one implementation, the traction status information is determined based on the status feedback signal of the coupling actuator 102b, the coupling completion signal, and / or the presence information of the towed object.

[0064] For example, when the coupling actuator 102b outputs a locking feedback signal and detects that the target luggage trailer has established a connection with the vehicle, it can be determined that the vehicle is in a towing state; when the coupling actuator 102b outputs an unlocking feedback signal, or when no towing object is detected to still be connected to the vehicle, it can be determined that the vehicle is in a non-towing state. This provides a clear state basis for the decision planning module 101 to switch path planning constraints.

[0065] In non-traction mode, the decision planning module 101 generates a global path or local trajectory according to the first path planning constraints. At this time, the vehicle performs path planning according to the single vehicle working condition to take into account both traffic efficiency and steering flexibility.

[0066] In traction mode, the decision planning module 101 switches to generating a global path or local trajectory according to the second path planning constraints. The second path planning constraints include increasing the turning radius threshold relative to the first path planning constraints and / or applying curvature change continuity constraints to the local trajectory, thereby reducing bucket sway, increased following error, and operational instability caused by sharp turns or trajectory changes during traction operation.

[0067] Furthermore, in a preferred embodiment, the decision planning module 101 can dynamically adjust the second path planning constraints based on the number of towed objects, the total length of the convoy, the total mass of the trailer, and / or the current operating speed.

[0068] For example, when only a single baggage trailer is attached, the second path planning constraint can use the first set of traction constraint parameters; when multiple baggage trailers are attached, the total length of the formation increases, or the total mass of the trailers increases, the decision planning module 101 can further increase the turning radius threshold, reduce the upper limit of allowable curvature, and / or reduce the upper limit of curvature change rate, so that the planning results match the actual operating capacity under longer formations or larger traction loads.

[0069] Therefore, the second path planning constraint is not fixed, but can be adaptively adjusted according to changes in the traction formation status.

[0070] In one example, when the vehicle is in a traction state and the number of traction objects increases, the decision planning module 101 can further improve the path smoothness requirements based on the existing second path planning constraints, and prioritize candidate trajectories with larger turning radii and more continuous curvature changes. When the vehicle is in a traction state but the current operating speed decreases, the decision planning module 101 can appropriately relax the local trajectory adjustment range under the premise of meeting safety constraints, so as to take into account both the alignment accuracy and operation stability under low-speed conditions.

[0071] During operation, after receiving the task instruction, the decision planning module 101 first combines the feedback information from the coupling actuator 102b and the information on the existence of the towed object to determine whether the vehicle is currently in a non-traction state or a traction state. When the vehicle is in a non-traction state, a global path or local trajectory is generated according to the first path planning constraint. When the vehicle completes coupling and enters the traction state, it switches to generating a global path or local trajectory according to the second path planning constraint. If the traction grouping parameters change further, such as an increase in the number of towed objects or an increase in the grouping length, the second path planning constraint is dynamically adjusted to output a comprehensive control sequence that is more suitable for the current traction conditions.

[0072] The remaining structure is the same as that in Example 2.

[0073] Example 4

[0074] Reference Figure 2 This is the fourth embodiment of the present invention, which differs from the third embodiment in that the decision-making unit 100 further includes an environmental perception module 105 and a positioning and mapping module 106. The environmental perception module 105 is used to identify ground markings in the work area and output ground marking information to the input interface 101a; the positioning and mapping module 106 is used to output positioning information based on high-precision maps and differential positioning information; the decision planning module 101 performs preliminary docking control based on the positioning information and performs fine-tuning of the docking position based on the ground marking information to generate a comprehensive control sequence containing position fine-tuning control instructions.

[0075] The environmental perception module 105 is used to acquire information related to the environment around the vehicle. The ground marking recognition unit in the environmental perception module 105 is used to recognize the ground markings set in the work area, and output the corresponding ground marking information to the input interface 101a.

[0076] In one embodiment, the ground marking recognition unit includes a vision camera positioned facing downwards and forwards of the vehicle, and an image processing module connected to the vision camera. The vision camera is used to acquire image information of the ground area in front of the vehicle, and the image processing module is used to identify ground markings from the acquired ground images.

[0077] In this embodiment, the ground markings are preferably pre-marked lines on the ground within the work area, such as road guide lines, lane boundary lines, parking space border lines, parking space center lines, stop lines, attachment area boundary lines, and parking space border lines.

