Campus network connection unmanned vehicle dynamic scheduling planning method and system

Abstract

The application relates to the technical field of unmanned vehicle scheduling, in particular to a park networked unmanned vehicle dynamic scheduling planning method and system, which comprises the following steps: collecting dynamic obstacle data through a multi-source sensor and a V2X device to generate a four-dimensional passability graph, formulating avoidance strategy weights and optimizing multi-vehicle trajectories, combining edge nodes and cloud adjustment paths, and evaluating threat values to trigger a resident area reconstruction. The system comprises a perception module, a decision module, a storage module, a distribution module, a regulation and control module and a verification module, and an abnormality processing module is linked with a threat value change rate to optimize resource distribution. The application can dynamically adjust the avoidance strategy according to the threat value, improve scheduling accuracy and resource distribution accuracy, guarantee vehicle safety and reduce the task failure rate.

Description

Technical Field

[0001] This invention relates to the field of intelligent transportation and autonomous driving technology, and in particular to a dynamic scheduling and planning method and system for networked unmanned vehicles in industrial parks. Background Technology

[0002] Currently, the networked unmanned vehicle dispatching system in industrial parks generally relies on preset path planning and static parking area allocation strategies. While this improves vehicle operating efficiency to some extent, it can easily lead to path planning instability under the combined interference of dynamic obstacles and multi-vehicle trajectory conflicts. To improve the stability and adaptability of the dispatching system, corresponding measures need to be taken to address issues such as sudden obstacles intruding into preset parking areas, multiple vehicles simultaneously requesting parking leading to insufficient capacity, and conflicts between parking areas and the movement trajectories of dynamic obstacles.

[0003] Therefore, a dynamic scheduling method based on spatiotemporal grid maps can be used to optimize the operation of autonomous vehicles. When a sudden obstacle is detected intruding into the parking area, the path and parking area allocation are adjusted in real time to maintain system stability. This is generally handled by a distributed decision-making unit. However, the intrusion of a sudden obstacle into the preset parking area may cause frequent path replanning; multiple vehicles applying for parking at the same time will also intensify competition for parking area resources, affecting overall traffic efficiency; and conflicts between the parking area and the trajectory of dynamic obstacles may cause local congestion or even mission failure.

[0004] However, conventional dynamic scheduling methods have limited applicability in complex scenarios. The park environment, characterized by the randomness of dynamic obstacles, the need for multi-vehicle coordination, and spatial limitations, places higher demands on the scheduling system. For example, when construction vehicles suddenly occupy parking spaces or vehicles apply for parking during peak hours, existing scheduling methods are prone to uneven resource allocation, leading to excessively long waiting times for some vehicles. Furthermore, conflict resolution strategies are less efficient and more likely to cause system instability.

[0005] For example, the combined interference of dynamic obstacles and multi-vehicle trajectory conflicts leads to path planning instability. Problems such as sudden obstacles intruding into the preset stopping area, multiple vehicles simultaneously requesting stopping leading to insufficient capacity, and conflicts between the stopping area and the movement trajectory of dynamic obstacles further highlight the limitations of existing technologies. These situations indicate that existing technologies need further optimization when dealing with complex dynamic scenarios to improve the robustness and adaptability of the scheduling system. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art, solve or at least alleviate the problem of path planning instability of connected unmanned vehicles in parks under the combined interference of dynamic obstacles and multi-vehicle trajectory conflicts, and provide a dynamic scheduling and planning method and system for connected unmanned vehicles in parks.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a dynamic scheduling and planning method for connected unmanned vehicles in a park, comprising the following steps:

[0008] S1. Collect three-dimensional trajectory data of dynamic obstacles through multi-source sensors and V2X communication equipment, and generate a four-dimensional accessibility map based on spatiotemporal grid map;

[0009] S2. Based on the motion state of the dynamic obstacle and the planned arrival time, remaining battery power, and importance of the cargo carried by the unmanned vehicle, different avoidance strategy weights are formulated.

[0010] S3. Apply the obtained avoidance strategy weights to the distributed decision-making unit, and optimize the multi-vehicle trajectory through a mixed integer programming model and asymmetric game strategy.

