Multi-mode collaborative operation method for offshore unmanned system

Abstract

The invention relates to the technical field of offshore unmanned system operation, and discloses an offshore unmanned system multi-mode collaborative operation method, which comprises the following steps of: carrying out topological calculation on an operation target based on a dynamic task graph and equipment parameters of unmanned equipment to obtain a task sequence; adjusting the operation state of the unmanned equipment based on the task sequence feedback; performing feedback correction on the task sequence based on the operation state and the progress deviation of the task sequence to obtain an optimized task sequence; coupling and resolving the marine environment sensing data and the equipment parameters to obtain flight environment parameters of the unmanned equipment; and based on the optimization task sequence and the flight environment parameters, adjusting the navigation trajectory of the unmanned device to obtain a synchronous collision avoidance path of the unmanned device. According to the invention, the problems of low operation efficiency and poor environmental adaptability of the offshore unmanned system during multi-mode collaborative operation can be solved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned maritime system operations, and in particular to a multi-modal collaborative operation method of an unmanned maritime system. Background Art

[0002] In the field of unmanned maritime system operations, existing technologies generally suffer from low operational efficiency and poor environmental adaptability. Traditional methods often rely on a single data source or fixed mission planning model, making them difficult to cope with the complex and changing marine environment. For example, existing systems often limit their collection of marine environmental perception data to a single observation method and lack the coordinated integration of multi-source data. This results in a one-sided understanding of the environment and an inability to construct accurate dynamic mission scenarios. This makes unmanned equipment prone to mission execution deviations due to environmental misjudgment when operating in complex waters.

[0003] At the same time, existing mission planning and trajectory adjustment mechanisms have significant deficiencies. Traditional methods typically use static mission sequence generation, which fails to fully consider the dynamic coupling relationship between unmanned equipment parameters and the marine environment. This makes it difficult to optimize mission processes in real time based on operational status, leading to resource conflicts and progress delays during mission execution. Furthermore, in collision avoidance path planning, existing technologies often make decisions based on single-moment environmental data, lacking comprehensive consideration of short-term environmental forecasts and equipment maneuverability boundaries. This results in weak synchronized collision avoidance capabilities for unmanned equipment in complex sea conditions, seriously impacting operational safety and collaborative efficiency. Summary of the Invention

[0004] The present invention provides a multi-modal collaborative operation method for an unmanned marine system, the main purpose of which is to solve the problems of low operating efficiency and poor environmental adaptability during the multi-modal collaborative operation of an unmanned marine system.

[0005] To achieve the above objectives, the present invention provides a multi-modal collaborative operation method for an unmanned maritime system, comprising: S1: Acquire ocean environment perception data and generate dynamic mission maps to obtain the operation targets of unmanned equipment; S2: performing topological calculation on the operation target based on the dynamic task graph and the equipment parameters of the unmanned equipment to obtain a task sequence; S3: adjusting the operation status of the unmanned equipment based on the task sequence feedback; S4: Based on the progress deviation between the job status and the task sequence, performing feedback correction on the task sequence to obtain an optimized task sequence; S5: Couple and solve the ocean environment perception data and the device parameters to obtain flight environment parameters of the unmanned device; S6: Adjusting the navigation trajectory of the unmanned equipment based on the optimized task sequence and the flight environment parameters to obtain a synchronous collision avoidance path for the unmanned equipment.

[0006] The acquiring of ocean environment perception data and generating of a dynamic task graph includes: Perform multi-source data collaborative collection operations on the preset ocean stereo observation network to obtain ocean environment perception data; Based on the marine environment perception data, collaborative situational spatiotemporal modeling is performed on the target sea area corresponding to the unmanned equipment to obtain a dynamic task map.

[0007] The obtaining of the operation target of the unmanned equipment includes: Performing target allocation for the unmanned equipment based on the dynamic task graph to generate a target allocation plan; The operation target of the unmanned equipment is determined based on the target allocation plan.

[0008] The topological calculation of the operation target based on the dynamic task graph and the equipment parameters of the unmanned equipment to obtain a task sequence includes: Building tasks for the operation targets based on the dynamic task graph and the equipment parameters of the unmanned equipment to obtain an operation target list for the unmanned equipment; Based on the fluid vortex conduction characteristics of the target sea area, the operation target list is divided into levels to obtain hierarchical tasks for the unmanned equipment; The mutually exclusive dynamic resource points in the hierarchical tasks are resolved to obtain the task sequence of the unmanned equipment.

[0009] The adjusting the operating state of the unmanned equipment based on the task sequence feedback includes: Monitoring the operational control state parameters of the unmanned equipment based on the task sequence to obtain closed-loop control core parameters; Reconstruct and solve the control instructions of the unmanned equipment based on the closed-loop control core parameters to obtain dynamic allocation instructions for the unmanned equipment; Adjust the operating status of the unmanned equipment based on the dynamic allocation instruction.

[0010] The step of performing feedback correction on the task sequence based on the progress deviation between the job status and the task sequence to obtain an optimized task sequence includes: Calculating the progress difference between the operation status and the progress planning standard of the task sequence to obtain a key delayed task node identifier; Performing perturbation source analysis on the operation state of the unmanned equipment based on the ocean environment perception data and the key delayed task node identifier to obtain a core induced delay factor; The task sequence is dynamically optimized based on the core induced delay factor to obtain an optimized task sequence.

[0011] The dynamically optimizing the task sequence based on the core induced delay factor to obtain an optimized task sequence includes: Re-assigning the task execution priority of the operation state of the unmanned equipment based on the core induced delay factor and the operating condition parameters of the unmanned equipment to obtain the task structure of the unmanned equipment; Performing a risk analysis on the operation status based on the short-term prediction results of the marine environment perception data and the task structure to obtain a sequence operation report of the unmanned equipment; Verifying the node integrity of the task sequence in sequence based on the sequence operation report; Optimize the optimization task sequence of the unmanned equipment based on the verification result of the verification.

[0012] The steps of obtaining the short-term forecast statistical results include: Predicting the ocean environment perception data based on a pre-trained adversarial network to obtain a comprehensive prediction result of the ocean environment perception data; Abnormal data in the comprehensive prediction result is eliminated to obtain a short-term prediction result of the prediction result.

