Perception and safety protection system and method for obstacles above unmanned vehicle at port

Abstract

The invention discloses a sensing and safety protection system and method for an obstacle above a port unmanned vehicle, which integrates a sensing module of a laser radar to form an obstacle recognition and response capability independent of a quay crane task signal, so that the vehicle can autonomously sense the upper space and independently judge the height risk of the obstacle; whether emergency stop or speed limit is started or not is decided in real time through the task management module, the reliability and response timeliness of the system are remarkably improved, the setting of a dynamic height safety threshold value improves the action efficiency on the premise of guaranteeing the safety, the system synchronously uploads a judgment result to the FMS module, an operator can conveniently check and recognize the risk state in real time, and the safety of the system is improved. And in the case of misjudgment, a'manual release 'operation is initiated through a front-end display screen, so that double security assurance of'local instant response + cloud manual bottom taking' is formed, and guarantee is provided for operation flexibility under extreme working conditions.

Description

Technical Field

[0001] This invention belongs to the field of intelligent driving technology and relates to obstacle perception and safety protection for unmanned vehicles. Specifically, it relates to an obstacle perception and safety protection system and method for unmanned vehicles in ports. Background Technology

[0002] Currently, port autonomous driving systems primarily rely on two-dimensional spatial information for obstacle recognition and avoidance. This approach is applicable to most scenarios, processes relatively little data, and can effectively handle 90% of situations. Its basic assumption is that the vehicle will not encounter obstacles exceeding its own height during operation; therefore, the system focuses solely on detecting planar obstacles. However, in port operations, the lifting and unloading of spreader equipment typically occur above or in front of the vehicle, a characteristic that existing two-dimensional systems have not fully considered.

[0003] To ensure smooth vehicle passage, existing technologies typically filter out obstruction areas created by spreader cranes directly from two-dimensional obstacle detection. This approach ignores the potential threat to vehicle safety posed by spreader cranes, fails to provide effective real-time monitoring, and consequently lacks identification of overhead obstacles. In existing measures to prevent pull-off accidents, the transport vehicle determines whether it can start based on task signals, such as checking whether the vehicle has completed loading / unloading tasks and whether the spreader crane height exceeds the vehicle's safe height. However, this judgment is instantaneous, considering only the spreader crane height at the moment of vehicle start-up, failing to account for potential height changes during subsequent operations. Furthermore, existing technologies rely on task confirmation signals with the quay crane for safety, but signal interpretation in terminals often presents problems, prone to misjudgments or signal delays, thus increasing the likelihood of pull-off accidents or other safety risks. Therefore, although the system can prevent vehicle start-up in certain situations, changes in spreader crane height can still lead to scraping or pull-off accidents in actual operation. A pull-off accident occurs when a transport vehicle (such as a container truck) starts moving while the spreader crane is not completely detached from the container, causing physical contact between the spreader crane and the container or transport equipment, resulting in structural damage.

[0004] The aforementioned limitations in continuously monitoring and identifying risks related to the height of spreader expose the inadequacy of existing technologies in fully ensuring the safe operation of vehicles in a port environment. Summary of the Invention

[0005] To address the aforementioned problems, the main objective of this invention is to design an obstacle perception and safety protection system and method for unmanned vehicles in ports, thereby solving the problem of safety accidents such as pulling and scraping caused by the inability of autonomous driving systems to effectively identify high-altitude obstacles such as lifting equipment.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: An obstacle perception and safety protection system for unmanned vehicles in ports includes: The perception module, consisting of LiDAR sensors deployed above the autonomous vehicle, is used to detect obstacles within a detection area centered on the vehicle and output the detection results. The task management module receives the detection results output by the perception module, uploads the detection results to the FMS module, and determines whether the vehicle is in a safe state based on the detection results. If the vehicle is determined to be in a safe state, no instructions are given and monitoring continues. If the vehicle is determined to be in a dangerous state, a red light stop instruction is triggered to control the vehicle's traffic. The FMS module is a cloud-based management platform for autonomous vehicles. It centrally displays the detection results of all vehicles and the decision results of the task management module, allowing operators to view and intervene in real time. The route planning module receives task information and red light traffic control instructions from the task management module, and executes immediate stopping or starting to restore the original planned route.