[0011] S4: Edge nodes monitor vehicle trajectory execution in real time and adjust the path based on cloud-based global optimization results;

[0012] S5. While determining the avoidance strategy weights, the dynamic threat value of the residence area is evaluated. When the threat value exceeds the set threshold, the residence area reconstruction process is triggered.

[0013] To further realize the present invention, the following technical solutions may be preferred:

[0014] Preferably, in step S5, the comprehensive threat value is calculated based on the direction and speed of the movement of the dynamic obstacle and its distance from the preset dwelling area;

[0015] A region is classified as low-threat if its threat value is less than the first threshold, as medium-threat if its threat value is between the first and second thresholds, and as high-threat if its threat value is greater than the second threshold.

[0016] Preferably, step S5 further includes the following steps:

[0017] When the threat value is in a low-threat zone and the length of the residency application queue is less than the first judgment value, the avoidance strategy weight is changed.

[0018] When the threat value is in a high-threat area and the length of the stay application queue is greater than the second judgment value, the avoidance strategy weight is changed.

[0019] Change the avoidance strategy weight when the threat value is in the medium threat zone and the length of the residency application queue is stable;

[0020] The weight of the avoidance strategy is greater when the threat value is in a low-threat zone than when the threat value is in a high-threat zone, and the weight of the avoidance strategy is greater when the threat value is in a high-threat zone than when the threat value is in a medium-threat zone.

[0021] Preferably, in step S3, the avoidance strategy includes an emergency detour weight, a deceleration and parking weight, and a lateral offset weight. The emergency detour weight, the deceleration and parking weight, and the lateral offset weight are multiplied by the avoidance strategy weight to obtain the actual avoidance strategy weight.

[0022] Preferably, step S3 further includes the following steps:

[0023] The real-time position of the dynamic obstacle is compared with the emergency detour weight and the deceleration and stop weight. When the real-time position threat value is greater than the deceleration and stop weight, it enters the lateral offset state.

[0024] When the real-time location threat value is less than the emergency detour weight, the planned arrival time of the unmanned vehicle is compared with the deceleration and parking weight. When the planned arrival time is less than the deceleration and parking weight, the vehicle enters the deceleration and parking state. When the planned arrival time is greater than the deceleration and parking weight, the vehicle enters the emergency detour state.

[0025] When the real-time location threat value is greater than the emergency detour weight but less than the deceleration and parking weight, the planned arrival time of the unmanned vehicle is compared with the deceleration and parking weight. When the planned arrival time is less than the deceleration and parking weight, the vehicle enters the lateral offset state. When the planned arrival time is greater than the deceleration and parking weight, the vehicle enters the deceleration and parking state.

[0026] Preferably, step S5 further includes the following steps:

[0027] The real-time location of dynamic obstacles is obtained through V2X communication devices, and the threat value change rate of the current time window is obtained. The threat value change rate of each time window is recorded in sequence.

[0028] The threat value change rate of the current time window is compared with the threat value change rates of the previous multiple time windows. If the threat value change rates of multiple time windows are not all positive and do not all exceed the set value, the residence zone allocation status is maintained.

[0029] When the threat value change rate of multiple time windows is positive and exceeds the set value, the residence area allocation status is changed to emergency reconstruction status, and the residence area allocation status remains in emergency reconstruction status in the first subsequent time window.

[0030] The status of the residence area allocation in the subsequent second time window will be changed based on the threat value change rate in the subsequent first and second time windows.

[0031] Preferably, step S5 further includes the following steps:

[0032] Determine the positive or negative status of the threat value change rate in the first subsequent time window. If the threat value change rate is positive, determine the positive or negative status of the threat value change rate in the second subsequent time window. If the threat value change rate is positive, maintain the emergency reconstruction status for the garrison area allocation. If the threat value change rate is negative, determine the absolute value of the threat value change rate. If the absolute value of the threat value change rate is less than the set value, maintain the emergency reconstruction status. If the absolute value of the threat value change rate is greater than the set value, change back to the original garrison area allocation status.

