Heterogeneous bus scheduling method based on open avionics system of multipurpose unmanned aerial vehicle

Abstract

The application discloses a heterogeneous bus scheduling method based on a multipurpose unmanned aerial vehicle open avionics system and belongs to the technical field of unmanned aerial vehicle avionics system design, which comprises the following steps: in the flight control bus scheduling, each end system on the flight control bus adopts a static deterministic scheduling strategy of a local micro time slot scheduling table, strict time sequence deterministic control of flight key data is realized at the sending side of the end system, and thus any influence of task load change on the flight control communication time sequence is avoided; in the task bus scheduling, a dynamic reconfiguration mechanism of traffic shaping and event triggering with key sensing is introduced, and a hybrid scheduling strategy of a static scheduling window and a dynamic scheduling window is used for online reconfiguration, so that the task bus can fully utilize bandwidth resources and adapt to the dynamic change of the task load of each load end system, and thus optimal balance between performance and efficiency is realized at the system level.

Description

Technical Field

[0001] This invention belongs to the field of avionics system design technology for unmanned aerial vehicles (UAVs), and specifically relates to a heterogeneous bus scheduling method based on an open avionics system for multi-purpose UAVs. Background Technology

[0002] As the complexity of unmanned aerial vehicle (UAV) missions increases, their avionics systems need to simultaneously handle two vastly different communication requirements: one is core control commands crucial for flight safety (such as navigation and servo control), which demands extremely high determinism and reliability; the other is diverse mission payload data (such as electro-optical reconnaissance video and radar point clouds), which requires high bandwidth and dynamic flexibility. Traditional avionics network architectures, when addressing these hybrid critical communication needs, primarily employ the following three typical solutions and their inherent limitations: Single-event triggered bus architecture (such as using a standard CAN bus to carry all data): This type of bus uses a priority-based arbitration mechanism, and its network latency and jitter become unpredictable as the load increases. While high-priority messages are sent first, low-priority messages may face indefinite waiting (starvation). When critical control commands coexist with sudden large volumes of payload data, the worst-case response time of the control commands will deteriorate drastically and even become uncontrollable, failing to meet the stringent deterministic requirements of the flight control system and posing a serious safety hazard.

[0003] A single static time-triggered bus architecture (such as a pure TTEthernet with global synchronization): This architecture provides strict time slots for all communications through an offline-generated global schedule table, achieving excellent determinism. However, its schedule table is fixed and periodically repeating, which cannot effectively adapt to the dynamic and sudden communication needs of the task system. Reserving fixed bandwidth for possible bursts of data leads to low bandwidth utilization; if not reserved, dynamic tasks cannot be transmitted in a timely manner, greatly limiting the system's ability to handle diverse and non-periodic task loads.

[0004] To avoid interference between different types of services, some existing designs configure completely independent physical communication networks for the flight control system and the mission system. While this approach achieves service isolation at the physical level, it also introduces problems such as resource redundancy, increased system weight, power consumption, and cost. Furthermore, data interaction between different physical networks typically relies on dedicated gateway devices, further increasing system complexity and communication latency, which is detrimental to the miniaturization and integration of unmanned aerial vehicle platforms.

[0005] In summary, existing technological solutions, while simultaneously pursuing the dual goals of strong determinism and high-bandwidth flexibility, have fallen into a dilemma: pursuing determinism sacrifices flexibility and efficiency, while pursuing flexibility jeopardizes safety and reliability. This contradiction has become a key bottleneck restricting the development of avionics systems for multi-purpose, high-performance unmanned aerial vehicles. Therefore, how to achieve logical isolation and differentiated assurance between flight-critical communications and mission payload communications under a unified physical interconnection network has become an urgent technical problem to be solved. Summary of the Invention

[0006] To address the aforementioned problems in the existing technology, this invention provides a heterogeneous bus scheduling method based on an open avionics system for multi-purpose unmanned aerial vehicles (UAVs). The technical problem to be solved by this invention is achieved through the following technical solution: This invention provides a heterogeneous bus scheduling method based on an open avionics system for multi-purpose unmanned aerial vehicles (UAVs). The method is applied to an open avionics system for multi-purpose UAVs, which includes a flight control bus subsystem, an integrated processing module, a mission bus subsystem, multiple payload systems on the mission bus, and navigation, servo, and power supply systems on the flight control bus. The corresponding heterogeneous bus scheduling method includes: The flight control bus scheduling process includes: in the offline phase, each terminal system on the flight control bus generates its own local micro-hour slot scheduling table through the flight control bus subsystem; in the operational phase, each terminal system on the flight control bus communicates and schedules with the integrated processing module according to the corresponding local micro-hour slot scheduling table and local clock. The task bus scheduling process includes: for the task data streams requesting access from each payload system on the task bus, the task bus subsystem introduces a dynamic reconfiguration mechanism that combines critical awareness traffic shaping and event triggering to assess system load and task requirements in real time, proactively adjust resource allocation and communication scheduling strategies, and use a hybrid scheduling strategy of static and dynamic scheduling windows for online reconfiguration. Under the premise of not affecting the security of flight control bus communication, the remaining bandwidth resources of the system are adaptively optimized.