[0078] Specifically, the image processing module can perform distortion correction, region of interest cropping, edge extraction, and marking contour recognition on the acquired ground images to extract the position and orientation information of the target ground markings. Furthermore, the ground marking recognition unit pre-stores the intrinsic and extrinsic parameters of the visual camera, as well as the camera's mounting pose parameters relative to the vehicle reference point. Based on the identified target ground markings, it transforms the marking features in the image coordinate system to the vehicle coordinate system or the ground coordinate system, thereby calculating the lateral deviation Δx, longitudinal deviation Δy, and orientation deviation Δθ of the vehicle relative to the target marking.

[0079] Among them, the lateral deviation Δx is used to characterize the left-right offset of the vehicle's centerline relative to the target marker's centerline, the longitudinal deviation Δy is used to characterize the front-back distance offset of the vehicle's reference point relative to the stop line or target parking position, and the orientation deviation Δθ is used to characterize the angular deviation between the vehicle's current heading and the target marker's direction. The ground marking information includes at least the lateral deviation Δx, longitudinal deviation Δy, and orientation deviation Δθ.

[0080] For example, when the identified object is a parking space, the center line of the parking space can be determined based on the left and right boundary lines of the parking space, and the lateral deviation Δx can be obtained based on the positional relationship between the vehicle's center point and the center line of the parking space; the longitudinal deviation Δy can be obtained based on the distance from the vehicle's reference point to the parking space's stop line; and the orientation deviation Δθ can be obtained based on the angle between the vehicle's current heading and the direction of the parking space's center line.

[0081] When the identified object is the attachment area, the lateral deviation Δx, longitudinal deviation Δy, and orientation deviation Δθ of the vehicle relative to the attachment area can be obtained based on the center guide line and boundary line of the attachment area.

[0082] The positioning and mapping module 106 is used to determine the vehicle's current position based on a high-precision map and differential positioning information, and outputs the positioning information. The high-precision map provides the vehicle's basic path and target point information within the work area, while the differential positioning information improves the vehicle's basic positioning accuracy. Therefore, the positioning information output by the positioning and mapping module 106 supports the vehicle's global path planning and initial docking control within the work area.

[0083] After receiving the positioning information, the decision planning module 101 can perform initial docking control based on the positioning information. When the vehicle approaches the target docking area, it further adjusts the comprehensive control sequence based on the lateral deviation Δx, longitudinal deviation Δy, and orientation deviation Δθ to generate control content for fine-tuning direction, low-speed approach, and parking correction. In other words, the positioning and mapping module 106 provides the basic positioning basis for the vehicle within the work area, while the lateral deviation Δx, longitudinal deviation Δy, and orientation deviation Δθ output by the ground marking recognition unit are used to finely correct the vehicle's docking position near the target area, thereby improving the docking accuracy and control stability of the vehicle near parking spaces, hook-up areas, or parking positions.

[0084] During use, the positioning and mapping module 106 first outputs the positioning information corresponding to the vehicle's current position based on the high-precision map and differential positioning information. The decision planning module 101 performs global path planning and local behavior decision-making based on the positioning information. When the vehicle runs near the target work area, the ground marking recognition unit in the environmental perception module 105 identifies the corresponding ground markings and inputs the lateral deviation Δx, longitudinal deviation Δy, and orientation deviation Δθ calculated from the identified ground markings as ground marking information to the decision planning module 101 via the input interface 101a. Subsequently, the decision planning module 101 combines the positioning information with the lateral deviation Δx, longitudinal deviation Δy, and orientation deviation Δθ to make a comprehensive judgment on the vehicle's current position, approach direction, and parking position, and generates a comprehensive control sequence suitable for precise parking. This sequence is output to the execution side 102 via the output interface 101b to control the vehicle to gradually complete the approach, position correction, and parking actions.

[0085] Example 5

[0086] A fifth embodiment of the present invention further provides an unmanned driving control method for a baggage towing vehicle, comprising: Receive mission instructions, environmental perception data, positioning information, and traction status information; Global path planning and local behavior decisions are made based on task instructions, environmental perception data, positioning information, and traction status information. When the vehicle is in a non-traction state, a global path or local trajectory is generated according to the first path planning constraint; when the vehicle is in a traction state, a global path or local trajectory is generated according to the second path planning constraint. Generate a comprehensive control sequence, which includes vehicle driving control commands and coupling control commands; The integrated control sequence is output to the execution side 102 to control the vehicle actuator 102a and the attached actuator 102b to perform corresponding actions.