[0013] The coupled solution of the ocean environment perception data and the device parameters to obtain the flight environment parameters of the unmanned device includes: performing coupling calibration on the unmanned equipment based on the ocean environment perception data and the equipment parameters to obtain an equivalent projection component of the unmanned equipment; performing a maneuverability critical boundary calculation on the equivalent projection component to obtain a maximum heading correction margin of the unmanned equipment in the corresponding sea area; Performing flight envelope synthesis on the navigation trajectory of the unmanned device based on the maximum heading correction margin and the attitude oscillation attenuation period of the unmanned device to obtain a safe maneuvering boundary parameter of the unmanned device; The safety maneuvering boundary parameters are tuned by equipment state feedback to obtain flight environment parameters.

[0014] The step of adjusting the navigation trajectory of the unmanned equipment based on the optimized task sequence and the flight environment parameters to obtain a synchronous collision avoidance path for the unmanned equipment includes: Performing multi-dimensional obstacle buffering on the navigation trajectory based on the flight environment parameters and the equipment parameters to obtain a dynamic threat envelope of the navigation trajectory; Performing a track deformation operation on the navigation trajectory based on the optimized task sequence and the dynamic threat envelope to obtain an adaptive track of the unmanned device; performing phase shift elimination on the navigation trajectory based on the ocean environment perception data and the adaptive track to obtain a conflict elimination path for the unmanned equipment; Based on the conflict resolution path and the flight environment parameters, a spatiotemporal synchronization verification operation is performed on the navigation trajectory to obtain a synchronous collision avoidance path for the unmanned equipment.

[0015] Beneficial effects

[0016] 1. This solution significantly improves the operational efficiency and environmental adaptability of unmanned systems at sea. By collaboratively collecting multi-source data to build a three-dimensional ocean observation network and combining it with collaborative spatiotemporal modeling to generate a dynamic task graph, it breaks through the limitations of a single traditional data source, achieves accurate understanding of the target sea environment, and provides comprehensive and real-time support for task planning. At the same time, topological calculations are performed based on the dynamic task graph and equipment parameters. By hierarchically dividing tasks and eliminating mutually exclusive resource points to form a reasonable task sequence, and combining closed-loop control to adjust the operating status in real time, it effectively resolves problems such as resource conflicts and delayed progress under the traditional static planning model, allowing unmanned equipment to more efficiently execute operational objectives.

[0017] 2. By coupling and solving ocean environment perception data with equipment parameters to obtain flight environment parameters, the team then optimizes the task sequence to adjust the trajectory, construct a dynamic threat envelope, and complete conflict resolution and spatiotemporal synchronization verification to generate a synchronized collision avoidance path. This process fully considers short-term ocean environment predictions and equipment safety maneuvering boundaries, avoiding the limitations of traditional collision avoidance planning that relies solely on single-time data. This significantly reduces the collision risk of unmanned equipment in complex sea conditions, ensures the synchronization and stability of multi-equipment collaborative operations, and further expands the operational capabilities of unmanned maritime systems in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A flowchart of a multi-modal collaborative operation method for an unmanned maritime system is provided in accordance with one embodiment of the present invention. DETAILED DESCRIPTION

[0019] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0020] The embodiment of the present application provides a method for multimodal collaborative operation of an unmanned marine system. The execution subject of the multimodal collaborative operation method of an unmanned marine system includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present application. In other words, the multimodal collaborative operation method of an unmanned marine system can be executed by software or hardware installed on a terminal device or a server device. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0021] Reference Figure 1 FIG. 1 is a flow chart of a multi-modal collaborative operation method for a maritime unmanned system according to an embodiment of the present invention. In this embodiment, the multi-modal collaborative operation method for a maritime unmanned system includes: S1: Acquire ocean environment perception data and generate dynamic mission maps to obtain the operation targets of unmanned equipment; In this embodiment, the acquisition of ocean environment perception data and generation of a dynamic task graph includes: Perform multi-source data collaborative collection operations on the preset ocean stereo observation network to obtain ocean environment perception data; Based on the marine environment perception data, collaborative situational spatiotemporal modeling is performed on the target sea area corresponding to the unmanned equipment to obtain a dynamic task map.

[0022] Specifically, the ocean stereoscopic observation network is a comprehensive observation network composed of sensors, unmanned equipment and data transmission equipment deployed at different levels of the ocean.

[0023] Multi-source data include various types of data such as temperature, salinity, current speed and direction, wave height and period, meteorological data, sea area image data, obstacle information, etc. collected from different sensors and equipment in the ocean stereo observation network.

[0024] Marine environmental perception data is obtained through the collaborative collection of multi-source data and is a comprehensive data set that can reflect the environmental conditions of the target sea area.

[0025] The target sea area is the specific ocean area where the unmanned equipment is about to operate, and its scope is determined according to operational requirements.

[0026] The dynamic mission map graphically displays the target sea area's environmental situation, mission objectives, unmanned equipment status, and the relationships between tasks. It updates in real time as the environment changes and the mission progresses.

[0027] Through a pre-defined ocean stereo observation network, various sensors and unmanned equipment within the network are controlled to operate simultaneously according to a coordinated collection strategy. For example, unmanned vessels sail on the ocean surface to collect data on sea surface temperature, salinity, and currents; underwater robots collect underwater topography and water quality data at a certain depth; and drones capture aerial images of the ocean and obtain meteorological data. These devices aggregate the collected data in real time through data transmission equipment, achieving coordinated multi-source data collection and ultimately generating comprehensive ocean environmental perception data.

[0028] Based on acquired marine environmental perception data, a spatiotemporal modeling algorithm is used to analyze and model the target sea area. In the temporal dimension, this considers how environmental parameters vary over time, such as current and wave variations at different moments. In the spatial dimension, a three-dimensional model of the target sea area is constructed, integrating environmental data from different locations. This collaborative spatiotemporal modeling transforms environmental data into an intuitive dynamic mission map, clearly presenting the environmental situation in the target sea area and providing a basis for subsequent mission planning.