[0007] The beneficial effects of this invention are: This invention proposes a heterogeneous bus scheduling method based on an open avionics system for multi-purpose unmanned aerial vehicles (UAVs). This method breaks through the limitations of traditional single scheduling strategies. Through an innovative heterogeneous bus and heterogeneous scheduling architecture, it simultaneously achieves millisecond / microsecond-level strong deterministic guarantees for flight control communication and high-bandwidth, high-efficiency, and dynamically adaptable transmission for mission payload communication within an integrated system. This maximizes system mission performance and resource utilization while ensuring the highest level of safety. Specifically, in the flight control bus scheduling of this invention, each end system on the flight control bus adopts a static deterministic scheduling strategy using a local micro-hour slot scheduling table. This achieves strict timing deterministic control of critical flight data at the end system sending side, thereby avoiding any impact of mission load changes on flight control communication timing. In the mission bus scheduling, a dynamic reconfiguration mechanism incorporating criticality awareness flow shaping and event triggering is adopted. A hybrid scheduling strategy using static and dynamic scheduling windows is employed for online reconfiguration. The mission bus can fully utilize bandwidth resources to adapt to the dynamic changes in mission payloads of each end system, thus achieving an optimal balance between performance and efficiency at the system level. This invention achieves flight control bus scheduling and mission bus scheduling within the same open avionics system for a multi-purpose unmanned aerial vehicle (UAV), resolving the core contradiction of traditional solutions that cannot simultaneously accommodate both scheduling methods. This invention deeply decouples the flight control and mission systems in terms of logic and scheduling. Adding or replacing payload systems only requires adjustments to the mission scheduling strategy, without affecting the already certified flight control communication architecture. This reduces the complexity and risk of system upgrades, and improves overall reliability and maintainability.

[0008] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0009] Figure 1 This is a flowchart illustrating a heterogeneous bus scheduling method based on an open avionics system for multi-purpose unmanned aerial vehicles provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of the open avionics system for a multi-purpose unmanned aerial vehicle provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the flight control bus scheduling process provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the task bus scheduling process provided in an embodiment of the present invention. Detailed Implementation

[0010] The present invention will be further described in detail below with reference to specific embodiments, but the implementation of the present invention is not limited thereto.

[0011] Please see Figure 1 This invention provides a heterogeneous bus scheduling method based on an open avionics system for multi-purpose unmanned aerial vehicles (UAVs), applicable to, for example... Figure 2The open avionics system of the multi-purpose UAV shown includes a flight control bus subsystem, an integrated processing module, a mission bus subsystem, multiple payload end systems on the mission bus, and navigation end systems, servo end systems, and power supply and distribution end systems on the flight control bus; the corresponding heterogeneous bus scheduling method includes: The flight control bus scheduling process includes: in the offline phase, each terminal system on the flight control bus generates its own local micro-hour slot scheduling table through the flight control bus subsystem; in the operational phase, each terminal system on the flight control bus communicates and schedules with the integrated processing module according to the corresponding local micro-hour slot scheduling table and local clock. The task bus scheduling process includes: for the task data streams requesting access from each payload system on the task bus, the task bus subsystem introduces a dynamic reconfiguration mechanism that combines critical awareness traffic shaping and event triggering to assess system load and task requirements in real time, proactively adjust resource allocation and communication scheduling strategies, and use a hybrid scheduling strategy of static and dynamic scheduling windows for online reconfiguration. Under the premise of not affecting the security of flight control bus communication, the remaining bandwidth resources of the system are adaptively optimized.

[0012] In the open avionics system of multi-purpose unmanned aerial vehicles (UAVs), the integrated processing module serves as a heterogeneous communication hub, internally integrating the flight control bus interface and the mission bus interface, and realizing cross-bus time base mapping. Internally, the integrated processing module uses a unified clock management unit to map the static time slot cycle of the flight control bus as a global reference to the scheduling window of the mission bus, thereby achieving time base alignment and deterministic data exchange across protocol buses.