[0087] The task instructions may include at least one of the following: target hook-up area, target parking position, target unhooking area, and task execution sequence. In the operational scenario of this embodiment, the hook-up actuator 102b is used to perform the hook-up and unhooking actions between the baggage tractor and the target baggage trailer.

[0088] In one embodiment, the coupling actuator 102b is disposed at the rear of the baggage tractor and includes a coupling seat, a locking member that can switch between a locked position and an unlocked position, and an electric linear actuator for driving the locking member. Furthermore, the coupling actuator 102b may also include a locking position sensor, an unlocking position sensor, and a coupling object presence sensor for feedback on the current state of the coupling mechanism.

[0089] In one embodiment, the engagement control command includes an engagement control command and a disengagement control command. The engagement control command controls the electric linear actuator to drive the locking member from the unlocked position to the locked position, thereby locking the towing connector of the target baggage trailer within the engagement mount. The disengagement control command controls the electric linear actuator to drive the locking member from the locked position to the unlocked position, thereby releasing the towing connector of the target baggage trailer. The engagement control command can be a switching action command, and may further include displacement control information for defining the target travel position of the locking member.

[0090] In one implementation, when the decision planning module 101 determines that the vehicle has reached the preset docking position based on the positioning information, environmental perception data and ground marking information, and the relative position between the vehicle and the target luggage trailer meets the docking conditions, it outputs a docking control command to control the docking actuator 102b to perform a locking action; when the locking position sensor outputs locking completion feedback, it determines that the docking action is completed and switches the vehicle state to traction state.

[0091] When the vehicle reaches the preset uncoupling position and the current state meets the uncoupling safety conditions, an uncoupling control command is output to control the coupling actuator 102b to perform the unlocking action; when the unlocking sensor outputs the unlocking completion feedback, it is determined that the uncoupling action is completed and the vehicle state is switched to non-traction state.

[0092] In one embodiment, the traction status information can be determined based on the locking feedback signal of the coupling actuator 102b, the presence sensor signal of the coupling object, and / or the trailer connection status information identified by the environmental perception module. When the coupling actuator 102b outputs locking feedback and detects that the target luggage trailer is in a coupled state, it can be determined that the vehicle is in a traction state; when the coupling actuator 102b outputs unlocking feedback, or does not detect that the target luggage trailer is still connected to the vehicle, it can be determined that the vehicle is in a non-traction state.

[0093] During operation, the vehicle first proceeds to the target docking area according to the task instructions. Upon reaching the preset docking position, it outputs a docking control command to lock the target luggage trailer using the docking actuator 102b. After docking, the vehicle guides the target luggage trailer to the target parking position according to the second path planning constraint. After parking, if the task instructions also include unhooking, the vehicle continues to the target unhooking area and outputs a unhooking control command upon reaching the preset unhooking position to unlock and unhook the trailer using the docking actuator 102b.

[0094] For example, in an airport baggage transfer scenario, the cloud-based dispatch center 200 issues a task instruction to a baggage towing vehicle in the waiting area. The task instruction can include the vehicle departing from its current location, proceeding to the designated hook-up area, hooking up with the target baggage trailer, towing the target baggage trailer to the baggage loading / unloading area corresponding to the target parking position, and then proceeding to the designated unhooking area to perform the unhooking operation. The task instruction can include at least the current starting location, the target hook-up area, the target parking position, the target unhooking area, and the task execution sequence information. After being received by the communication module 103, the task instruction can be cached by the local task management module 104 so that the corresponding task can continue to be executed in the event of a communication interruption.

[0095] Upon receiving the aforementioned task instruction, the vehicle is currently in a non-towing state. At this time, the positioning and mapping module 106 outputs the vehicle's current position and current vehicle orientation information. For example, the vehicle is located near the exit of the waiting area, with its front facing the main channel, and according to the high-precision map, it can be determined that there are passable road sections, turning connection sections, and approach sections between the current position and the docking area. At the same time, the environmental perception module 105 outputs current environmental perception data, such as no fixed obstacles occupying the main channel, a support vehicle passing at low speed a certain distance ahead, ground markings for guiding docking and parking near the docking area, and the target luggage trailer located within the docking area. After receiving the aforementioned task instruction, environmental perception data, and positioning information, the decision planning module 101 first parses the current stage objective as reaching the docking area and completing the pre-dock parking alignment, and then determines the global path from the current position to the docking area based on the high-precision map. This global path can be planned, for example, as traveling straight along the road section from the exit of the waiting area, passing through the turning connection section, entering the approach section, and finally reaching the preset docking position in front of the docking area.