[0029] In this embodiment, obtaining the operation target of the unmanned equipment includes: Performing target allocation for the unmanned equipment based on the dynamic task graph to generate a target allocation plan; The operation target of the unmanned equipment is determined based on the target allocation plan.

[0030] Specifically, the target allocation plan is a plan that assigns specific task goals to each unmanned device based on the dynamic task map and factors such as the performance parameters and operating capabilities of the unmanned equipment.

[0031] The operating objectives of unmanned equipment are the specific tasks that the unmanned equipment needs to complete during this operation, such as patrolling a specific area, detecting specific targets, collecting samples, etc.

[0032] Based on the generated dynamic task graph, the task requirements and the characteristics of each unmanned device are analyzed, and a reasonable target allocation algorithm is used to assign tasks to the unmanned devices. For example, tasks requiring large-scale sea patrols can be assigned to unmanned vessels with long endurance and high speed; tasks requiring detailed underwater detection can be assigned to underwater robots with high-precision detection equipment. This allocation process generates a target allocation plan.

[0033] Based on the generated target allocation plan, the specific operational objectives that each unmanned device needs to complete during operation are clearly defined. The determination of operational objectives must be combined with the environmental information in the dynamic task graph to ensure that the objectives are achievable within the current physical environment. For example, setting operational objectives to avoid areas with known dangerous obstacles can be crucial.

[0034] S2: performing topological calculation on the operation target based on the dynamic task graph and the equipment parameters of the unmanned equipment to obtain a task sequence; In this embodiment, a topological solution is performed on the operation target based on the dynamic task graph and the equipment parameters of the unmanned equipment to obtain a task sequence, including: Building tasks for the operation targets based on the dynamic task graph and the equipment parameters of the unmanned equipment to obtain an operation target list for the unmanned equipment; Based on the fluid vortex conduction characteristics of the target sea area, the operation target list is divided into levels to obtain hierarchical tasks for the unmanned equipment; The mutually exclusive dynamic resource points in the hierarchical tasks are resolved to obtain the task sequence of the unmanned equipment.

[0035] Specifically, the equipment parameters of unmanned equipment refer to the physical performance indicators of unmanned equipment, including endurance, load limit, maneuverability, environmental adaptation threshold, etc.

[0036] The work target list refers to breaking down abstract work targets into a set of specific executable sub-tasks, such as the list-based expression of "water quality sampling in sea area A + topographic mapping in sea area B", which clearly defines the spatial coordinates, execution time, required equipment functions, etc. of each sub-task.

[0037] Based on the environmental constraints reflected in the dynamic task graph, such as avoiding strong vortex areas, and the performance limits of equipment parameters, such as excluding deep-water robots from shallow areas, the operation objectives are structured and broken down. By matching equipment capabilities with environmental feasibility, the operation objectives are converted into specific subtask items to form an operation objective list. For example, if the dynamic task graph shows that the current speed in a certain area exceeds the unmanned vessel's resistance threshold, the sampling task in that area will be assigned to an underwater robot with stronger resistance capabilities.

[0038] More specifically, the direction of the current in the dynamic task graph determines the order of the subtasks, with downstream tasks taking priority; The device's battery life parameters limit the geographical scope of subtasks. For example, tasks far from the base station require reserved battery power for return. The water depth data of the target sea area is used to screen and adapt equipment, such as excluding deep-diving equipment in shallow water areas.

[0039] Furthermore, the fluid vortex conduction characteristics of the target sea area refer to the physical parameters of the intensity, propagation direction, impact range and energy attenuation law of the vortex in the sea area, such as the vortex rotation speed, the moving trajectory of the vortex center, the disturbance intensity of the vortex on the surrounding water flow, etc.

[0040] Hierarchical tasks are a hierarchical structure of tasks divided according to spatiotemporal dependencies and the degree of environmental impact, such as "basic environmental detection layer → core area sampling layer → data return layer". The upper-level tasks provide basic environmental data for the lower-level tasks.

[0041] The task list is sorted and grouped based on the fluid's vortex conduction characteristics. First, the impact of vortices on task execution is analyzed. Tasks in strong vortex areas should be prioritized or adjusted to a window of opportunity after the vortex has decayed. Then, tasks are divided into levels based on "environmental dependencies." For example, peripheral environmental detection tasks unaffected by vortices are performed first, followed by sampling tasks in core areas more affected by vortices, forming a hierarchical task structure.

[0042] More specifically, the direction of vortex propagation determines the order of task execution, such as arranging tasks in advance against the vortex propagation direction; The vortex's influence range defines the mission's safety boundary, and hierarchical tasks must avoid the vortex's core disturbance zone; The viscosity coefficient of the fluid affects the vortex decay rate, which in turn determines the level interval of the mission time window.

[0043] Furthermore, mutually exclusive dynamic resource points include limited resources or conflicting spaces that are competed for by multiple tasks, such as simultaneous access requests to the same sampling point, conflicts in the occupancy of shared communication frequency bands, and simultaneous use of device charging ports.

[0044] A task sequence refers to the linearized order of task execution after eliminating resource conflicts, clarifying the start time, execution duration, resource allocation plan, and equipment scheduling path of each task.

[0045] A dynamic resource scheduling algorithm identifies resource conflicts within hierarchical tasks. For spatial conflicts, such as when two tasks need to enter the same narrow waterway at the same time, task time windows are adjusted based on ocean current speeds and equipment maneuverability, prioritizing faster equipment. For resource conflicts, such as shared detection equipment, usage windows are allocated based on task priority, with buffer time reserved for resource switching, such as equipment calibration. Ultimately, hierarchical tasks are transformed into a conflict-free, linear execution sequence.

[0046] More specifically, the spatial narrowness of the target sea area exacerbates path conflicts, which need to be resolved through time staggering; The attenuation characteristics of communication signals require that a communication window be reserved in the task sequence; The tidal cycle affects the execution time of tasks in the shallow area, and the tidal rise and fall times need to be matched in the sequence.

[0047] S3: adjusting the operation status of the unmanned equipment based on the task sequence feedback; In this embodiment, adjusting the operating state of the unmanned equipment based on the task sequence feedback includes: Monitoring the operational control state parameters of the unmanned equipment based on the task sequence to obtain closed-loop control core parameters; Reconstruct and solve the control instructions of the unmanned equipment based on the closed-loop control core parameters to obtain dynamic allocation instructions for the unmanned equipment; Adjust the operating status of the unmanned equipment based on the dynamic allocation instruction.