[0013] In the open avionics system of multi-purpose unmanned aerial vehicles (UAVs), data exchange between various terminal systems and the integrated processing module is achieved through an interconnection network. The interconnection network adopts a heterogeneous dual-bus architecture, specifically divided into: Flight control bus: Used to transmit critical flight data between the navigation system, servo system, power supply system and integrated processing module; this type of data is characterized by being fixed, highly periodic, and having extremely low tolerance for delay and jitter. Task bus: Used to transmit task payload data between various payload end systems and the integrated processing module; this type of data is characterized by heterogeneity, high bandwidth requirements, potential bursts, and diverse real-time requirements.

[0014] Based on a unified physical interconnection network, two fundamentally different communication subsystems, flight control bus and mission bus, are constructed through logical division. In response to their differentiated needs, two scheduling mechanisms are collaboratively designed and applied within the same open avionics system architecture of multi-purpose UAVs, thus forming a heterogeneous collaborative scheduling method for heterogeneous buses, rather than a simple application of a single bus or a single scheduling algorithm.

[0015] To address the core requirements of strong determinism and extremely low jitter in flight control bus communication, this invention employs a static deterministic scheduling strategy based on a local time-triggered table for each end system. This strategy differs from global time-triggered scheduling mechanisms that rely on global time synchronization and a unified scheduling table across the network. Instead, it independently generates and executes a local micro-time slot scheduling table within each end system. This local micro-time slot scheduling table uses only the end system's own local clock as a reference, and does not require time slot alignment or synchronous execution between different end systems or network devices in absolute time. This achieves strict deterministic timing control of critical flight data at the end system transmission side, thereby preventing any impact of mission load changes on flight control communication timing.

[0016] To address the high bandwidth and dynamic uncertainty issues faced by the mission bus when supporting diverse and reconfigurable payloads, this invention does not employ the traditional passive event-triggered or fixed-priority scheduling method for mission bus scheduling. Instead, it introduces a key-aware proactive traffic shaping and event-triggered dynamic reconfiguration mechanism to assess system load and mission requirements in real time, proactively adjust resource allocation and scheduling strategies, and utilize a hybrid scheduling strategy of static and dynamic scheduling windows for online reconfiguration. This enables adaptive optimization of the system's remaining bandwidth resources for mission payload communication without compromising flight control communication security.

[0017] Next, we will introduce in detail the implementation process of flight control bus scheduling and mission bus scheduling.

[0018] In the flight control bus subsystem of this invention, such as Figure 2 As shown, the system includes a local clock, a local micro-hour slot scheduler, a flow shaper, a flight control scheduler, and a physical layer transmission queue. During the system offline phase, an offline, periodic local micro-hour slot scheduler is generated locally on each end system based on the communication contracts of all flight-critical virtual links. During system operation, each end system on the flight control bus strictly adheres to the local micro-hour slot scheduler, triggering message transmission at pre-allocated, conflict-free, and precise times. This completely avoids the competition and uncertainties caused by bus arbitration, thus ensuring from the source that the transmission delay of all flight control commands is constant and jitter is bounded.

[0019] In the offline phase of flight control bus scheduling in this embodiment of the invention, each terminal system on the flight control bus generates its own local micro-hour slot scheduling table through the flight control bus subsystem, such as... Figure 3As shown, the process includes: obtaining the key communication contract parameters of the virtual links corresponding to each end system on the flight control bus; the key communication contract parameters of each virtual link include the bandwidth allocation interval, the maximum frame length, and the maximum transmit jitter allowed by the system requirements; calculating the minimum time slot length t_slot that satisfies the complete transmission of the maximum frame in the virtual link based on the maximum frame length, and using the least common multiple of the bandwidth allocation intervals of all virtual links as the total scheduling period; generating a micro-time slot sequence based on the total scheduling period and the minimum time slot length; specifically, discretizing the communication time axis into equal-length, continuous, and non-overlapping micro-time slot sequences in units of t_slot, where the minimum time slot length t_slot is uniquely determined by the contract parameters to ensure that any micro-time slot can carry the maximum frame data of the virtual link; allocating transmit time slots for each virtual link within the total scheduling period based on the bandwidth allocation interval and the micro-time slot sequence; generating a local micro-time slot scheduling table based on all transmit time slot allocation results, and storing the local micro-time slot scheduling table in the non-volatile memory corresponding to each end system on the flight control bus.

[0020] In the offline phase of flight control bus scheduling, the constraints that each terminal system in the flight control bus subsystem must follow when allocating transmission time slots for each virtual link on the flight control bus within the total scheduling cycle include: periodicity constraint, conflict-free constraint, and jitter constraint. The periodicity constraint is that the interval between adjacent transmission time slots allocated to each virtual link is equal to the corresponding bandwidth allocation interval or an integer fraction of the bandwidth allocation interval. The conflict-free constraint is that within any single micro-hour slot in the micro-hour slot sequence, at most one transmission time slot is allocated to a virtual link to ensure no bus contention. The jitter constraint is that the deviation between the actual transmission interval and the bandwidth allocation interval of each virtual link is less than the maximum transmission jitter allowed by the system; in the optimal case, zero jitter can be achieved for the virtual link.