[0007] As a further description of the present invention, the detection area is defined as a rectangular three-dimensional enclosed space, including a vertical height range, a horizontal width range, and a horizontal length range; The vertical height range is configured between 0 and 10m. The horizontal width range is 1.35m to the left and right of the vehicle center; The horizontal length range is 7.5m on each side, front and rear of the vehicle center.

[0008] As a further description of the present invention, the detection results include function off, function abnormal, obstacle detected, and no obstacle in the detection area. When the detection result is that an obstacle is detected, the height information of the obstacle is output simultaneously.

[0009] As a further description of the present invention, the judgment result of the task management module includes a dangerous state and a safe state; wherein, the judgment condition for a dangerous state is that the detection result output by the perception module is that an obstacle is detected and the height of the obstacle is lower than the set height safety threshold, or the detection result is that the function is abnormal; the judgment condition for a safe state is that the detection result output by the perception module is that an obstacle is detected and the height of the obstacle is higher than the set height safety threshold, or the detection result is that there is no obstacle in the detection area, or the function is turned off. The traffic control instruction response mechanism is as follows: In a dangerous situation, a red light instruction to prohibit vehicles from starting or an immediate stop instruction will be issued; in a safe situation, no red light will be issued, or the red light will be canceled and an instruction to allow vehicles to pass will be issued, or a stop restriction instruction will be lifted.

[0010] As a further description of the present invention, the dangerous state includes a speed limit control mechanism and an FMS command processing mechanism; Speed ​​limit control mechanism: If the detection result indicates that an obstacle has been detected and the height of the obstacle is lower than the set height safety threshold, the vehicle speed will be reduced to the set speed limit value until the height of the obstacle recovers to exceed the set height safety threshold, at which point the speed limit will be lifted; FMS command processing mechanism: If the operator manually confirms that the vehicle is passable based on the information from the FMS module, the FMS module will issue a leaveFlag command to the task management module. If the result is safe in the next ten consecutive frames, or if the vehicle's driving height exceeds the set height safety threshold, the leaveFlag command will be automatically canceled.

[0011] As a further description of the present invention, the height safety threshold is dynamically calculated and set according to the actual height of the autonomous vehicle, and the expression is: height safety threshold = vehicle height + safety buffer zone.

[0012] As a further description of the invention, the FMS module displays the current altitude protection function status on the vehicle's display screen: A gray light indicates that the function is not enabled or the current status cannot be determined. The red light indicates that the height requirement is not met, posing a risk. The yellow light indicates that the height requirement is not met, but it has been manually confirmed that passage is permitted. Operators can manually cancel the dangerous status by clicking on the red light, which triggers the task management module to issue the leaveFlag command, thus lifting the traffic restriction.

[0013] As a further description of the present invention, the route planning module performs the following actions based on the traffic control instruction response mechanism, speed limit control mechanism, and FMS instruction processing mechanism of the task management module: Upon receiving an immediate stop command, the vehicle will stop immediately. Once the task management module removes the traffic command restriction, the original planned route will resume execution.