[0048] Specifically, the operational control state parameters include the speed, heading, altitude, depth, attitude angle, power system output power, battery charge, etc. of the unmanned equipment. In the complex ocean physical environment, these parameters are constantly affected by factors such as ocean currents, waves, sea breezes, and magnetic fields. For example, the flow of ocean currents will change the actual navigation speed and direction of the unmanned equipment, resulting in changes in speed and heading parameters; the ups and downs of the waves may cause fluctuations in the altitude or depth of the unmanned equipment; the size and direction of the sea breeze will affect the attitude angle of the drone. In order to maintain stable flight, the power system output power will also be adjusted accordingly. The closed-loop control core parameters are the key parameters selected from these operational control state parameters and are used for the subsequent precise control of the equipment. Parameters such as speed, heading, attitude angle, etc. that directly affect the motion trajectory and operating capabilities of the equipment are usually included.

[0049] During offshore operations, unmanned equipment is equipped with various sensors to monitor these operational control parameters in real time. For example, GPS and inertial measurement units (IMUs) acquire position, velocity, and attitude information; pressure sensors measure the underwater equipment's depth; power sensors monitor the power output of the propulsion system; and battery sensors monitor the battery charge. These sensors continuously collect data and transmit it to the equipment's control system. The control system analyzes and filters this massive amount of data, extracting the core closed-loop control parameters, which provide the key basis for subsequent control command reconstruction and resolution. This process enables unmanned equipment to understand its own operational status in real time, laying the foundation for precise operations.

[0050] Furthermore, the core parameters of the closed-loop control are used as input in this link, and the task priority in the task sequence, the task execution time requirements, the real-time change data of the ocean environment, and the performance parameters of the unmanned equipment itself are also involved. These parameters are interrelated and jointly affect the generation of dynamic allocation instructions. For example, if the task priority is high and requires to be completed in a short time, and the current direction of the ocean is conducive to quickly reaching the task location, then when generating dynamic allocation instructions, it will tend to let the unmanned equipment move towards the task location at a higher speed and a larger turning angle.

[0051] After receiving the core closed-loop control parameters, the control system combines them with the other relevant parameters mentioned above and performs complex calculations and analyses based on specific control algorithms and strategies. First, the unmanned device determines the task it should currently focus on, based on task priority and execution time requirements. Then, the control instructions are reconstructed and resolved, taking into account real-time changes in the ocean environment and the device's own performance. For example, if the current speed is high and the direction is opposite to the mission location, the control system will calculate the need to increase power output to overcome current resistance, while adjusting the heading angle to maintain the correct direction of travel. This series of calculations and analyses ultimately generates a dynamic allocation instruction appropriate to the current situation. This instruction specifies the unmanned device's subsequent specific operational requirements in terms of speed, heading, power, and other aspects. This step allows the unmanned device to make reasonable decisions and take action based on various information.

[0052] Furthermore, the dynamic allocation instructions include speed adjustment parameters, heading adjustment parameters, power system adjustment parameters, and equipment operating mode switching parameters. These parameters directly determine the direction and degree of change in the unmanned equipment's operating status. In the marine environment, different operating tasks and environmental conditions require different parameter settings. For example, during a maritime patrol mission, if a suspected target area is discovered, the dynamic allocation instructions may instruct the unmanned equipment to reduce speed, adjust its heading to approach the target area, switch to high-resolution detection mode, and appropriately increase the power system output power to maintain a stable detection posture.

[0053] The actuators of the unmanned vehicle adjust their movements accordingly based on dynamically assigned commands. For speed adjustments, if the command calls for acceleration, the motors or propellers increase power output to increase the vehicle's speed; if deceleration is required, power output is reduced. For course adjustments, the servos or steering mechanism change the vehicle's direction of travel to maintain the commanded course. When power system parameters are adjusted, the operating state of the motors or propellers changes accordingly, such as adjusting motor speed or propeller blade angle, to achieve the desired power output. If the command includes parameters for switching device operating modes, the device automatically switches to the corresponding operating mode. For example, when switching from cruise mode to detection mode, the device activates the appropriate detection equipment and adjusts sensor operating parameters to meet the requirements of the detection mission. This process allows the unmanned vehicle to quickly respond to commands, adjusting its own state to complete various operational tasks, ensuring efficient and accurate mission execution in complex marine environments.

[0054] S4: Based on the progress deviation between the job status and the task sequence, performing feedback correction on the task sequence to obtain an optimized task sequence; In this embodiment, the feedback correction of the task sequence based on the progress deviation between the job status and the task sequence to obtain an optimized task sequence includes: Calculating the progress difference between the operation status and the progress planning standard of the task sequence to obtain a key delayed task node identifier; Performing perturbation source analysis on the operation state of the unmanned equipment based on the ocean environment perception data and the key delayed task node identifier to obtain a core induced delay factor; The task sequence is dynamically optimized based on the core induced delay factor to obtain an optimized task sequence.

[0055] Specifically, the operation status refers to the actual operation data set of the unmanned equipment at the current moment, including location coordinates, real-time progress of task execution, equipment operation parameters, etc.

[0056] The schedule planning standard for a task sequence refers to a task time node table preset based on a dynamic task diagram and equipment parameters, which includes the planned start time, planned completion time, planned execution path, etc. of each task node.

[0057] The key delayed task node identifier is the unique identifier of the task node that lags behind the planning standard and has a significant impact on the overall task process, which is screened out after the progress difference calculation.

[0058] The operating status data of unmanned equipment is collected in real time, and compared with the progress planning standards of the task sequence node by node. The degree of deviation of each task node is calculated through indicators such as time deviation rate and task completion deviation. Finally, by setting the deviation threshold, the task nodes that exceed the threshold are marked as critical lagging task nodes and a unique identifier is generated.

[0059] More specifically, if a strong current suddenly occurs in the target sea area, the speed of the unmanned ship is forced to reduce, resulting in the actual time to reach a certain sampling point far exceeding the planned time. The task node of the sampling point will be marked as a critical lagging node.