[0021] In the operational phase of the flight control bus scheduling in this embodiment of the invention, each terminal system on the flight control bus communicates and schedules with the integrated processing module according to its corresponding local micro-hour slot scheduling table and local clock, such as... Figure 3 As shown, this includes: starting a local cyclic counter; the flight control scheduler adjusting the phase of the local cyclic counter based on the received absolute time alignment pulse; and tracking the current micro-hour slot of each virtual link based on the local clock and the adjusted local cyclic counter. The absolute time alignment pulse is periodically broadcast to the flight control bus by the integrated processing module. When entering the next micro-hour slot, the flight control scheduler queries the local micro-hour slot scheduling table for the transmission slot allocation result corresponding to the current micro-hour slot, and stores the frame to be transmitted in the physical layer transmission queue according to the transmission slot allocation result, so as to realize the communication scheduling between the various end systems on the flight control bus and the integrated processing module through the physical layer transmission queue. More specifically: After the open avionics system of a multi-purpose UAV is powered on, each terminal system on the flight control bus performs deterministic communication scheduling based on the locally stored micro-hour slot scheduling table and the local clock. Specifically, this includes the following steps: Local Clock Driven and Macroscopic Synchronization Calibration: At the microscopic execution level, the scheduling process is driven solely by the local high-precision clocks of each end system on the flight control bus, ensuring extremely low transmission jitter. During system initialization, a pre-defined local micro-timeslot scheduling table is loaded, and a local cyclic counter synchronized with the local micro-timeslots is started. To address the issue of accumulated clock drift caused by the aging of physical crystal oscillators in each end system during long-term operation, a macroscopic synchronization calibration mechanism is introduced: the integrated processing module, leveraging its role as a heterogeneous communication hub, periodically broadcasts absolute time alignment pulses to each end system on the flight control bus, using the macro-period of the flight control bus as a reference. Upon receiving these absolute time alignment pulses, the flight control scheduler of each end system performs phase fine-tuning or reset of its local cyclic counter to eliminate the accumulated drift error of the local high-precision clock, thereby ensuring that the entire distributed network does not experience time slot overlap or misalignment conflicts during long-endurance flight.

[0022] Strict table-triggered execution: The flight controller scheduler continuously tracks the current micro-slot based on the local clock and local loop counter. The scheduler's operation logic is entirely driven by the local micro-slot scheduling table. When entering a new micro-slot, the scheduler immediately queries the local micro-slot scheduling table for the slot's allocation status. If the micro-slot is allocated to a virtual link, the scheduler unconditionally and immediately delivers the frames to be transmitted for that virtual link to the physical layer transmission queue. This enables communication scheduling between the various systems on the flight control bus and the integrated processing module through the physical layer transmission queue. This process involves no priority arbitration and no dynamic decision delay.

[0023] Traffic policing and shaping: Upstream of the flight controller scheduler, a traffic shaper is installed. This traffic shaper is used for rate constraint and buffer management of transmission requests. Its function can be implemented by existing network interface controllers or communication middleware of various end systems. The traffic shaper ensures that the rate of frame transmission requests from the application layer for virtual links does not exceed the contracted bandwidth specified by the bandwidth allocation interval. Any early-arriving frames will be buffered in the traffic shaper and wait until the next transmission slot of their respective virtual links arrives, thus ensuring the determinism of bus load and the effectiveness of the scheduling table from the source.

[0024] Furthermore, the task bus subsystem of the present invention, as follows: Figure 2As shown, the system includes a criticality-aware proactive traffic shaper, a scheduling ready queue, a hybrid scheduler, and a task bus controller. The hybrid scheduler comprises a static scheduling window, a dynamic scheduling window, and a dynamic decision engine. By introducing a criticality-aware traffic shaping and event-triggered dynamic reconfiguration mechanism, the system assesses system load and task requirements in real time, proactively adjusts resource allocation and scheduling strategies, and utilizes static and dynamic scheduling windows, along with a multi-objective optimization decision model, to dynamically optimize and allocate remaining bandwidth resources while strictly ensuring the real-time performance of high-criticality task data streams.