[0033] If the threat value change rate is negative, the system will determine whether the threat value change rate is positive or negative in the second time window. If the threat value change rate is negative, the garrison zone allocation status will change back to the original garrison zone allocation status. If the threat value change rate is positive, the system will determine the absolute value of the threat value change rate. If the absolute value of the threat value change rate is greater than the set value, the emergency reconfiguration status will be maintained. If the absolute value of the threat value change rate is less than the set value, the system will change back to the original garrison zone allocation status.

[0034] Preferably, in step S5, while assessing the rate of change of threat value, the following steps are also performed:

[0035] The intrusion depth of dynamic obstacles within the monitoring time window is monitored. When the intrusion depth exceeds the warning threshold, the status of the dwell area allocation is changed to emergency reconstruction status.

[0036] A dynamic scheduling and planning system for connected unmanned vehicles in a park, used in the aforementioned dynamic scheduling and planning method, includes a perception module, a decision-making module, a storage module, an allocation module, a control module, and a verification module;

[0037] The perception module integrates LiDAR, millimeter-wave radar and camera, and constructs a four-dimensional motion model of dynamic obstacles by aligning with timestamps.

[0038] The decision-making module is deployed on edge nodes and achieves collaborative optimization of multi-vehicle trajectories through a trajectory optimization computing cluster.

[0039] The storage module stores three types of avoidance strategies and spatiotemporal raster maps, and maintains data consistency through a dynamic update mechanism.

[0040] The allocation module is deployed on edge nodes and implements dynamic buffer management through the residential area allocation engine.

[0041] The control module achieves dynamic coupling between manual configuration and adaptive optimization, and optimizes the allocation of the dwell area through a weight adjustment algorithm;

[0042] The verification module supports SUMO+ROS co-simulation testing and verifies trajectory stability through digital twin technology.

[0043] Preferably, it further includes an exception handling module, which is connected in parallel with the allocation module;

[0044] During emergency reconfiguration, the response frequency of the anomaly handling module is linked to the threat value change rate, suppressing local congestion and accelerating the release of resources in the dwell area.

[0045] The beneficial effects of this invention are:

[0046] This invention distinguishes park operation scenarios through dynamic obstacle threat value recognition technology, dynamically adjusts the actual avoidance strategy weight according to the threat value type, and enters different scenario scheduling states according to the actual avoidance strategy weight, thereby improving the accuracy of scheduling actions, reducing the misjudgment rate and reducing the task failure rate while ensuring vehicle safety.

[0047] Meanwhile, this invention assesses the rate of change of threat values ​​to correct the allocation status of the residence area, predicts the intrusion of dynamic obstacles, and adjusts the scheduling status in combination with the characteristics of the park environment to ensure vehicle safety.

[0048] In addition, the dynamic scheduling planning system of the present invention makes matching adjustments according to different scheduling states, which not only plays a better scheduling role, but also improves the accuracy of resource allocation. Attached Figure Description

[0049] Figure 1 This is a flowchart of the dynamic scheduling planning method of the present invention.

[0050] Figure 2 This is a flowchart of step S5 of the present invention.

[0051] Figure 3 This is a flowchart of the avoidance strategy weight adjustment process of the present invention.

[0052] Figure 4 This is a flowchart for assessing the rate of change of threat value in this invention.

[0053] Figure 5 This is a flowchart illustrating the weight calculation of the avoidance strategy according to the present invention. Detailed Implementation

[0054] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0055] Example 1

[0056] The networked unmanned vehicle dispatching system in industrial parks generally relies on preset path planning and static parking area allocation strategies. While this improves vehicle operating efficiency to some extent, it can easily lead to path planning instability under the combined interference of dynamic obstacles and multi-vehicle trajectory conflicts. To improve the stability and adaptability of the dispatching system, corresponding measures need to be taken to address issues such as sudden obstacles intruding into preset parking areas, multiple vehicles simultaneously requesting parking leading to insufficient capacity, and conflicts between parking areas and the movement trajectories of dynamic obstacles.

[0057] Conventional dynamic scheduling methods have limited applicability in complex scenarios. The park environment, characterized by the randomness of dynamic obstacles, the need for multi-vehicle coordination, and spatial limitations, places higher demands on the scheduling system. For example, when construction vehicles suddenly occupy parking spaces or vehicles apply for parking during peak hours, existing scheduling methods are prone to uneven resource allocation, leading to excessively long waiting times for some vehicles. Furthermore, conflict resolution strategies are less efficient and more likely to cause system instability.