[0060] If the drone's sensor accuracy decreases due to sudden rainfall and the image acquisition task in a certain area is not completed as planned, the image acquisition node will be identified as a critical lagging node.

[0061] Furthermore, core delay-inducing factors are fundamental environmental or equipment factors that cause deviations in key lagging mission nodes, such as strong ocean currents, sudden strong winds, equipment power system failures, and obstacle avoidance.

[0062] The task execution period and geographic location corresponding to the key delayed task node identifier are temporally and spatially correlated with the ocean environment perception data of the same period. For example, the ocean current data and meteorological data of the area when a delayed sampling task is executed are extracted. Interference factors that may cause delays are then checked based on the associated data. If the task node is located in an area with turbulent ocean currents, analyze whether the ocean current speed exceeds the device's anti-current threshold; if there are obstacles in the task path, confirm whether the path is detoured due to obstacle avoidance. Finally, a causal relationship model is used to quantify the influence of each factor on the delay, and the 1-2 factors with the highest weight are selected as the core delay-inducing factors.

[0063] More specifically, if the key delay node is the terrain detection task of the underwater robot, and combined with the ocean environment perception data, it is found that the current speed in the area exceeds the robot's ability to resist the current, then the core induced delay factor is "strong current causing the speed to decrease."

[0064] If the drone inspection mission is delayed and environmental data shows that the wind speed during the mission period exceeds the drone's safe wind speed, the core delay-inducing factor is "sudden strong winds causing flight obstruction."

[0065] Furthermore, the optimized task sequence is a new task execution plan that is dynamically adjusted to adapt to the current environment and equipment status, including task sequence adjustment, time node update, path optimization, etc.

[0066] Adjust the order of tasks based on the core-induced delay factors. For example, if the core factor is "strong ocean currents causing delays in tasks in area A", then tasks in area B, which are less affected by the ocean currents, will be executed in advance, and tasks in area A will be executed after the ocean currents weaken. And recalculate the planned time for each task based on the degree of influence of the core factors. For example, if a task is delayed by 30 minutes due to ocean currents, the planned start time of subsequent tasks will be postponed accordingly, and buffer time will be reserved. If the delay factor is equipment performance limitations, spare equipment will be deployed for the lagging task nodes or the task load will be adjusted. Finally, the feasibility of executing the optimized task sequence in the current environment will be simulated to ensure that there are no new resource conflicts or environmental adaptation issues.

[0067] More specifically, if the core delay-inducing factor is "excessively large waves causing the unmanned vessel's sampling efficiency to decrease," optimizing the task sequence will adjust the sampling task to a period with smaller waves and shorten the duration of a single sampling.

[0068] If the path detour is delayed due to reef avoidance, the optimized task sequence will replan the shortest path to avoid the reef and update the time nodes of subsequent tasks.

[0069] In this embodiment, the dynamically optimizing the task sequence based on the core induced delay factor to obtain the optimized task sequence includes: Re-assigning the task execution priority of the operation state of the unmanned equipment based on the core induced delay factor and the operating condition parameters of the unmanned equipment to obtain the task structure of the unmanned equipment; Performing a risk analysis on the operation status based on the short-term prediction results of the marine environment perception data and the task structure to obtain a sequence operation report of the unmanned equipment; Verifying the node integrity of the task sequence in sequence based on the sequence operation report; Optimize the optimization task sequence of the unmanned equipment based on the verification result of the verification.

[0070] Specifically, the operating parameters of unmanned equipment refer to the real-time operating status indicators of the equipment, including the current battery life, propeller output power, equipment health, the deviation rate between the current speed and the rated speed, etc.

[0071] Combining core delay-inducing factors with operating parameters, a dynamic priority algorithm is used to sort and adjust the subtasks within the task sequence. For example, if the core delay factor is "strong ocean currents causing a decrease in ship speed," the priority of downstream tasks is increased, prioritizing tasks less affected by the current. If operating parameters indicate that a device's battery life is less than 20%, the priority of short-range tasks for that device is increased, and long-range tasks are temporarily stored or assigned to other devices. Conflicting tasks are sorted using a dual weighting of "environmental adaptability + device capability," with devices with strong anti-interference capabilities receiving priority for high-priority tasks. This priority redistribution creates a task structure that includes the order in which tasks are executed and the matching relationships between devices.

[0072] Furthermore, the short-term prediction results of marine environmental perception data are based on the environmental data of the next 3 to 6 hours generated by the pre-trained adversarial network, including current speed and direction prediction, wave height prediction, wind speed and direction prediction, vortex movement trajectory prediction, etc., and outliers need to be eliminated.

[0073] Based on short-term prediction results, the environmental risks of each node in the task structure are identified. For example, if a force 8 gale is predicted to occur in the target sea area during a certain task period, the task is marked as a high-risk delay. The matching degree between equipment capabilities and the environment is analyzed in combination with the task structure. For example, if an underwater task in the task structure is assigned to equipment with a current resistance capacity of 5 knots, but the current in the area is predicted to reach 6 knots, the risk of insufficient equipment capability is marked. A sequence operation report is also generated, which includes risk level, risk type, and risk impact duration.

[0074] Furthermore, the task sequence node is a key milestone in the task sequence, including the task start time, sampling point coordinates, data return node, equipment coordination intersection point, etc. Each node must meet the preset time and space constraints.

[0075] Verify the integrity of each node one by one in the order of task execution, check whether there are any missing nodes due to risks, and whether the spatiotemporal parameters meet the constraints; mark incomplete nodes, such as "Node A is missing sampling task" and "Node B time deviation exceeds the threshold by 10 minutes"; and output the verification results, which include a complete node list, missing node details, and analysis of the causes of deviation nodes.

[0076] Furthermore, for missing nodes, device resources are reallocated and task instructions are supplemented, for example, missing sampling tasks are assigned to adjacent idle devices and their navigation paths are adjusted; for deviation nodes, spatiotemporal parameters are corrected in combination with environmental prediction results, for example, the execution time of nodes affected by strong winds is postponed to a period when the wind speed decreases, and the start time of subsequent tasks is adjusted synchronously; for high-risk nodes, backup plans are activated, such as replacing task execution equipment and splitting complex nodes into sub-nodes, ultimately generating an optimized task sequence.