[0014] A method for obstacle perception and safety protection above unmanned vehicles in ports, based on the above-mentioned system, includes the following steps: S1: When the vehicle enters the high-level guard state, the perception module detects obstacle information above the vehicle based on the lidar sensor and outputs the detection results in real time; S2: The task management module receives the perceived obstacle information and detection results, uploads them to the FMS module for display on the vehicle's front end, and executes the following mechanisms: If the height of the obstacle is lower than the preset height safety threshold, the speed limit will be activated to reduce the vehicle speed to the speed limit value. If the height of the obstacle returns to above the preset height safety threshold, the speed limit will be lifted immediately. The vehicle's status is determined based on the criteria for determining a dangerous state and a safe state. If the state is dangerous, a red light prohibition command or an immediate stop command is issued. If the state is safe, no intervention is taken, or the previously issued red light permission command is canceled, or the parking restriction command is lifted. When the operator clicks the red light on the FMS module display to confirm that passage is permitted, the traffic control order is lifted by receiving the leaveFlag command issued by the FMS module and switching to the yellow light state. S3: The task management module determines whether the vehicle is in a dangerous state. If it is in a dangerous state, it enters the leaveFlag logic judgment. If it is in a safe state or the leaveFlag logic judgment is that it can pass, it determines whether the FMS module has issued an immediate stop command. If the FMS module has already issued an immediate stop command or the leaveFlag logic determines that the road is impassable, then an immediate stop command is sent to the route planning module to execute the stop operation. If the FMS module does not issue an immediate stop command, the immediate stop command will be canceled, the route planning module will resume the execution of the original planned route, and the cancellation of the stop and the route restoration operation will be uploaded to the FMS module simultaneously.

[0015] As a further description of the present invention, the leaveFlag logic judgment condition is that after the operators confirm that it is passable through the FMS module, the task management module issues the leaveFlag command; The leaveFlag command is canceled when the result of ten consecutive frames is a safe state, or when the vehicle's driving height exceeds the set height safety threshold. The leaveFlag command is triggered by the operator clicking the red light on the FMS module display screen. The yellow light indicates that the command is in effect.

[0016] Compared with the prior art, the technical advantages of the present invention are as follows: This invention provides a system and method for obstacle perception and safety protection of unmanned vehicles in ports. Through the collaborative work of a perception module, a task management module, an FMS module, and a path planning module, it achieves real-time detection, dynamic risk assessment, and traffic control response of obstacles above the vehicle. The invention integrates a LiDAR perception module, forming an obstacle recognition and response capability independent of the quay crane task signals. The vehicle can autonomously perceive the space above, independently judge the height risk of obstacles, and make real-time decisions on whether to initiate emergency braking or speed limits through the task management module, significantly improving the system's reliability and response time. The dynamic height safety threshold setting improves efficiency while ensuring safety. The system synchronously uploads the judgment results to the FMS module, allowing operators to view and identify risk status in real time. In case of misjudgment, a "manual clearance" operation can be initiated through the front-end display screen, forming a dual safety guarantee of "local instant response + cloud-based manual backup," providing assurance for operational flexibility under extreme conditions. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the sensing and security protection system method of the present invention. Detailed Implementation

[0018] The present invention will now be described in detail with reference to the accompanying drawings: In one embodiment of the present invention, a port unmanned vehicle obstacle perception and safety protection system is disclosed. This system, through the collaborative work of a perception module, a task management module, an FMS module, and a path planning module, achieves real-time detection of obstacles above the vehicle, dynamic risk assessment, and traffic control response. Specifically, this embodiment provides a detailed description of the collaboration of the above four modules.

[0019] I. Perception Module The perception module is deployed above the autonomous vehicle. Its core sensor is LiDAR (Light Detection and Ranging), which is used to detect obstacles in the detection area centered on the vehicle and output the detection results.

[0020] In this embodiment, the detection area is defined as a rectangular three-dimensional enclosed space, including a vertical height range, a horizontal width range, and a horizontal length range. The vertical height range is configured between 0 and 10m; the horizontal width range is 1.35m to the left and right of the vehicle center; and the horizontal length range is 7.5m to the front and rear of the vehicle center.

[0021] The system dynamically sets a reasonable height safety threshold, such as 450-480cm, based on the detection range and stability of the vehicle's sensors. To ensure data stability, multi-frame filtering and trend smoothing are performed on the sensor output data to avoid misjudgments caused by data jumps.

[0022] In this embodiment, the detection results include function off, function malfunction, obstacle detected, and no obstacle in the detection area. When the detection result is that an obstacle is detected, the height information of the obstacle is output synchronously. The specific output states are as follows: HangerDetection = DEFAULT: Functionality disabled; HangerDetection = FAILED: Functional malfunction; HangerDetection = DETECTING&&hanger_height_cm = X: An obstacle has been detected, and the height of the bottom of the obstacle is X centimeters. HangerDetection = READY_TO_MOVE: The detection area is free of obstacles.