[0025] In this invention's embodiments, during task bus scheduling, a dynamic reconfiguration mechanism based on criticality-aware traffic shaping and event triggering is introduced to assess system load and task requirements in real time, proactively adjusting resource allocation and communication scheduling strategies. Figure 4 As shown, the process includes: acquiring the mission data streams requesting admission from each payload system on the mission bus; each mission data stream has multi-dimensional attributes including data criticality level, data characteristic pattern, bandwidth requirement estimation, and transmission deadline constraints; a criticality-aware active flow shaper calculates the dynamic admission weight corresponding to each payload system's requesting admission mission data stream based on a preset dynamic weight function; the dynamic weight function takes the criticality level in the mission data stream, the current real-time load rate of the mission bus, the historical requirement fulfillment rate of the mission data stream, and the current flight phase status of the aircraft as input variables, and outputs a comprehensive evaluation of the mission data stream. The system dynamically assigns admission weights to task data streams under the current system state. Based on these weights, it stores the task data streams requesting admission from each load system in a scheduling ready queue, enabling communication scheduling between the load systems and the integrated processing module. When the criticality-aware active traffic shaper determines that the system is in a resource-strained early warning event, it stores the task data streams requesting admission from each load system in a scheduling ready queue based on the criticality level, bandwidth requirement estimation, data characteristic patterns, and dynamic admission weights, enabling communication scheduling between the load systems and the integrated processing module. More specifically: The task bus scheduling object of this invention is the task data stream from different payload systems on the task bus. Each task data stream has multi-dimensional attributes, including: data criticality level, data characteristic pattern, bandwidth requirement estimate, and transmission deadline constraint. The criticality level describes the degree of impact of the task data stream on system security or mission success, and can be represented by discrete levels or numerical forms, categorized into high-criticality, medium-criticality, and low-criticality task data streams, or mapped to numerical levels 3, 2, and 1. For example, a high-criticality task data stream refers to a data stream whose communication delay or loss may directly affect flight safety or mission success or failure, such as mission abort commands or payload safety control commands; a low-criticality task data stream refers to a data stream whose communication delay or loss will not directly affect flight safety or core missions, such as log data or non-real-time monitoring data. The data characteristic pattern characterizes the service form and transmission behavior characteristics of the task data stream, including periodic, bursty, continuous stream, large frame throughput, or small packet control. Bandwidth requirement estimation characterizes the average bandwidth, peak bandwidth, or bandwidth variation trend required by the task data stream within the current scheduling cycle or prediction time window. Transmission deadline constraints limit the latest possible completion time for the task data stream within the current task context. These multi-dimensional attributes collectively form the basis for the task bus's decisions regarding admission control, scheduling, resource allocation, and online reconfiguration.

[0026] Before mission data streams from different payload modules enter the scheduling ready queue, a criticality-aware active flow shaper is set up to implement admission control. Unlike the contractual flow shaper in the flight control bus, the criticality-aware active flow shaper is used not only for rate constraints but also for triggering admission control and scheduling strategies based on criticality level and system load status. The system maintains a dynamic criticality baseline, which can be represented numerically or in terms of levels, such as initially set to "low criticality" or the value 1, to represent the lowest acceptable mission criticality level for the current system.

[0027] For a requesting task data stream, the criticality-aware active traffic shaper calculates its dynamic admission weight based on a preset dynamic weight function. This invention's dynamic weight function takes the criticality level of the task data stream, the real-time load rate of the current task bus, the historical demand fulfillment rate of the task data stream, and the current flight phase state of the system as input variables. The output is a dynamic admission weight used to comprehensively evaluate the admission priority of the task data stream under the current system state. The dynamic weight function is a monotonic function incorporating task environment awareness; its output increases with the criticality level of the task data stream and adaptively adjusts with the system load state. The flight phase state variable acts as a scenario multiplier factor in the function, adjusting the weight amplification coefficient of specific task data streams according to changes in the environment perceived by the system. For example, when the aircraft is in a high-risk or high-dynamic flight phase, the core monitoring data stream of that phase is given the highest priority dynamic admission weight. When the criticality level is higher than the dynamic criticality benchmark maintained by the current system, the corresponding dynamic admission weight receives a significant gain, thus enabling high-criticality task data streams to gain priority admission and enter the scheduling ready queue.

[0028] When the criticality-aware proactive traffic shaper determines that system communication resources are about to become strained through a load prediction model, or predicts through runtime monitoring mechanisms that high-criticality task data flows that have requested access may not be able to meet their transmission deadline constraints, a resource strain warning event is triggered. The criticality-aware proactive traffic shaper is a shaping module with the ability to identify and control the criticality of data flows. It not only constrains the rate of data flows but also proactively determines the access, delay, or suppression strategies for data flows based on their criticality level and system operating status. After a resource strain warning event is triggered, the system proactively and temporarily suppresses or delays the access of low-criticality task data flows with lower dynamic access weights, based on the criticality level, bandwidth requirement estimation, data characteristic patterns, and dynamic access weights of each candidate task data flow. This reserves and ensures communication resources for high-criticality task data flows that have already requested access or are about to request access, achieving a shift from passively discarding data after congestion to proactively preventing congestion before it occurs.