[0096] After the global path is determined, the decision-making and planning module 101 further makes local behavior decisions. Specifically, when the environmental perception module 105 detects that a support vehicle is passing through the main channel 20 meters ahead, the decision-making and planning module 101 will not immediately execute according to the maximum traffic efficiency, but will first generate a local trajectory of deceleration and waiting, so that the vehicle can resume driving along the global path after the support vehicle has passed; when the main channel is detected to be clear again, a local trajectory for continuing to move forward is generated. That is to say, in this stage, the global path solves the problem of the vehicle moving from its current position to the preset docking position, while the local behavior decision solves the problem of whether to continue moving forward, decelerate and wait, or approach at a low speed at the current moment. Subsequently, the decision-making and planning module 101 generates a corresponding integrated control sequence, such as first outputting a start command, then outputting a command to accelerate to the preset low-speed cruising speed, outputting deceleration and steering commands when approaching the turning connection section, outputting a brake holding command when the support vehicle is detected to have passed, and outputting a command to continue moving forward after traffic is restored. After the integrated control sequence is sent to the execution side 102 via the output interface 101b, the vehicle's actuator 102a is controlled to complete the corresponding actions.

[0097] When the vehicle enters the vicinity of the docking area, the ground marking recognition unit in the environmental perception module 105 identifies the parking guidance markings in the docking area, such as the docking center line, stop line, or docking area number markings. At this time, although the positioning and mapping module 106 has already provided the basic positioning result of the vehicle being near the docking area, the decision planning module 101 will further judge the lateral deviation, longitudinal deviation, and vehicle orientation deviation between the vehicle's current position and the preset docking position by combining the recognized ground marking information. If the judgment result is that the vehicle is 0.25 meters to the left of the docking center line, the vehicle's orientation deviation is 3 degrees, and the distance to the stop line is still 1.2 meters, then the decision planning module 101 will adjust the local trajectory accordingly. For example, it will reduce the vehicle speed to a lower approach speed, first generate a slight right turn correction trajectory, then generate a small straight approach trajectory, and finally generate a braking and stopping trajectory to make the vehicle accurately stop at the preset docking position. At this time, the control content in the integrated control sequence is no longer mainly based on conventional traffic control, but instead shifts to position fine-tuning control, such as low-speed forward command, small-angle correction command, and parking braking command.

[0098] When the decision planning module 101 determines, based on the positioning information and ground marking information, that the vehicle has reached the preset docking position and the relative position between the vehicle and the target luggage trailer meets the docking conditions, it outputs a docking control command to the docking execution mechanism 102b to execute the corresponding docking action. After docking, the vehicle's operating state switches from non-traction state to traction state. At this point, the task enters the next stage, which is towing the target luggage trailer from the docking area to the target parking position. When replanning the path, the decision planning module 101 no longer generates a path according to the first path planning constraint in the non-traction state, but switches to the second path planning constraint in the traction state. Specifically, under this constraint, the planning result will try to avoid excessively sharp turns, excessively small turning radii, and excessively large trajectory curvature changes, and instead prioritize paths with larger turning radii and smoother trajectories. For example, if there are two feasible routes from the docking area to the target parking position, where one route is shorter but contains a sharp turn, and the other route is slightly longer but has a gentler turn, the decision planning module 101 will prioritize the latter route to reduce the risk of trailer swaying or unstable following in the traction state.

[0099] During towing, the environmental perception module 105 continuously monitors the surrounding environment. For example, if the system detects two workers temporarily entering the edge area of ​​the work lane ahead, and equipment parked on the right occupies part of the passable space, the decision-making and planning module 101 adjusts the local behavior based on real-time environmental perception data. It can first generate a deceleration trajectory, and if necessary, briefly stop at a safe location until the personnel leave the danger zone before continuing. Since the vehicle is in a towing state, the local trajectory adjustment also follows the second path planning constraint, that is, to use a gentler directional correction as much as possible to avoid excessive sweeping of the trailer tail while the vehicle can pass. As a result, the integrated control sequence is dynamically updated, including a series of continuous control commands such as reducing speed, maintaining traction stability, slow steering, and resuming cruise control, rather than a single fixed instruction.