[0058] This embodiment discloses an invention that provides a dynamic scheduling and planning system for networked unmanned vehicles in a park. In practical applications, the system operates with unmanned vehicles within the park as the target, and achieves dynamic scheduling and path optimization through the collaborative work of multiple modules.

[0059] A dynamic scheduling and planning system for networked unmanned vehicles in a park includes a perception module, a decision-making module, a storage module, an allocation module, a control module, and a verification module.

[0060] The perception module integrates LiDAR, millimeter-wave radar and camera, and constructs a four-dimensional motion model of dynamic obstacles by aligning with timestamps.

[0061] The decision-making module is deployed on edge nodes and achieves collaborative optimization of multi-vehicle trajectories through a trajectory optimization computing cluster.

[0062] The storage module stores three types of avoidance strategies and spatiotemporal raster maps, and maintains data consistency through a dynamic update mechanism.

[0063] The allocation module is deployed on edge nodes and implements dynamic buffer management through the residential area allocation engine.

[0064] The control module achieves dynamic coupling between manual configuration and adaptive optimization, and optimizes the allocation of the dwell area through a weight adjustment algorithm;

[0065] The verification module supports SUMO+ROS co-simulation testing and verifies trajectory stability through digital twin technology.

[0066] It also includes an exception handling module, which is connected in parallel with the allocation module;

[0067] During emergency reconfiguration, the response frequency of the anomaly handling module is linked to the threat value change rate, suppressing local congestion and accelerating the release of resources in the dwell area.

[0068] In a real-world application scenario, suppose an autonomous vehicle is performing a cargo transportation task within a park, encountering multiple dynamic obstacles along the way. The perception module collects 3D trajectory data of the obstacles using multi-source sensors and uploads the data to the decision-making module. The decision-making module calculates the optimal avoidance strategy weight based on the obstacle's motion state and the autonomous vehicle's task priority, and transmits the result to the allocation module. The allocation module dynamically adjusts the allocation status of the dwell area based on the avoidance strategy data provided by the storage module. Simultaneously, the control module monitors the usage of the dwell area in real time and adjusts the dwell area allocation status according to the threat value change rate. The verification module uses digital twin technology to simulate and verify the entire process, ensuring the stability and security of path planning. Throughout the process, the anomaly handling module is on standby. If an abnormal threat value change rate or intrusion depth is detected, an emergency reconfiguration process is immediately initiated to ensure the safe operation of the autonomous vehicle.

[0069] Example 2

[0070] Reference Figure 1 - Figure 5 To improve the performance of the aforementioned system, this embodiment provides a dynamic scheduling and planning method for connected unmanned vehicles in a park, which includes the following steps:

[0071] S1. Collect three-dimensional trajectory data of dynamic obstacles through multi-source sensors and V2X communication equipment, and generate a four-dimensional accessibility map based on spatiotemporal grid map;

[0072] S2. Based on the motion state of the dynamic obstacle and the planned arrival time, remaining battery power, and importance of the cargo carried by the unmanned vehicle, different avoidance strategy weights are formulated.

[0073] S3. Apply the obtained avoidance strategy weights to the distributed decision-making unit, and optimize the multi-vehicle trajectory through a mixed integer programming model and asymmetric game strategy.

[0074] S4: Edge nodes monitor vehicle trajectory execution in real time and adjust the path based on cloud-based global optimization results;

[0075] S5. While determining the avoidance strategy weights, the dynamic threat value of the residence area is evaluated. When the threat value exceeds the set threshold, the residence area reconstruction process is triggered.

[0076] In step S5, the comprehensive threat value is calculated by the direction and speed of the movement of the dynamic obstacle and its distance from the preset dwelling area.

[0077] A region is classified as low-threat if its threat value is less than the first threshold, as medium-threat if its threat value is between the first and second thresholds, and as high-threat if its threat value is greater than the second threshold.

[0078] Step S5 further includes the following steps:

[0079] When the threat value is in a low-threat zone and the length of the residency application queue is less than the first judgment value, the avoidance strategy weight is changed.

[0080] When the threat value is in a high-threat area and the length of the stay application queue is greater than the second judgment value, the avoidance strategy weight is changed.