[0029] In the task bus scheduling of this invention, the criticality-aware active traffic shaper determines that the system is in a resource shortage warning event, including: the criticality-aware active traffic shaper determines that the system communication resources are about to become strained through a load prediction model, or predicts through a runtime monitoring mechanism that the data streams of high-critical-level tasks that have been admitted may not be able to meet their transmission deadline constraints.

[0030] Furthermore, in the task bus scheduling, this embodiment of the invention utilizes a hybrid scheduling strategy combining static and dynamic scheduling windows for online reconfiguration, such as... Figure 4As shown, the process includes: Within each scheduling cycle of the task bus, a static scheduling window is defined, and the remaining bandwidth outside the static window constitutes a dynamic scheduling window. The static scheduling window is used to transmit task data streams with strict periodicity and high deterministic requirements, such as power-on / off commands and operating mode switching commands for each load system. The dynamic scheduling window is used to transmit non-periodic, bursty, or dynamically changing bandwidth-demanding task data streams. The time position of the dynamic scheduling window is fixed with the static window, but its available bandwidth and internal scheduling order dynamically change with system load and task status. Within the static scheduling window, a static scheduling table is generated, and the data packet transmission sequence for the next scheduling cycle is calculated based on the static scheduling table. The data packet transmission sequence is then executed by the task bus controller. The static scheduling table is generated offline based on the scheduling cycle of the task data stream, providing deterministic guarantees similar to a time-triggered mechanism. Within the dynamic scheduling window, based on the task bus's multi-objective optimization model, the data packet transmission sequence for the next scheduling cycle is dynamically calculated, and the data packet transmission sequence is then executed by the task bus controller. The task bus's multi-objective optimization model aims to ensure both system security and transmission efficiency.

[0031] Within the dynamic scheduling window of this embodiment of the invention, a dynamic decision engine runs. The scheduling decisions of this dynamic decision engine are based on a multi-objective optimization model of the task bus. The optimization objectives of the multi-objective optimization model of the task bus include a primary objective, a core performance objective, a quality of service objective, and a secondary fairness objective. The primary objective is to maximize the on-time delivery rate of all high-criticality task data streams in the system. The core performance objective is to maximize the total bandwidth utilization of the task bus within the current scheduling cycle. The quality of service objective is to minimize the overall average transmission delay and latency jitter of all task data streams. The secondary fairness objective is to ensure fairness in service access among low-criticality task data streams, while satisfying the primary objective, core performance objective, and quality of service objective. More specifically: This invention's dynamic decision engine, based on the multidimensional attributes of task data streams in the current ready queue, the urgency of each task data stream's remaining deadline, and the system's historical operating status, employs a lightweight heuristic scheduling algorithm to dynamically calculate and output the data packet transmission sequence for the next scheduling cycle. This heuristic scheduling algorithm can, for example, be built upon the Earliest Deadline First (EDF) algorithm, and further incorporate a task data stream criticality weight factor, transforming the aforementioned multi-objective optimization model into an executable scheduling priority calculation rule: when deadlines are close or conflicting, high-criticality task data streams receive higher scheduling priority than low-criticality task data streams, thereby improving the overall scheduling efficiency of the task bus while ensuring the real-time performance of critical tasks. The hybrid scheduler controls the task bus controller to execute corresponding data transmission operations based on the data packet transmission sequence.

[0032] Furthermore, the task bus scheduling process in this embodiment of the invention further includes: when the system detects that the criticality level of a newly requested task data stream is higher than the current dynamic criticality benchmark, or higher than the criticality level of a task data stream currently being scheduled in the system, the newly requested task data stream is reconfigured online based on data characteristic patterns, bandwidth requirement estimation, transmission deadline constraints, and the remaining bandwidth status of the current dynamic scheduling window. More specifically: When the system detects the activation of a new payload module, the declaration of a new task data stream, or the generation of a new service mode switching request by an existing payload module, i.e., the system detects a new request for access to the task data stream, if the criticality level of the new request for access to the task data stream is higher than the current dynamic criticality benchmark, or higher than the criticality level of some task data streams currently being scheduled in the system, the task scheduler immediately and proactively triggers a global state reassessment within the task bus scheduling scope. This reassessment not only compares the criticality level of the new request for access to the task data stream, but also combines the data characteristic pattern, bandwidth requirement estimation, transmission deadline constraint, and the remaining bandwidth status of the current dynamic scheduling window of the new request for access to the task data stream to reconfigure online the bandwidth allocation ratio of existing task data streams, the scheduling ready queue order, the transmission order within the dynamic scheduling window, and related scheduling parameters. This ensures that the new request for access to the high-criticality task data stream can seamlessly access the system without degrading the performance of existing critical tasks of the same or higher criticality level.