[0100] As the vehicle approaches the loading / unloading area corresponding to the target parking position, the environmental perception module 105 re-identifies the ground markings of the parking position, such as the parking position number, parking frame lines, or loading / unloading docking lines. The positioning and mapping module 106 provides a basic position result indicating that the vehicle has reached the vicinity of the target parking position. For example, the vehicle is still 1.8 meters away from the center of the target parking frame, and the vehicle's orientation is deflected by 2 degrees relative to the center line of the parking frame. The decision-making and planning module 101, after combining the ground marking information, determines that the vehicle cannot stop directly at this point, otherwise the trailer will not fall accurately into the loading / unloading docking range. Therefore, it further generates a local trajectory for final parking correction. For example, it first controls the vehicle to continue moving forward at a very low speed for 0.8 meters, then makes a small-angle directional correction to make the vehicle and the towed trailer coincide with the center line of the target parking frame, and then continues to move forward at a low speed for 0.9 meters and applies the parking brake. In this way, the vehicle can finally stop at the predetermined position of the target parking position, meeting the needs of subsequent loading and unloading operations.

[0101] After parking at the target parking position, if the task instruction also includes proceeding to the uncoupling area to perform uncoupling, the decision planning module 101 continues to generate the path and control sequence from the current parking position to the uncoupling area. When it is determined that the vehicle has reached the preset uncoupling position, a uncoupling control instruction is output to control the coupling actuator 102b to perform the uncoupling action, thereby completing the task. If a vehicle-to-cloud communication interruption occurs during any of the above stages, for example, if the vehicle temporarily loses its communication connection with the cloud dispatch center 200 while traveling from the coupling area to the target parking position, the local task management module 104 calls the pre-cached task instruction, causing the decision planning module 101 to continue executing subsequent control according to the determined current task objective and task path until communication is restored or the current task stage is completed.

[0102] Example 6

[0103] Reference Figures 3-7 This is the sixth embodiment of the present invention, which also provides an unmanned access device for a baggage towing vehicle: including an unmanned controller 300.

[0104] The dual CAN access unit 301 has a first CAN interface 301a and a second CAN interface 301b. The first CAN interface 301a is used to electrically connect with CAN1-H and CAN1-L in the vehicle electrical system, and the second CAN interface 301b is used to electrically connect with CAN2-H and CAN2-L in the vehicle electrical system.

[0105] The first CAN interface 301a corresponds to the original vehicle driving control related communication link, which is used to exchange messages with the control nodes related to power, braking, steering and driving status; the second CAN interface 301b corresponds to the original vehicle body status and auxiliary control related communication link, which is used to exchange messages with the control nodes related to body control, status feedback and auxiliary functions.

[0106] Furthermore, the dual CAN access unit 301 can transparently transmit and / or rewrite the control messages output by the autonomous driving controller 300 according to the original vehicle communication protocol format and send them to the corresponding control node, and receive the status feedback messages returned by the original vehicle control node.

[0107] In this context, "transparent transmission" refers to the dual CAN access unit 301 directly forwarding the corresponding control message when the message format output by the autonomous driving controller 300 is consistent with the original vehicle link protocol. "Protocol-modified transmission" refers to the dual CAN access unit 301 performing protocol adaptation on the control message before transmission when the message identifier, data field arrangement, byte definition, or checksum format output by the autonomous driving controller 300 is inconsistent with the original vehicle link protocol. In one example, the original vehicle communication protocol can be the J1939 protocol message or the original vehicle's custom CAN message format; this embodiment does not limit this.

[0108] The discrete signal acquisition unit 302 is used to electrically connect to the discrete signal terminal of the vehicle electrical system and acquire vehicle status signals and manual operation signals. The vehicle status signals and manual operation signals include at least ON signal, ST signal, handbrake signal, braking signal, emergency stop signal, and forward / reverse operation signal.

[0109] The execution access unit 303 is used for electrical connection with the execution component of the attached execution mechanism 102b. The execution access unit 303 is also used to collect control signals and / or status feedback signals from the attached execution mechanism 102b. Status feedback signals may include lock-in feedback, unlock-in feedback, attachment object presence feedback, and / or actuator fault feedback.