[0023] II. Task Management Module The task management module is responsible for judging the detection results of the perception module and issuing traffic control instructions. The processing logic of traffic control instructions includes the traffic control instruction response mechanism, the speed limit control mechanism, and the FMS instruction processing mechanism.

[0024] Specifically, in this embodiment, the task management module receives the detection results output by the perception module, uploads the detection results to the FMS module, and simultaneously determines whether the vehicle is in a safe state based on the detection results. If the vehicle is determined to be in a safe state, no instructions are given and monitoring continues. If the vehicle is determined to be in a dangerous state, a red light stop instruction is triggered to control the vehicle's traffic. That is, the determination results of the task management module include both dangerous and safe states.

[0025] Criteria for determining a dangerous state: The detection result output by the perception module is that an obstacle has been detected (DETECTING), and the height of the obstacle (hanger_height_cm) is lower than the set height safety threshold. Or the test result is a functional abnormality (FAILED); If any of the above conditions are met, the system determines that there is a risk of an obstacle above and prohibits the vehicle from starting.

[0026] Criteria for determining a safe state: The detection result output by the perception module is that an obstacle has been detected (DETECTING), and the height of the obstacle (hanger_height_cm) is higher than the set safety threshold. Or the detection result is that there are no obstacles in the detection area (READY_TO_MOVE); Or the function is disabled (DEFAULT); If any of the above conditions are met, the system determines that there is no risk of obstacles above and allows passage.

[0027] It should be noted that the aforementioned height safety threshold is dynamically calculated and set based on the actual height of the autonomous vehicle. The expression is: Height safety threshold = Vehicle height + Safety buffer zone (usually 10-20cm). This height safety threshold is recorded in the task management module and adjusted according to the actual height of different vehicles.

[0028] In this embodiment, the traffic control command processing logic for autonomous vehicle control is as follows: The traffic control instruction response mechanism is as follows: Dangerous situation: Immediately issue a red light through the virtual traffic control interface, and issue a command to prohibit vehicles from starting or to stop immediately; Safety status: No red light is issued (no intervention is made) and vehicles are allowed to continue to follow the passage instruction, or the previously issued red light is cancelled to allow vehicles to pass, or the parking restriction instruction is lifted.

[0029] Dangerous situations also include speed limit control mechanisms and FMS command processing mechanisms; Speed ​​limit control mechanism: If the detection result indicates that an obstacle has been detected and the height of the obstacle is lower than the set height safety threshold, the vehicle speed will be reduced to the set speed limit (e.g., 10 km / h) until the height of the obstacle returns to above the set height safety threshold, at which point the speed limit will be lifted. FMS instruction processing mechanism (Leave): If the operator manually confirms that the passage is safe based on the information from the FMS module, the FMS module will issue a leaveFlag command to the task management module. If the result is safe in the next ten consecutive frames, or if the vehicle exceeds the set height safety threshold, the leaveFlag command will be automatically canceled.

[0030] III. FMS Module The FMS module enables visualization and manual deactivation of the system. As a cloud-based management platform, the FMS (Fleet Management System) collects and displays the perception and control status of all vehicles, allowing operators to view and intervene in real time. The system displays the current high-level protection status on each vehicle's display screen (traffic control section). Gray light: Function not enabled or current status cannot be determined; Red light: Height does not meet requirements, posing a risk; Yellow light: Height does not meet requirements, but passage has been manually confirmed. Operators can initiate a manual "cancel dangerous status" operation by clicking on the red light, which triggers the task management module to issue the leaveFlag command, thus lifting the restrictions of highly relevant traffic orders.