[0077] In this embodiment, the step of obtaining the short-term forecast statistical results includes: Predicting the ocean environment perception data based on a pre-trained adversarial network to obtain a comprehensive prediction result of the ocean environment perception data; Abnormal data in the comprehensive prediction result is eliminated to obtain a short-term prediction result of the prediction result.

[0078] Specifically, the pre-trained adversarial network is a deep learning model trained on marine environmental data. It consists of a generator and a discriminator. The generator is responsible for generating short-term future environmental forecasts based on historical and real-time environmental data, while the discriminator optimizes the generator's performance by comparing real data with predicted data, ultimately outputting a highly accurate comprehensive forecast.

[0079] The comprehensive prediction result is a set of future short-term ocean environment data output by the adversarial network, including predicted ocean current change trends, wave intensity, wind speed evolution, vortex movement trajectory, etc., covering the spatiotemporal dynamic characteristics of the target sea area.

[0080] Real-time ocean environment perception data is fed into a pre-trained adversarial network. The network uses built-in algorithms to extract spatiotemporal features from the data and perform predictions based on historical environmental changes. A generator generates preliminary predictions based on correlations with the ocean's physical environment. A discriminator then compares the distribution of real-world environmental data, revising the predictions and ultimately outputting a comprehensive prediction encompassing multiple environmental parameters.

[0081] Based on the physical constraints of the ocean environment and statistical analysis methods, the comprehensive forecast results are verified parameter by parameter. For example, if the ocean current speed at a certain forecast point far exceeds the historical maximum value of the same period in the sea area and there is no reasonable physical cause, it will be marked as abnormal data. The identified abnormal data is then removed from the comprehensive forecast results, and the missing parts in the data series are supplemented and corrected using interpolation methods, ultimately generating continuous and reliable short-term forecast results. For example, after eliminating the "level 10 instantaneous wind speed" outlier, the wind speed forecast trend in the surrounding sea area is revised to a reasonable "level 5-6 wind speed".

[0082] S5: Couple and solve the ocean environment perception data and the device parameters to obtain flight environment parameters of the unmanned device; In this embodiment, the coupled solution of the ocean environment perception data and the device parameters to obtain the flight environment parameters of the unmanned device includes: performing coupling calibration on the unmanned equipment based on the ocean environment perception data and the equipment parameters to obtain an equivalent projection component of the unmanned equipment; performing a maneuverability critical boundary calculation on the equivalent projection component to obtain a maximum heading correction margin of the unmanned equipment in the corresponding sea area; Performing flight envelope synthesis on the navigation trajectory of the unmanned device based on the maximum heading correction margin and the attitude oscillation attenuation period of the unmanned device to obtain a safe maneuvering boundary parameter of the unmanned device; The safety maneuvering boundary parameters are tuned by equipment state feedback to obtain flight environment parameters.

[0083] Specifically, equipment parameters refer to the physical properties of unmanned equipment, including mass, dimensions, propulsion power, wind resistance, flow resistance, and sensor installation location.

[0084] The equivalent projected component refers to the projection of the ocean environment force and the device's own power onto the unmanned device's motion coordinate system, including longitudinal, transverse, and vertical components. For example, the projection of the transverse force of the ocean current onto the device's coordinate system is the transverse equivalent projected component.

[0085] Based on ocean environment perception data, such as ocean currents and wind, and combined with parameters such as the unmanned vehicle's dimensions and center of gravity, mechanical modeling is used to calculate the projection of environmental forces on the device's motion axes. Simultaneously, the projection of the device's own power on each axis is calculated based on the device's propulsion power and power output characteristics. The environmental force projection is then vector-synthesized with the device's power projection to obtain equivalent projection components, completing the coupled calibration. This step achieves a quantitative conversion of the physical interactions between the environment and the device, providing a foundation for subsequent solutions.

[0086] Furthermore, the maximum heading correction margin is the maximum angle range within which the unmanned vehicle can safely adjust its heading in the current sea environment. This parameter is limited by both the intensity of environmental interference and the device's maneuverability.

[0087] Based on the lateral force and torque parameters in the equivalent projected components and the steering mechanism performance of the unmanned vehicle, a model of the vehicle's heading controllability was established. Simulations were performed to calculate the stability of the vehicle at various heading correction angles. When the heading correction angle exceeds a certain threshold, the vehicle may experience heading oscillation or loss of control due to environmental interference. By gradually testing critical states, the maximum heading adjustment angle, or maximum heading correction margin, was determined to ensure stability and effectively mitigate risks under current sea conditions.

[0088] Furthermore, the attitude oscillation attenuation period is the time period for the flight attitude of the unmanned equipment to change from oscillation to stability after being disturbed by the environment, reflecting the dynamic stability of the equipment.

[0089] The safe maneuvering boundary parameters include a set of parameters such as the maximum allowable speed, minimum turning radius, maximum crawling rate, and safe heading angle range of the unmanned equipment, forming the maneuvering capability boundary of the equipment in the current environment.

[0090] The maximum heading correction margin is used as the heading boundary constraint, and the time dimension constraint is determined in combination with the attitude oscillation decay period. Based on the unmanned vehicle's power system performance and environmental resistance parameters, the maximum safe speed under different headings is calculated. The minimum turning radius is calculated based on the maximum heading correction margin and the vehicle's steering performance. These parameters are integrated to create a three-dimensional maneuvering envelope of the unmanned vehicle's speed, heading, and attitude under current sea conditions. From this, the safe maneuvering boundary parameters are extracted, clarifying the device's safe motion limits.

[0091] Furthermore, the safe maneuvering boundary parameters refer to the boundary values ​​of the speed, turning radius, heading range, etc. obtained above.

[0092] The equipment status data refers to the current speed, heading angle, attitude angle, power output power, battery power and other operating status data transmitted back by the unmanned equipment in real time.

[0093] The flight environment parameters are the final output parameters including the safe speed range after environmental adaptability correction, dynamic heading adjustment threshold, maneuver response delay compensation value, etc., which are directly used to guide the trajectory adjustment of unmanned equipment.