[0110] The control switching unit 304 is connected to the dual CAN access unit 301, the discrete signal acquisition unit 302, and the execution access unit 303, and is used to switch between automatic control state, manual takeover state, and safety control state.

[0111] In the automatic control state, the unmanned driving controller 300 is allowed to output vehicle driving control commands via the dual CAN access unit 301 and output docking control commands via the execution access unit 303; in the manual takeover state, the unmanned driving controller 300 is prohibited from or blocked from continuing to automatically output vehicle driving control commands, and the current state of the docking actuator 102b is maintained; in the safety control state, the automatic output of vehicle driving control commands and / or docking control commands is prohibited or blocked, so that the vehicle enters a restricted operation, parking or standby state.

[0112] The unmanned driving controller 300 is used to output vehicle driving control commands via dual CAN access unit 301 and to output hook-up control commands via execution access unit 303 under the automatic control state permitted by control switching unit 304; the control switching unit 304 is used to prohibit or block the automatic output of vehicle driving control commands and / or hook-up control commands when a manual takeover signal and / or abnormal signal is detected.

[0113] In one embodiment, the control switching unit 304 is electrically connected to the emergency stop signal and is used to prohibit or block the autonomous driving controller 300 from continuing to automatically output vehicle driving control commands and coupling control commands when the emergency stop signal is triggered, so that the vehicle enters a safe control state.

[0114] In other words, once an emergency stop signal is detected to be valid, the control switching unit 304 can prioritize cutting off the automatic output link, so that the vehicle will no longer continue to operate according to the automatic driving or docking instructions of the unmanned driving controller 300, thereby ensuring safety in emergency situations.

[0115] In one embodiment, the control switching unit 304 is connected to the discrete signal acquisition unit 302 and is configured to switch to manual control state when the detection of a braking signal, handbrake signal and / or forward / reverse operation signal meets the conditions for manual control. At the same time, it prohibits or blocks the automatic output of vehicle driving control commands by the unmanned driving controller 300 through the dual CAN access unit 301 and maintains the current state of the attached actuator 102b.

[0116] Maintaining the current state of the coupling actuator 102b means that, in the manual takeover state, no new automatic coupling control commands or automatic uncoupling control commands will be sent to the coupling actuator 102b. Instead, the existing locked or unlocked state of the coupling actuator 102b at the moment of switching will be maintained to prevent the coupling mechanism from continuing to act automatically during the manual takeover process.

[0117] In one implementation, the autonomous driving controller 300 can perform pedal status diagnosis based on the accelerator pedal feedback signal acquired by the discrete signal acquisition unit 302; at the same time, it can perform abnormal diagnosis of the coupling action based on the control signal and / or status feedback signal of the coupling actuator 102b acquired by the execution access unit 303.

[0118] For example, when an abnormal accelerator pedal feedback is detected, it can be determined that there is an abnormality in the vehicle driving control link; when it is detected that the coupling actuator 102b fails to output locking feedback after receiving the locking control command for more than a preset time, or fails to output unlocking feedback after receiving the unlocking control command for more than a preset time, it can be determined that the coupling action is abnormal.

[0119] When an anomaly is diagnosed, the autonomous driving controller 300 outputs an anomaly signal to the control switching unit 304 to trigger a manual takeover state and / or a safety control state. Therefore, this embodiment can not only perform safety control on the vehicle driving control link, but also establish separate safety control logic for the engagement action.

[0120] During use, when the vehicle is in a condition that allows autonomous driving, the control switching unit 304 first determines whether the current automatic control state is met. If it is, the unmanned driving controller 300 sends vehicle driving control commands to the original vehicle's CAN1 bus and / or CAN2 bus via the dual CAN access unit 301 according to the upper-level decision planning results, so as to control the relevant control nodes of the original vehicle to perform driving actions. At the same time, when it is necessary to perform the coupling operation, the coupling control command is output to the coupling execution mechanism 102b via the execution access unit 303.

[0121] Meanwhile, the discrete signal acquisition unit 302 continuously acquires ON signal, ST signal, handbrake signal, braking signal, emergency stop signal, forward / reverse operation signal, and accelerator pedal feedback signal to monitor the vehicle status and manual operation status in real time; the execution access unit 303 continuously acquires the control signal and / or status feedback signal of the coupling actuator 102b to monitor the coupling action status in real time.