[0081] Change the avoidance strategy weight when the threat value is in the medium threat zone and the length of the residency application queue is stable;

[0082] The weight of the avoidance strategy is greater when the threat value is in a low-threat zone than when the threat value is in a high-threat zone, and the weight of the avoidance strategy is greater when the threat value is in a high-threat zone than when the threat value is in a medium-threat zone.

[0083] In step S3, the avoidance strategy includes emergency detour weight, deceleration and parking weight, and lateral offset weight. The emergency detour weight, deceleration and parking weight, and lateral offset weight are multiplied by the avoidance strategy weight to obtain the actual avoidance strategy weight.

[0084] Step S3 further includes the following steps:

[0085] The real-time position of the dynamic obstacle is compared with the emergency detour weight and the deceleration and stop weight. When the real-time position threat value is greater than the deceleration and stop weight, it enters the lateral offset state.

[0086] When the real-time location threat value is less than the emergency detour weight, the planned arrival time of the unmanned vehicle is compared with the deceleration and parking weight. When the planned arrival time is less than the deceleration and parking weight, the vehicle enters the deceleration and parking state. When the planned arrival time is greater than the deceleration and parking weight, the vehicle enters the emergency detour state.

[0087] When the real-time location threat value is greater than the emergency detour weight but less than the deceleration and parking weight, the planned arrival time of the unmanned vehicle is compared with the deceleration and parking weight. When the planned arrival time is less than the deceleration and parking weight, the vehicle enters the lateral offset state. When the planned arrival time is greater than the deceleration and parking weight, the vehicle enters the deceleration and parking state.

[0088] Step S5 further includes the following steps:

[0089] The real-time location of dynamic obstacles is obtained through V2X communication devices, and the threat value change rate of the current time window is obtained. The threat value change rate of each time window is recorded in sequence.

[0090] The threat value change rate of the current time window is compared with the threat value change rates of the previous multiple time windows. If the threat value change rates of multiple time windows are not all positive and do not all exceed the set value, the residence zone allocation status is maintained.

[0091] When the threat value change rate of multiple time windows is positive and exceeds the set value, the residence area allocation status is changed to emergency reconstruction status, and the residence area allocation status remains in emergency reconstruction status in the first subsequent time window.

[0092] The status of the residence area allocation in the subsequent second time window will be changed based on the threat value change rate in the subsequent first and second time windows.

[0093] Step S5 further includes the following steps:

[0094] Determine the positive or negative status of the threat value change rate in the first subsequent time window. If the threat value change rate is positive, determine the positive or negative status of the threat value change rate in the second subsequent time window. If the threat value change rate is positive, maintain the emergency reconstruction status for the garrison area allocation. If the threat value change rate is negative, determine the absolute value of the threat value change rate. If the absolute value of the threat value change rate is less than the set value, maintain the emergency reconstruction status. If the absolute value of the threat value change rate is greater than the set value, change back to the original garrison area allocation status.

[0095] If the threat value change rate is negative, the status of the threat value change rate in the second subsequent time window is determined. If the threat value change rate is negative, the garrison zone allocation status is changed back to the original garrison zone allocation status. If the threat value change rate is positive, the absolute value of the threat value change rate is determined. If the absolute value of the threat value change rate is greater than the set value, the emergency reconfiguration status is maintained. If the absolute value of the threat value change rate is less than the set value, the garrison zone allocation status is changed back to the original garrison zone allocation status.

[0096] In step S5, while assessing the rate of change of threat value, the following steps are also performed:

[0097] The intrusion depth of dynamic obstacles within the monitoring time window is monitored. When the intrusion depth exceeds the warning threshold, the status of the dwell area allocation is changed to emergency reconstruction status.

[0098] This invention distinguishes park operation scenarios through dynamic obstacle threat value recognition technology, dynamically adjusts the actual avoidance strategy weight according to the threat value type, and enters different scenario scheduling states according to the actual avoidance strategy weight, thereby improving the accuracy of scheduling actions, reducing the misjudgment rate and reducing the task failure rate while ensuring vehicle safety.