[0031] IV. Path Planning Module The route planning module is responsible for executing actions based on the task information and traffic control instructions issued by the task management module, including immediately stopping or starting to restore the original planned route. Specifically, based on the traffic control instruction response mechanism, speed limit control mechanism, and FMS instruction processing mechanism of the task management module, it executes the following actions: Upon receiving an immediate stop command (stop = EMERGENCY), the vehicle will stop immediately (non-periodic response, triggered processing). Once the traffic restrictions are lifted by the task management module, the original planned route will resume execution.

[0032] It should also be noted that this embodiment is not limited to obtaining obstacle information through a sensing module, but also includes a remote height determination method where the FMS module receives signals from the quay crane's ECS via the cloud. In this mode, the FMS module accesses and subscribes in real time to the spreader height signals issued by the port quay crane's Equipment Control System (ECS) or PLC (Programmable Logic Controller). The FMS module continuously monitors the height data of each working spreader in the cloud and makes real-time judgments against preset height safety thresholds. Once a spreader is detected to be below the set height safety threshold, it is considered that there is a risk above, and the FMS issues an emergency stop or speed limit command to the target vehicle.

[0033] The specific implementation path is as follows: Spreader height acquisition (ECS) → Cloud-based FMS module receives and determines the data → Traffic control instructions are sent to the vehicle-side task management module → Vehicles execute parking / speed limit commands.

[0034] This approach can acquire height information without relying entirely on the perception algorithms of vehicle LiDAR, and it has good deployment feasibility in terminal projects with close docking and high integration of ECS and FMS.

[0035] In another embodiment of the present invention, a method for obstacle perception and safety protection above unmanned vehicles in ports is disclosed, such as... Figure 1 As shown, this method, based on the aforementioned system, includes the following steps: S1: When the vehicle enters the high-level guard state, the perception module detects obstacle information above the vehicle based on the lidar sensor and outputs the detection results in real time; S2: The task management module receives the perceived obstacle information and detection results, uploads them to the FMS module for display on the vehicle's front end, and executes the following mechanisms: If the height of the obstacle is lower than the preset height safety threshold, the speed limit will be activated to reduce the vehicle speed to the speed limit value. If the height of the obstacle returns to above the preset height safety threshold, the speed limit will be lifted immediately. The vehicle's status is determined based on the criteria for determining a dangerous state and the criteria for determining a safe state. If it is a dangerous state, a red light prohibition command or an immediate stop command is issued. If it is a safe state, no intervention is made or the issued red light permission command is canceled, or the stop restriction command is lifted to allow the vehicle to start. When the operator clicks the red light on the FMS module display to confirm that passage is permitted, the traffic control order is lifted by receiving the leaveFlag command issued by the FMS module and switching to the yellow light state. S3: The task management module determines whether the vehicle is in a dangerous state. If it is in a dangerous state, it enters the leaveFlag logic judgment. If it is in a safe state or the leaveFlag logic judgment is that it can pass, it determines whether the FMS module has issued an immediate stop command. If the FMS module has already issued an immediate stop command or the leaveFlag logic determines that the road is impassable, then an immediate stop command is issued to the route planning module to execute the stop operation; If the FMS module does not issue an immediate stop command, the immediate stop command will be canceled, the route planning module will resume the execution of the original planned route, and the cancellation of the stop and the route restoration operation will be uploaded to the FMS module simultaneously.

[0036] Specifically, in this embodiment, the determination conditions for the vehicle's driving status in step S3 include two types: vehicle status and obstacle situation. If either condition is met, the vehicle is determined to be in a safe state. Vehicle status: The vehicle is not within the precise stopping period of the spreader; Obstacle status: The height of the obstacle is greater than the set height safety threshold, or there is no obstacle above.

[0037] The logic condition for leaveFlag is: after the operations personnel confirm that the vehicle is ready to run through the FMS module, the task management module issues the leaveFlag command; The leaveFlag command is canceled when: the result of ten consecutive frames is a safe state, or the vehicle's driving height exceeds the set height safety threshold. The leaveFlag command is triggered by the operator clicking the red light on the FMS module display screen. The yellow light indicates that the command is in effect.