[0033] Furthermore, the task bus scheduling process in this embodiment of the invention further includes: when the system detects that a requested-admission task data stream may miss its transmission deadline constraint, the requested-admission task data stream is reconfigured online based on its criticality level, data characteristic pattern, bandwidth requirement estimate, and the urgency of the remaining deadline. More specifically: The task scheduler continuously monitors the remaining deadlines of high-criticality task data streams that have requested access. If it determines that a high-criticality task data stream may miss its transmission deadline under the current scheduling state, a deadline warning event is triggered. This deadline warning event is a runtime correction mechanism for accessed task data streams. After it is triggered, the dynamic decision engine performs targeted scheduling intervention based on the criticality level, data characteristic patterns, bandwidth requirement estimates, and the urgency of the remaining deadline of the requested access task data stream. Specifically, the scheduling intervention measures include: proactively increasing the scheduling priority of the requested access task data stream, or, in extreme cases, preempting its execution resources to ensure that the criticality level constraints are not violated, thereby turning potential failure impacts from post-event handling into pre-event avoidance.

[0034] In summary, the heterogeneous bus scheduling method based on the open avionics system of multi-purpose UAVs proposed in this invention breaks through the limitations of traditional single scheduling strategies. Through an innovative heterogeneous bus and heterogeneous scheduling architecture, it simultaneously achieves millisecond / microsecond-level strong deterministic guarantee of flight control communication and high-bandwidth, high-efficiency, and dynamically adaptable transmission of mission payload communication within an integrated system. This maximizes system mission efficiency and resource utilization while ensuring the highest level of safety. Specifically, in the flight control bus scheduling of this invention, each end system on the flight control bus adopts a static deterministic scheduling strategy with a local micro-hour slot scheduling table. This achieves strict timing deterministic control of critical flight data at the end system sending side, thereby avoiding any impact of mission load changes on flight control communication timing. In the mission bus scheduling, a dynamic reconfiguration mechanism with criticality awareness flow shaping and event triggering is adopted. A hybrid scheduling strategy of static and dynamic scheduling windows is used for online reconfiguration, enabling the mission bus to fully utilize bandwidth resources and adapt to the dynamic changes in mission payloads of each payload end system, thereby achieving an optimal balance between performance and efficiency at the system level. This invention achieves flight control bus scheduling and mission bus scheduling within the same open avionics system for a multi-purpose unmanned aerial vehicle (UAV), resolving the core contradiction of traditional solutions that cannot simultaneously accommodate both scheduling methods. This invention deeply decouples the flight control and mission systems in terms of logic and scheduling. Adding or replacing payload systems only requires adjustments to the mission scheduling strategy, without affecting the already certified flight control communication architecture. This reduces the complexity and risk of system upgrades, and improves overall reliability and maintainability.

[0035] In the description of this invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0036] Although the invention has been described herein in conjunction with various embodiments, those skilled in the art, by reviewing the specification and accompanying drawings, will understand and implement other variations of the disclosed embodiments in carrying out the claimed invention. In the specification, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude a plurality. While certain measures are described in different embodiments, this does not mean that these measures cannot be combined to produce good results.

[0037] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A heterogeneous bus scheduling method based on an open avionics system for multi-purpose unmanned aerial vehicles, characterized in that, Applied to the open avionics system of multi-purpose UAVs, the open avionics system of multi-purpose UAVs includes a flight control bus subsystem, an integrated processing module, a mission bus subsystem, multiple payload end systems on the mission bus, and a navigation end system, a servo end system, and a power supply and distribution end system on the flight control bus; Corresponding heterogeneous bus scheduling methods include: The flight control bus scheduling process includes: in the offline phase, each terminal system on the flight control bus generates its own local micro-hour slot scheduling table through the flight control bus subsystem; in the operational phase, each terminal system on the flight control bus communicates and schedules with the integrated processing module according to the corresponding local micro-hour slot scheduling table and local clock. The task bus scheduling process includes: for the task data streams requesting access from each payload system on the task bus, the task bus subsystem introduces a dynamic reconfiguration mechanism that combines critical awareness traffic shaping and event triggering to assess system load and task requirements in real time, proactively adjust resource allocation and communication scheduling strategies, and use a hybrid scheduling strategy of static and dynamic scheduling windows for online reconfiguration. Under the premise of not affecting the security of flight control bus communication, the remaining bandwidth resources of the system are adaptively optimized.