[0099] Meanwhile, this invention assesses the rate of change of threat values ​​to correct the allocation status of the residence area, predicts the intrusion of dynamic obstacles, and adjusts the scheduling status in combination with the characteristics of the park environment to ensure vehicle safety.

[0100] In addition, the dynamic scheduling planning system of the present invention makes matching adjustments according to different scheduling states, which not only plays a better scheduling role, but also improves the accuracy of resource allocation.

[0101] Example 3

[0102] To enable those skilled in the art to fully understand and implement this invention, the specific implementation principles of this invention are further supplemented below with a specific application scenario.

[0103] During the operation of the park's connected unmanned vehicle dynamic scheduling system, the perception module first collects real-time location information of dynamic obstacles within the park through the collaborative efforts of LiDAR, millimeter-wave radar, and cameras. LiDAR generates high-precision 3D point cloud data, millimeter-wave radar detects the speed and distance of obstacles, and cameras are responsible for identifying and classifying the appearance features of obstacles. This sensor data is then uploaded to edge nodes after being timestamped. The edge nodes generate a four-dimensional accessibility map based on a spatiotemporal grid map. This map not only contains the spatial location information of dynamic obstacles but also covers their speed, direction, and temporal trends, providing fundamental data support for the subsequent decision-making module. As the core input component of the entire system, the perception module's output is directly transmitted to the decision-making module.

[0104] Upon entering the decision-making module, the system calculates the weight of the optimal avoidance strategy based on the motion state of the dynamic obstacle and the task priority of the autonomous vehicle. For example, when the overall threat value of the dynamic obstacle is greater than the deceleration and parking weight, the system adjusts the autonomous vehicle to a lateral drift state; when the threat value is less than the emergency detour weight and the planned arrival time is less than the deceleration and parking weight, the system adjusts the autonomous vehicle to a deceleration and parking state; when the planned arrival time is greater than the deceleration and parking weight, the system adjusts the autonomous vehicle to an emergency detour state. The decision-making module optimizes multi-vehicle trajectory coordination through a mixed-integer programming model and asymmetric game strategy to ensure efficient vehicle operation in complex scenarios.

[0105] The main function of the storage module is to store the avoidance strategy and the spatio-temporal grid map, and maintain data consistency through a dynamic update mechanism. For example, when the threat value in a certain area continues to rise, the storage module will automatically adjust the weight of the avoidance strategy for that area and synchronize the updated data to the allocation module. The allocation module dynamically adjusts the allocation status of the residence area according to the output result of the decision-making module and the avoidance strategy data provided by the storage module. For example, when the threat value is in a low-threat area and the length of the residence application queue is less than the first judgment value, the allocation module will change the weight of the avoidance strategy; when the threat value is in a high-threat area and the length of the residence application queue is greater than the second judgment value, the allocation module will also change the weight of the avoidance strategy; and when the threat value is in a medium-threat area and the length of the residence application queue is stable, the allocation module will also adjust the weight of the avoidance strategy accordingly. This dynamic adjustment mechanism ensures the efficient utilization of residence area resources.

[0106] The function of the regulation module is to achieve the dynamic coupling of manual configuration and adaptive optimization, and optimize the residence area allocation through a weight adjustment algorithm. For example, when the change rates of the threat values in multiple time windows are all positive and exceed the set value, the regulation module will change the allocation status of the residence area to the emergency reconstruction state and maintain the emergency reconstruction state in the first subsequent time window. In addition, the regulation module also supports the linkage response of the exception handling module. When the change rate of the threat value is negative and the absolute value is less than the set value, the regulation module will change the allocation status of the residence area to the original residence area allocation status. The operation interface of the regulation module is designed to be simple and intuitive, facilitating manual intervention by operators.

[0107] The verification module verifies the trajectory stability through the SUMO+ROS joint simulation test, combined with digital twin technology. The verification module is connected to the decision-making module and the regulation module through a dedicated communication protocol, and can obtain the trajectory execution situation and the residence area allocation status of the unmanned vehicle in real time. Through the analysis of the simulation results, the verification module evaluates the accuracy and stability of the path planning and feeds back the evaluation results to the decision-making module and the regulation module for further optimizing the scheduling strategy . For example, in a certain specific scenario, the verification module finds that the lateral offset frequency of the unmanned vehicle is too high, which may affect the task completion efficiency. At this time, the verification module will send an adjustment suggestion to the decision-making module to reduce the usage frequency of the lateral offset weight.