[0122] If the control switching unit 304 detects that signals such as braking, handbrake, forward / reverse operation meet the conditions for manual takeover, it switches to the manual takeover state, prohibits or blocks the autonomous driving controller 300 from continuing to automatically output vehicle driving control commands, and maintains the current state of the coupling actuator 102b; if an emergency stop signal is detected, or the autonomous driving controller 300 diagnoses an abnormal pedal state or coupling action, the control switching unit 304 triggers the safety control state, prohibits or blocks the automatic output of vehicle driving control commands and / or coupling control commands, so that the vehicle enters the safety control state.

[0123] It is important to note that the constructions and arrangements of this application shown in several different exemplary embodiments are merely illustrative. Although only a few embodiments are described in detail in this disclosure, those who consult this disclosure will readily understand that many modifications are possible (e.g., changes in the size, dimensions, structure, shape and proportion of various elements, as well as parameter values ​​(e.g., temperature, pressure, etc.), mounting arrangements, use of materials, color, orientation, etc.) without substantially departing from the novel teachings and advantages of the subject matter described in this application). For example, an element shown as integrally formed may be composed of multiple parts or elements, the position of elements may be inverted or otherwise changed, and the nature or number or position of discrete elements may be altered or changed. Therefore, all such modifications are intended to be included within the scope of the invention. The order or sequence of any process or method steps may be changed or rearranged according to alternative embodiments. Therefore, the invention is not limited to the particular embodiments but extends to a variety of modifications that still fall within the scope of the appended claims.

[0124] Furthermore, in order to provide a concise description of exemplary embodiments, not all features of actual embodiments (i.e., those features that are not relevant to the currently considered best mode for carrying out the invention, or those features that are not relevant to implementing the invention) may be omitted.

[0125] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. An unmanned driving system for a baggage towing vehicle, characterized in that: Includes a decision-making unit (100), said decision-making unit (100) comprising, The decision planning module (101) is used to perform global path planning and local behavior decision-making based on task instructions, environmental perception data, positioning information and traction status information, and generate a comprehensive control sequence, which includes vehicle driving control instructions and coupling control instructions. The decision planning module (101) includes, The input interface (101a) is used to receive task instructions, environmental perception data, positioning information and traction status information; the output interface (101b) is used to output the integrated control sequence to the execution side (102), the execution side (102) includes the vehicle execution mechanism (102a) and the coupling execution mechanism (102b). The decision planning module (101) is configured to generate a global path or local trajectory according to the first path planning constraint when the vehicle is in a non-traction state; and to generate a global path or local trajectory according to the second path planning constraint when the vehicle is in a traction state. The second path planning constraint includes increasing the turning radius threshold relative to the first path planning constraint and applying a curvature change continuity constraint to the local trajectory; The decision-making unit (100) also includes a communication module (103) for receiving the task instructions issued by the cloud scheduling center (200); The local task management module (104) is used to cache the task instructions and provide the cached task instructions to the input interface (101a) when the communication module (103) detects a communication interruption, so as to continue to drive the decision planning module (101) to generate the integrated control sequence; The decision unit (100) further includes an environmental perception module (105), which is used to identify ground markings in the work area and output ground marking information to the input interface (101a). The decision planning module (101) generates guidance and control content for vehicles approaching the target area based on the ground marking information; The decision unit (100) also includes a positioning and mapping module (106) for outputting positioning information based on high-precision maps and differential positioning information; The decision planning module (101) performs preliminary docking control based on the positioning information and fine-tunes the docking position based on the ground marking information to generate the comprehensive control sequence containing position fine-tuning control instructions.

2. The unmanned driving system for baggage towing vehicles according to claim 1, characterized in that: The traction status information is determined based on the status feedback signal of the coupling actuator (102b), the coupling completion signal, and / or the existence information of the towed object.

3. An unmanned driving control method for a baggage towing vehicle, applicable to the unmanned driving system for a baggage towing vehicle as described in claim 1 or 2, characterized in that: include, Receive mission instructions, environmental perception data, positioning information, and traction status information; Global path planning and local behavior decisions are performed based on the task instructions, the environmental perception data, the positioning information, and the traction status information. When the vehicle is in a non-traction state, a global path or local trajectory is generated according to the first path planning constraint; when the vehicle is in a traction state, a global path or local trajectory is generated according to the second path planning constraint. Generate a comprehensive control sequence, which includes vehicle driving control commands and coupling control commands; The integrated control sequence is output to the execution side (102) to control the vehicle actuator (102a) and the coupling actuator (102b) to perform corresponding actions.