[0094] Real-time data on the status of unmanned equipment is collected and compared with safe maneuvering boundary parameters. If the equipment's current speed approaches the maximum limit, or the attitude oscillation amplitude exceeds expectations, the feedback tuning mechanism is triggered. Based on real-time changes in ocean environment perception data, the safe maneuvering boundary parameters are dynamically corrected. For example, when ocean currents intensify, the maximum safe speed is reduced, and the adjustment interval corresponding to the attitude oscillation decay period is shortened. Through control algorithms or adaptive control strategies, the corrected boundary parameters are converted into specific flight environment parameters, ensuring that the parameters can adapt to changes in the ocean environment and equipment status in real time. The final output flight environment parameters can be directly used for trajectory optimization and collision avoidance decisions of unmanned equipment.

[0095] S6: Adjusting the navigation trajectory of the unmanned equipment based on the optimized task sequence and the flight environment parameters to obtain a synchronous collision avoidance path for the unmanned equipment.

[0096] In this embodiment, adjusting the navigation trajectory of the unmanned device based on the optimized task sequence and the flight environment parameters to obtain a synchronized collision avoidance path for the unmanned device includes: Performing multi-dimensional obstacle buffering on the navigation trajectory based on the flight environment parameters and the equipment parameters to obtain a dynamic threat envelope of the navigation trajectory; Performing a track deformation operation on the navigation trajectory based on the optimized task sequence and the dynamic threat envelope to obtain an adaptive track of the unmanned device; performing phase shift elimination on the navigation trajectory based on the ocean environment perception data and the adaptive track to obtain a conflict elimination path for the unmanned equipment; Based on the conflict resolution path and the flight environment parameters, a spatiotemporal synchronization verification operation is performed on the navigation trajectory to obtain a synchronous collision avoidance path for the unmanned equipment.

[0097] Specifically, the navigation trajectory is the initial driving path preset by the unmanned equipment, which contains a series of continuous spatial coordinate points and corresponding time information, and is the basis for subsequent adjustments.

[0098] The dynamic threat envelope is a multi-dimensional buffer area centered on the navigation trajectory, constructed after comprehensively considering flight environment parameters and equipment parameters. It is used to characterize the range of possible obstacle threats around the trajectory and will be updated in real time with environmental changes and equipment status.

[0099] Based on obstacle information from the flight environment parameters and the safety buffer distance requirements from the equipment parameters, a multi-dimensional obstacle buffering operation is performed on the flight trajectory. Spatially, this operation extends the trajectory horizontally and vertically to a certain extent. Temporally, a time buffer is reserved for early obstacle avoidance, taking into account the equipment's speed and the speed of obstacle movement. This operation integrates the threat ranges of all potential obstacles around the trajectory, forming a dynamic threat envelope that visually presents the danger zone surrounding the trajectory.

[0100] Furthermore, the adaptive trajectory is the path obtained by adjusting the initial navigation trajectory according to the optimized task sequence and dynamic threat envelope. It can adapt to the mission requirements and changes in environmental threats and has a certain degree of flexibility and adaptability.

[0101] Based on the current task priority and target location specified by the optimized task sequence, and in conjunction with the danger zones defined by the dynamic threat envelope, the navigation trajectory is deformed. A path planning algorithm is used to adjust the trajectory toward the task target specified by the optimized task sequence while avoiding the danger zones covered by the dynamic threat envelope. For example, if the optimized task sequence prioritizes a sampling task, and the original trajectory requires crossing a rocky area within the dynamic threat envelope, the trajectory is deformed by bending or detouring, allowing the new trajectory to avoid the rocky threat while still progressing toward the sampling task target, ultimately resulting in an adaptive trajectory for the unmanned device.

[0102] Furthermore, phase shift elimination is to eliminate the phase conflict between the adaptive track and the actual environment caused by the dynamic changes of the ocean environment, ensuring that the track is synchronized with the environmental changes.

[0103] The conflict resolution path is the path obtained after the phase shift resolution operation, which resolves the potential conflict between the adaptive track and the real-time ocean environment, such as the problem of the track deviating from the predetermined direction due to a sudden increase in ocean currents.

[0104] Based on the real-time updated ocean environment perception data, environmental parameters such as current speed and direction, wave intensity, etc., are analyzed for their impact on the adaptive track. When it is found that the adaptive track may conflict with the actual feasible path due to environmental changes, a phase shift elimination operation is performed on the navigation track. The offset effect of the environment on the device is calculated based on the ocean current vector data, and then a reverse compensation adjustment is made to each coordinate point of the adaptive track. For example, the track is offset by a certain angle toward the ocean current to offset the driving effect of the ocean current, ensuring that the device can reach the mission target as expected, and ultimately obtaining a conflict elimination path.

[0105] Furthermore, the spatiotemporal synchronization verification operation method verifies the feasibility of the conflict resolution path from both time and space dimensions, ensuring that the path matches the task node requirements in time and does not overlap with the dynamic threat envelope in space.

[0106] The synchronized collision avoidance path refers to the final path obtained after time-space synchronization verification. It can completely avoid all potential obstacles while meeting the time requirements of the optimized task sequence, thereby achieving safe and efficient navigation of unmanned equipment.

[0107] Based on the conflict resolution path and flight environment parameters, the navigation trajectory is verified in time and space. In the time dimension, combined with the maximum safe speed of the equipment in the flight environment parameters and the time node requirements of the optimized task sequence, it is verified whether the length of the conflict resolution path can allow the equipment to reach the mission target within the specified time. If there is a time deviation, the speed or local path is adjusted. In the spatial dimension, the conflict resolution path is checked again to see if it completely avoids the dynamic threat envelope. If there is still an overlapping area, the path is further fine-tuned. Through repeated time and space verification and adjustment, it is ensured that the path meets the requirements of safe collision avoidance and mission execution in both time and space, and ultimately a synchronous collision avoidance path for the unmanned equipment is obtained.

[0108] In the several embodiments provided by the present invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the module division is merely a logical function division, and other division methods may be used in actual implementation.

[0109] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed across multiple network elements. Some or all of the modules may be selected to achieve the purpose of the solution of this embodiment according to actual needs.