[0038] The above embodiments disclose the technical solution of the present invention, proposing a height protection system and method for obstacles above automated guided vehicles (AIVs) in ports, which has real-time perception capabilities completely independent of traditional quay crane linkage signal mechanisms. Compared with the existing technology that relies on instantaneous judgment based on two-dimensional information and operation signals, this system, starting from the vehicle end, realizes a full-link self-perception, self-decision-making, and self-response mechanism. The specific improvements and innovations are as follows: 1. The "height protection" function based on vehicle-side perception enables independent real-time assessment of spreader risks; This invention independently integrates a perception module, a task management module, an FMS module, and a path planning module on the intelligent driving vehicle, forming a "high-altitude protection" system that completely eliminates reliance on quay crane operation signals or the "loading / unloading complete" feedback from traditional ECS systems. The vehicle can autonomously perceive the overhead space, independently assess the height and risk of obstacles, and make real-time decisions on whether to initiate emergency braking or speed limiting. This innovation breaks away from existing technologies that rely on instantaneous signals upon task completion to determine safety status, significantly improving system reliability and response time.

[0039] 2. Integrate a dynamic height safety threshold determination mechanism based on the vehicle's own cargo box status to improve system flexibility and adaptability; The height protection function of this invention does not use a fixed threshold when assessing safety risks. Instead, it dynamically configures the height determination based on factors such as the status of the container carried by the vehicle (e.g., empty / loaded) and the vehicle type. This dynamic threshold mechanism has stronger adaptability to traffic flow compared to fixed logic, and can improve efficiency while ensuring safety.

[0040] 3. The vehicle-side link design achieves millisecond-level low-latency response and reserves an entry point for manual intervention; The judgment process of this invention is completed locally on the vehicle. After the perception module outputs its data, it is directly transmitted to the task management module, and then quickly distributed to the path planning module, without relying on cloud relay. This method ensures that the system can respond within milliseconds when a dangerous situation is identified. At the same time, the system synchronously uploads the judgment result to the FMS module, which allows operators to view and identify risk status in real time. In case of misjudgment, a "manual clearance" operation (Leave command) can be initiated through the front-end display screen, forming a dual safety guarantee of "local instant response + cloud-based manual backup".

[0041] 4. Standardized output of high-level protection results and linkage with multi-functional modules form a closed-loop control; The system of this invention can output height protection results through standardized signal fields (such as the HangerDetection status field and the hanger_height_cm value), which can be recognized and processed by both the task management module and the FMS module front end. The task management module can initiate traffic commands (stopping or speed limit), the route planning module can respond in real time, and the FMS module realizes status visualization and manual intervention, making the vehicle's height protection a "bottom-line mechanism" for overhead safety perception, and working closely with the entire AIV intelligent driving system.

[0042] Compared with the traditional approach of current port automatic driving systems, which mainly rely on two-dimensional obstacle handling and PLC-related signal judgment for quay crane operations, this invention achieves breakthrough improvements in several aspects. Specifically, this invention has the following advantages: 1. This invention integrates a LiDAR perception module on the AIV vehicle side, forming an obstacle recognition and response capability independent of the quay crane mission signal, improving safety and reliability. The perception module can scan the space directly above the vehicle in real time, detect the height of the bottom surface of obstacles, and output structured status fields, breaking away from the limitations of traditional reliance on "ECS signal + instantaneous status judgment", solving the response lag problem caused by "signal delay and misjudgment" in high-risk times such as pulling, closing, and scraping, and realizing safety judgment based on real-scene perception; 2. In this invention, the judgment and control links of the perception → task management → planning modules are all completed locally on the vehicle. A trigger-based real-time processing mechanism is used to replace the periodic polling processing. The control link avoids relying on cloud relay, realizing a real-time response mechanism of the local closed-loop link on the vehicle. This significantly reduces decision delay and achieves millisecond-level stopping or speed limiting response after a dangerous state occurs. Compared with traditional solutions based on cloud logic and ECS feedback, it has faster and more stable decision execution capabilities. 3. The system of the present invention supports dynamically setting safety thresholds based on vehicle type (such as 3.5 generation, 5th generation) and current status of the carrying box (empty box / loaded box), flexibly determining passage safety, and calling them uniformly in the perception module and task management module, avoiding false blocking and false release caused by "fixed threshold judgment", and improving the system's adaptability and accuracy in different work tasks and equipment states. 4. This invention synchronizes all high-altitude protection results to the FMS module in real time and displays them on the front-end interface. It supports remote viewing by operators and allows them to manually trigger "cancel danger judgment" when necessary to lift traffic restrictions / parking control through the task management module. It has a "manual backup" mechanism, which takes into account both misjudgment handling and operational efficiency, forming a dual-channel control mechanism of "automatic judgment + manual review". While ensuring the bottom line of safety, it provides a guarantee for operational flexibility under extreme conditions. 5. Each module of this invention can use standardized communication protocols and fields, supports horizontal integration into the existing AIV system architecture, does not rely on dedicated quay crane docking protocols, has high compatibility with existing intelligent driving systems, is easy to deploy, facilitates later promotion and deployment, is adaptable to multiple vehicle types, multiple terminals, and multiple system environments, and has good versatility and scalability.