4. The unmanned driving control method for a baggage towing vehicle according to claim 3, characterized in that: The receiving task instruction includes receiving task instructions issued by the cloud scheduling center (200); and reading pre-cached task instructions and continuing to execute the corresponding task path when communication is interrupted.

5. The unmanned driving control method for a baggage towing vehicle according to claim 4, characterized in that: The traction status information is determined based on the status feedback signal of the coupling actuator (102b), the coupling completion signal, and / or the existence information of the traction object; the second path planning constraint includes increasing the turning radius threshold relative to the first path planning constraint and / or applying curvature change continuity constraint to the local trajectory.

6. The unmanned driving control method for a baggage towing vehicle according to claim 4 or 5, characterized in that: The environmental perception data includes ground marking information; It also includes preliminary parking control based on the positioning information, and fine-tuning the vehicle parking process based on the ground marking information, so that the vehicle can be parked at the target parking position or docking area.

7. The unmanned driving control method for a baggage towing vehicle according to claim 6, characterized in that: The integrated control sequence includes outputting a coupling control command when the vehicle reaches the preset coupling position and outputting a disengagement control command when the vehicle reaches the preset uncoupling position.

8. An unmanned access device for a baggage towing vehicle, applicable to the unmanned driving system for a baggage towing vehicle as described in claim 1 or 2, characterized in that: include, Unmanned driving controller (300); The dual CAN access unit (301) has a first CAN interface (301a) and a second CAN interface (301b). The first CAN interface (301a) is used to electrically connect to CAN1-H and CAN1-L in the vehicle electrical system, and the second CAN interface (301b) is used to electrically connect to CAN2-H and CAN2-L in the vehicle electrical system. The discrete signal acquisition unit (302) is used to be electrically connected to the discrete signal terminal of the vehicle electrical system and to acquire vehicle status signals and manual operation signals. The vehicle status signals and manual operation signals include at least ON signal, ST signal, handbrake signal, braking signal, emergency stop signal and forward / reverse operation signal. An execution access unit (303) is used to electrically connect to the execution component of the attached execution mechanism (102b); The control switching unit (304) is connected to the dual CAN access unit (301), the discrete signal acquisition unit (302), and the execution access unit (303); The unmanned driving controller (300) is used to output vehicle driving control commands via the dual CAN access unit (301) and to output hook-up control commands via the execution access unit (303) under the automatic control state permitted by the control switching unit (304); the control switching unit (304) is used to prohibit or block the automatic output of the vehicle driving control commands and / or the hook-up control commands when a manual takeover signal and / or an abnormal signal is detected.

9. The unmanned access device for a baggage towing vehicle according to claim 8, characterized in that: The control switching unit (304) is electrically connected to the emergency stop signal and is used to prohibit or block the autonomous driving controller (300) from continuing to automatically output the vehicle driving control command and / or the hook-up control command when the emergency stop signal is triggered, so that the vehicle enters a safe control state.

10. The unmanned access device for a baggage towing vehicle according to claim 8 or 9, characterized in that: The dual CAN access unit (301) is used to send the vehicle driving control command to the control node related to vehicle driving control via the CAN1 bus and / or CAN2 bus.

11. The unmanned access device for a baggage towing vehicle according to claim 8 or 9, characterized in that: The control switching unit (304) is connected to the discrete signal acquisition unit (302) and is configured to switch to manual control state when the braking signal, the handbrake signal and / or the forward / reverse operation signal meet the conditions for manual control. At the same time, it prohibits or blocks the automatic output of the vehicle driving control command by the unmanned driving controller (300) through the dual CAN access unit (301) and maintains the current state of the attached actuator (102b).

12. The unmanned access device for a baggage towing vehicle according to claim 11, characterized in that: The discrete signal acquisition unit (302) is also used to acquire feedback signals related to the accelerator pedal; the execution access unit (303) is also used to acquire control signals and / or status feedback signals of the coupling actuator (102b); the unmanned driving controller (300) is used to perform pedal status diagnosis based on the feedback signals, and to perform coupling action abnormality diagnosis based on the control signals and / or status feedback signals, and to output an abnormal signal to the control switching unit (304) when an abnormality is diagnosed, so as to trigger manual takeover state and / or safety control state.

Images