[0110] In addition, the functional modules in various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or hardware plus software functional modules.

[0111] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0112] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to achieve optimal results.

[0113] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A multi-modal collaborative operation method for unmanned maritime systems, characterized in that: The method comprises: S1: Acquire ocean environment perception data and generate dynamic mission maps to obtain the operation targets of unmanned equipment; S2: performing topological calculation on the operation target based on the dynamic task graph and the equipment parameters of the unmanned equipment to obtain a task sequence; S3: adjusting the operation status of the unmanned equipment based on the task sequence feedback; S4: Based on the progress deviation between the job status and the task sequence, performing feedback correction on the task sequence to obtain an optimized task sequence; S5: Couple and solve the ocean environment perception data and the device parameters to obtain flight environment parameters of the unmanned device; S6: Adjusting the navigation trajectory of the unmanned equipment based on the optimized task sequence and the flight environment parameters to obtain a synchronous collision avoidance path for the unmanned equipment.

2. A multi-modal collaborative operation method for unmanned maritime systems according to claim 1, characterized in that: The acquiring of ocean environment perception data and generating of a dynamic task graph includes: Perform multi-source data collaborative collection operations on the preset ocean stereo observation network to obtain ocean environment perception data; Based on the marine environment perception data, collaborative situational spatiotemporal modeling is performed on the target sea area corresponding to the unmanned equipment to obtain a dynamic task map.

3. The multi-modal collaborative operation method of a maritime unmanned system according to claim 1, characterized in that: The obtaining of the operation target of the unmanned equipment includes: Performing target allocation for the unmanned equipment based on the dynamic task graph to generate a target allocation plan; The operation target of the unmanned equipment is determined based on the target allocation plan.

4. The multi-modal collaborative operation method of an unmanned marine system according to claim 2, characterized in that: The topological calculation of the operation target based on the dynamic task graph and the equipment parameters of the unmanned equipment to obtain a task sequence includes: Building tasks for the operation targets based on the dynamic task graph and the equipment parameters of the unmanned equipment to obtain an operation target list for the unmanned equipment; Based on the fluid vortex conduction characteristics of the target sea area, the operation target list is divided into levels to obtain hierarchical tasks for the unmanned equipment; The mutually exclusive dynamic resource points in the hierarchical tasks are resolved to obtain the task sequence of the unmanned equipment.

5. The multi-modal collaborative operation method of a maritime unmanned system according to claim 1, characterized in that: The adjusting the operating state of the unmanned equipment based on the task sequence feedback includes: Monitoring the operational control state parameters of the unmanned equipment based on the task sequence to obtain closed-loop control core parameters; Reconstruct and solve the control instructions of the unmanned equipment based on the closed-loop control core parameters to obtain dynamic allocation instructions for the unmanned equipment; Adjust the operating status of the unmanned equipment based on the dynamic allocation instruction.

6. The multi-modal collaborative operation method of an unmanned marine system according to claim 1, characterized in that: The step of performing feedback correction on the task sequence based on the progress deviation between the job status and the task sequence to obtain an optimized task sequence includes: Calculating the progress difference between the operation status and the progress planning standard of the task sequence to obtain a key delayed task node identifier; Performing perturbation source analysis on the operation state of the unmanned equipment based on the ocean environment perception data and the key delayed task node identifier to obtain a core induced delay factor; The task sequence is dynamically optimized based on the core induced delay factor to obtain an optimized task sequence.

7. The multi-modal collaborative operation method of an unmanned marine system according to claim 6, characterized in that: The dynamically optimizing the task sequence based on the core induced delay factor to obtain an optimized task sequence includes: Re-assigning the task execution priority of the operation state of the unmanned equipment based on the core induced delay factor and the operating condition parameters of the unmanned equipment to obtain the task structure of the unmanned equipment; Performing a risk analysis on the operation status based on the short-term prediction results of the marine environment perception data and the task structure to obtain a sequence operation report of the unmanned equipment; Verifying the node integrity of the task sequence in sequence based on the sequence operation report; Optimize the optimization task sequence of the unmanned equipment based on the verification result of the verification.

8. The multi-modal collaborative operation method of an unmanned marine system according to claim 7, characterized in that: The steps of obtaining the short-term forecast statistical results include: Predicting the ocean environment perception data based on a pre-trained adversarial network to obtain a comprehensive prediction result of the ocean environment perception data; Abnormal data in the comprehensive prediction result is eliminated to obtain a short-term prediction result of the prediction result.

9. The multi-modal collaborative operation method of an unmanned marine system according to claim 1, characterized in that: The coupled solution of the ocean environment perception data and the device parameters to obtain the flight environment parameters of the unmanned device includes: performing coupling calibration on the unmanned equipment based on the ocean environment perception data and the equipment parameters to obtain an equivalent projection component of the unmanned equipment; performing a maneuverability critical boundary calculation on the equivalent projection component to obtain a maximum heading correction margin of the unmanned equipment in the corresponding sea area; Performing flight envelope synthesis on the navigation trajectory of the unmanned device based on the maximum heading correction margin and the attitude oscillation attenuation period of the unmanned device to obtain a safe maneuvering boundary parameter of the unmanned device; The safety maneuvering boundary parameters are tuned by equipment state feedback to obtain flight environment parameters.

10. The multi-modal collaborative operation method of an unmanned marine system according to claim 1, characterized in that: The step of adjusting the navigation trajectory of the unmanned equipment based on the optimized task sequence and the flight environment parameters to obtain a synchronous collision avoidance path for the unmanned equipment includes: Performing multi-dimensional obstacle buffering on the navigation trajectory based on the flight environment parameters and the equipment parameters to obtain a dynamic threat envelope of the navigation trajectory; Performing a track deformation operation on the navigation trajectory based on the optimized task sequence and the dynamic threat envelope to obtain an adaptive track of the unmanned device; Performing phase shift elimination on the navigation trajectory based on the ocean environment perception data and the adaptive track to obtain a conflict elimination path for the unmanned equipment; Based on the conflict resolution path and the flight environment parameters, a spatiotemporal synchronization verification operation is performed on the navigation trajectory to obtain a synchronous collision avoidance path for the unmanned equipment.

Images