[0108] Throughout the process, the exception handling module is on standby at any time. Once it detects that the intrusion depth of the dynamic obstacle in a certain time window is greater than the warning threshold, the exception handling module will immediately trigger the change process of the residence area allocation status and notify the regulation module to make corresponding adjustments. For example, when the intrusion depth of the dynamic obstacle exceeds the warning threshold, the exception handling module will respond quickly, suppress local congestion and accelerate the release of residence area resources, thus ensuring vehicle safety.

[0109] In a real-world application scenario, suppose an autonomous vehicle is performing a cargo transportation task within a park, encountering multiple dynamic obstacles along the way. The perception module collects 3D trajectory data of the obstacles using multi-source sensors and uploads the data to the decision-making module. The decision-making module calculates the optimal avoidance strategy weight based on the obstacle's motion state and the autonomous vehicle's task priority, and transmits the result to the allocation module. The allocation module dynamically adjusts the allocation status of the dwell area based on the avoidance strategy data provided by the storage module. Simultaneously, the control module monitors the usage of the dwell area in real time and adjusts the dwell area allocation status according to the threat value change rate. The verification module uses digital twin technology to simulate and verify the entire process, ensuring the stability and security of path planning. Throughout the process, the anomaly handling module is on standby. If an abnormal threat value change rate or intrusion depth is detected, an emergency reconfiguration process is immediately initiated to ensure the safe operation of the autonomous vehicle.

[0110] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A dynamic scheduling and planning method for connected unmanned vehicles in a park, characterized in that, Includes the following steps: S1. Collect three-dimensional trajectory data of dynamic obstacles through multi-source sensors and V2X communication equipment, and generate a four-dimensional accessibility map based on spatiotemporal grid map; S2. Based on the motion state of the dynamic obstacle and the planned arrival time, remaining battery power, and importance of the cargo carried by the unmanned vehicle, different avoidance strategy weights are formulated. S3. Apply the obtained avoidance strategy weights to the distributed decision-making unit, and optimize the multi-vehicle trajectory through a mixed integer programming model and asymmetric game strategy. S4: Edge nodes monitor vehicle trajectory execution in real time and adjust the path based on cloud-based global optimization results; S5. While determining the avoidance strategy weights, the dynamic threat value of the camp area is evaluated. When the threat value exceeds the set threshold, the camp area reconstruction process is triggered. In step S3, the avoidance strategy includes emergency detour weight, deceleration and parking weight, and lateral offset weight. The emergency detour weight, deceleration and parking weight, and lateral offset weight are multiplied by the avoidance strategy weight to obtain the actual avoidance strategy weight. Step S3 further includes the following steps: The real-time position threat value of the dynamic obstacle is compared with the emergency detour weight and the deceleration and parking weight. When the real-time position threat value is greater than the deceleration and parking weight, the obstacle enters the lateral offset state. When the real-time location threat value is less than the emergency detour weight, the planned arrival time of the unmanned vehicle is compared with the deceleration and parking weight. When the planned arrival time is less than the deceleration and parking weight, the vehicle enters the deceleration and parking state. When the planned arrival time is greater than the deceleration and parking weight, the vehicle enters the emergency detour state. When the real-time location threat value is greater than the emergency detour weight but less than the deceleration and parking weight, the planned arrival time of the unmanned vehicle is compared with the deceleration and parking weight. When the planned arrival time is less than the deceleration and parking weight, it enters the lateral offset state; when the planned arrival time is greater than the deceleration and parking weight, it enters the deceleration and parking state. Step S5 further includes the following steps: The real-time location of dynamic obstacles is obtained through V2X communication devices, and the threat value change rate of the current time window is obtained. The threat value change rate of each time window is recorded in sequence. The threat value change rate of the current time window is compared with the threat value change rates of the previous multiple time windows. If the threat value change rates of multiple time windows are not all positive and do not all exceed the set value, the residence zone allocation status is maintained. When the threat value change rate of multiple time windows is positive and exceeds the set value, the residence area allocation status is changed to emergency reconstruction status, and the residence area allocation status remains in emergency reconstruction status in the first subsequent time window. The status of the residence area allocation in the subsequent second time window will be changed based on the threat value change rate in the subsequent first and second time windows.