[0043] The above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Any other modifications or equivalent substitutions made by those skilled in the art to the technical solutions of the present invention, as long as they do not depart from the spirit and scope of the technical solutions of the present invention, should be covered within the scope of the claims of the present invention.

Claims

1. A port unmanned vehicle obstacle perception and safety protection system, characterized in that, include: The perception module, consisting of LiDAR sensors deployed above the autonomous vehicle, is used to detect obstacles within a detection area centered on the vehicle and output the detection results. The task management module is used to receive the detection results output by the perception module, upload the detection results to the FMS module, and determine whether the vehicle is in a safe state based on the detection results. If the vehicle is determined to be in a safe state, no intervention will be given and monitoring will continue. If the vehicle is determined to be in a dangerous state, a red light stop instruction will be triggered to control the vehicle's traffic. The FMS module is a cloud-based management platform for autonomous vehicles. It centrally displays the detection results of all vehicles and the decision results of the task management module, allowing operators to view and intervene in real time. The route planning module receives task information and red light traffic control instructions from the task management module, and executes immediate stopping or starting to restore the original planned route.

2. The obstacle perception and safety protection system for unmanned vehicles in ports according to claim 1, characterized in that: The detection area is defined as a rectangular three-dimensional enclosed space, including the vertical height range, the horizontal width range, and the horizontal length range; The vertical height range is configured between 0 and 10m. The horizontal width range is 1.35m to the left and right of the vehicle center; The horizontal length range is 7.5m on each side, front and rear of the vehicle center.

3. The obstacle perception and safety protection system for unmanned vehicles in ports according to claim 1, characterized in that: The detection results include function off, function abnormal, obstacle detected, and no obstacle in the detection area. When the detection result is that an obstacle is detected, the height information of the obstacle is output simultaneously.

4. The obstacle perception and safety protection system for unmanned vehicles in ports according to claim 3, characterized in that: The judgment results of the task management module include dangerous and safe states. The judgment conditions for dangerous state are that the detection result output by the perception module is that an obstacle is detected and the height of the obstacle is lower than the set height safety threshold, or the detection result is that the function is abnormal. The judgment conditions for safe state are that the detection result output by the perception module is that an obstacle is detected and the height of the obstacle is higher than the set height safety threshold, or the detection result is that there is no obstacle in the detection area, or the function is turned off. The traffic control instruction response mechanism is as follows: In a dangerous situation, a red light instruction to prohibit vehicles from starting or an immediate stop instruction will be issued; in a safe situation, no red light will be issued, or the red light will be canceled and an instruction to allow vehicles to pass will be issued, or a stop restriction instruction will be lifted.