2. The dynamic scheduling and planning method for networked unmanned vehicles in industrial parks according to claim 1, characterized in that, In step S5, the comprehensive threat value is calculated by the direction and speed of the movement of the dynamic obstacle and its distance from the preset dwelling area. A region is classified as low-threat if its threat value is less than the first threshold, as medium-threat if its threat value is between the first and second thresholds, and as high-threat if its threat value is greater than the second threshold.

3. The dynamic scheduling and planning method for networked unmanned vehicles in industrial parks according to claim 2, characterized in that, Step S5 further includes the following steps: When the threat value is in a low-threat zone and the length of the residency application queue is less than the first judgment value, the avoidance strategy weight is changed. When the threat value is in a high-threat area and the length of the stay application queue is greater than the second judgment value, the avoidance strategy weight is changed. Change the avoidance strategy weight when the threat value is in the medium threat zone and the length of the residency application queue is stable; The weight of the avoidance strategy is greater when the threat value is in a low-threat zone than when the threat value is in a high-threat zone, and the weight of the avoidance strategy is greater when the threat value is in a high-threat zone than when the threat value is in a medium-threat zone.

4. The dynamic scheduling and planning method for networked unmanned vehicles in industrial parks according to claim 1, characterized in that, Step S5 further includes the following steps: Determine the positive or negative status of the threat value change rate in the first subsequent time window. If the threat value change rate is positive, determine the positive or negative status of the threat value change rate in the second subsequent time window. If the threat value change rate is positive, maintain the emergency reconstruction status for the garrison area allocation. If the threat value change rate is negative, determine the absolute value of the threat value change rate. If the absolute value of the threat value change rate is less than the set value, maintain the emergency reconstruction status. If the absolute value of the threat value change rate is greater than the set value, change back to the original garrison area allocation status. If the threat value change rate is negative, the status of the threat value change rate in the second subsequent time window is determined. If the threat value change rate is negative, the garrison zone allocation status is changed back to the original garrison zone allocation status. If the threat value change rate is positive, the absolute value of the threat value change rate is determined. If the absolute value of the threat value change rate is greater than the set value, the emergency reconfiguration status is maintained. If the absolute value of the threat value change rate is less than the set value, the garrison zone allocation status is changed back to the original garrison zone allocation status.

5. The dynamic scheduling and planning method for networked unmanned vehicles in industrial parks according to claim 1, characterized in that, In step S5, while assessing the rate of change of threat value, the following steps are also performed: The intrusion depth of dynamic obstacles within the monitoring time window is monitored. When the intrusion depth exceeds the warning threshold, the status of the dwell area allocation is changed to emergency reconstruction status.

6. A dynamic scheduling and planning system for connected unmanned vehicles in a park, used in the dynamic scheduling and planning method for connected unmanned vehicles in a park as described in any one of claims 1-5, characterized in that, It includes a perception module, a decision-making module, a storage module, an allocation module, a control module, and a verification module; The perception module integrates LiDAR, millimeter-wave radar and camera, and constructs a four-dimensional motion model of dynamic obstacles by aligning with timestamps. The decision-making module is deployed on edge nodes and achieves collaborative optimization of multi-vehicle trajectories through a trajectory optimization computing cluster. The storage module stores three types of avoidance strategies and spatiotemporal raster maps, and maintains data consistency through a dynamic update mechanism. The allocation module is deployed on edge nodes and implements dynamic buffer management through the residential area allocation engine. The control module achieves dynamic coupling between manual configuration and adaptive optimization, and optimizes the allocation of the dwell area through a weight adjustment algorithm; The verification module supports SUMO+ROS co-simulation testing and verifies trajectory stability through digital twin technology.

7. The park-connected unmanned vehicle dynamic scheduling and planning system according to claim 6, characterized in that, It also includes an exception handling module, which is connected in parallel with the allocation module; During emergency reconfiguration, the response frequency of the anomaly handling module is linked to the threat value change rate, suppressing local congestion and accelerating the release of resources in the dwell area.

Images