5. The obstacle perception and safety protection system for unmanned vehicles in ports according to claim 4, characterized in that: Dangerous situations include speed limit control mechanisms and FMS command processing mechanisms; Speed ​​limit control mechanism: If the detection result indicates that an obstacle has been detected and the height of the obstacle is lower than the set height safety threshold, the vehicle speed will be reduced to the set speed limit value until the height of the obstacle recovers to exceed the set height safety threshold, at which point the speed limit will be lifted; FMS command processing mechanism: If the operator manually confirms that the vehicle is passable based on the information from the FMS module, the FMS module will issue a leaveFlag command to the task management module. If the result is safe in the next ten consecutive frames, or if the vehicle's driving height exceeds the set height safety threshold, the leaveFlag command will be automatically canceled.

6. A port unmanned vehicle obstacle perception and safety protection system according to claim 4 or 5, characterized in that: The height safety threshold is dynamically calculated and set based on the actual height of the autonomous vehicle, and the expression is: Height safety threshold = Vehicle height + Safety buffer zone.

7. The obstacle perception and safety protection system for unmanned vehicles in ports according to claim 5, characterized in that: The FMS module displays the current altitude protection function status on the vehicle's display screen: A gray light indicates that the function is not enabled or the current status cannot be determined. The red light indicates that the height requirement is not met, posing a risk. The yellow light indicates that the height requirement is not met, but it has been manually confirmed that passage is permitted. Operators can manually cancel the dangerous status by clicking on the red light, which triggers the task management module to issue the leaveFlag command, thus lifting the traffic restriction.

8. The obstacle perception and safety protection system for unmanned vehicles in ports according to claim 1, characterized in that: Based on the traffic control command response mechanism, speed limit control mechanism, and FMS command processing mechanism of the task management module, the route planning module performs the following actions: Upon receiving an immediate stop command, the vehicle will stop immediately. Once the task management module removes the traffic command restriction, the original planned route will resume execution.

9. A method for obstacle perception and safety protection above unmanned vehicles in ports, characterized in that, Based on the system according to any one of claims 1-8, the method comprises the following steps: S1: When the vehicle enters the high-level guard state, the perception module detects obstacle information above the vehicle based on the lidar sensor and outputs the detection results in real time; S2: The task management module receives the perceived obstacle information and detection results, uploads them to the FMS module for display on the vehicle's front end, and executes the following mechanisms: If the height of the obstacle is lower than the preset height safety threshold, the speed limit will be activated to reduce the vehicle speed to the speed limit value. If the height of the obstacle returns to above the preset height safety threshold, the speed limit will be lifted immediately. The vehicle's status is determined based on the criteria for determining a dangerous state and a safe state. If the state is dangerous, a red light prohibition command or an immediate stop command is issued. If the state is safe, no intervention is taken, or the previously issued red light permission command is canceled, or the parking restriction command is lifted. When the operator clicks the red light on the FMS module display to confirm that passage is permitted, the traffic control order is lifted by receiving the leaveFlag command issued by the FMS module and switching to the yellow light state. S3: The task management module determines whether the vehicle is in a dangerous state. If it is in a dangerous state, it enters the leaveFlag logic judgment. If it is in a safe state or the leaveFlag logic judgment is that it can pass, it determines whether the FMS module has issued an immediate stop command. If the FMS module has already issued an immediate stop command or the leaveFlag logic determines that the road is impassable, then an immediate stop command is sent to the route planning module to execute the stop operation. If the FMS module does not issue an immediate stop command, the immediate stop command will be canceled, the route planning module will resume the execution of the original planned route, and the cancellation of the stop and the route restoration operation will be uploaded to the FMS module simultaneously.

10. A method for obstacle perception and safety protection above unmanned vehicles in ports according to claim 9, characterized in that: The leaveFlag logic condition is that after the operations personnel confirm through the FMS module that passage is possible, the task management module issues the leaveFlag command; The leaveFlag command is canceled when the result of ten consecutive frames is a safe state, or when the vehicle's driving height exceeds the set height safety threshold. The leaveFlag command is triggered by the operator clicking the red light on the FMS module display screen. The yellow light indicates that the command is in effect.

Images