2. The heterogeneous bus scheduling method based on an open avionics system for multi-purpose unmanned aerial vehicles according to claim 1, characterized in that, During the offline phase of flight control bus scheduling, each system on the flight control bus generates its own local micro-hour slot scheduling table through the flight control bus subsystem, including: Obtain the key communication contract parameters of the virtual links corresponding to each system on the flight control bus; the key communication contract parameters of each virtual link include the bandwidth allocation interval, the maximum frame length, and the maximum transmit jitter allowed by the system. The minimum time slot length that satisfies the complete transmission of the maximum frame in the virtual link is calculated based on the maximum frame length. The least common multiple of the bandwidth allocation intervals of all virtual links is used as the total scheduling period. A micro time slot sequence is generated based on the total scheduling period and the minimum time slot length. Based on the bandwidth allocation interval and the micro-slot sequence, a transmission slot is allocated to each virtual link within the total scheduling period; A local micro-slot scheduling table is generated based on all transmission slot allocation results, and the local micro-slot scheduling table is permanently stored in the non-volatile memory corresponding to each terminal system on the flight control bus.

3. The heterogeneous bus scheduling method based on an open avionics system for multi-purpose unmanned aerial vehicles according to claim 2, characterized in that, During the offline phase of flight control bus scheduling, the constraints that must be followed when allocating transmission time slots for each virtual link within the total scheduling cycle include: periodic constraints, conflict-free constraints, and jitter constraints; among which, The periodic constraint is: the interval between adjacent transmission time slots allocated to each virtual link is equal to the corresponding bandwidth allocation interval or an integer fraction of the bandwidth allocation interval; The conflict-free constraint is that within any single micro-slot in a micro-slot sequence, at most one transmission slot can be allocated to a virtual link; The jitter constraint is: the deviation between the actual transmission interval and the bandwidth allocation interval of each virtual link is less than the maximum transmission jitter allowed by the system.

4. The heterogeneous bus scheduling method based on an open avionics system for multi-purpose unmanned aerial vehicles according to claim 1, characterized in that, During the flight control bus scheduling operation phase, each terminal system on the flight control bus communicates and schedules with the integrated processing module according to its corresponding local micro-hour slot scheduling table and local clock, including: Start the local loop counter; The flight controller scheduler adjusts the phase of the local cyclic counter based on the received absolute time alignment pulse, and tracks the current micro-hour slot of each virtual link based on the local clock and the adjusted local cyclic counter; the absolute time alignment pulse is periodically broadcast to the flight control bus by the integrated processing module. When entering the next micro-slot, the flight controller scheduler queries the local micro-slot scheduling table for the transmission slot allocation result corresponding to the current micro-slot, and stores the frame to be transmitted in the physical layer transmission queue according to the transmission slot allocation result, so as to realize the communication scheduling between the various end systems on the flight control bus and the integrated processing module through the physical layer transmission queue.

5. The heterogeneous bus scheduling method based on an open avionics system for multi-purpose unmanned aerial vehicles according to claim 1, characterized in that, In task bus scheduling, by introducing a critical awareness traffic shaping and event-triggered dynamic reconfiguration mechanism, system load and task requirements are assessed in real time, and resource allocation and communication scheduling strategies are proactively adjusted, including: Acquire the task data streams requesting admission from each payload system on the task bus; each task data stream has multi-dimensional attributes including data criticality level, data characteristic pattern, bandwidth requirement estimate, and transmission deadline constraint; The criticality-aware active flow shaper calculates the dynamic admission weights corresponding to the task data streams requesting admission from each payload system based on a preset dynamic weight function. The dynamic weight function takes the criticality level in the task data stream, the real-time load rate of the current task bus, the historical demand fulfillment rate of the task data stream, and the current flight phase status of the aircraft as input variables, and outputs the dynamic admission weights of the task data streams under the current system status as a comprehensive evaluation. Based on the dynamic admission weight, the task data stream requesting admission from each payload system is stored in the scheduling ready queue, so as to realize the communication scheduling between each payload system and the integrated processing module through the scheduling ready queue; When the critical perception active flow shaper determines that the system is in a resource shortage warning event, it stores the task data streams requesting access from each load end system in the scheduling ready queue according to the criticality level, bandwidth demand estimation, data characteristic pattern, and dynamic access weight, so as to realize the communication scheduling between each load end system and the integrated processing module through the scheduling ready queue.

6. The heterogeneous bus scheduling method based on an open avionics system for multi-purpose unmanned aerial vehicles according to claim 5, characterized in that, In task bus scheduling, the critical awareness-based proactive traffic shaper determines when the system is in a resource stress warning event, including: The criticality-aware proactive traffic shaper determines that system communication resources are about to become strained through load prediction models, or predicts through runtime monitoring mechanisms that high-criticality task data streams that have already been admitted may not be able to meet their transmission deadline constraints.

7. The heterogeneous bus scheduling method based on an open avionics system for multi-purpose unmanned aerial vehicles according to claim 1, characterized in that, In task bus scheduling, a hybrid scheduling strategy combining static and dynamic scheduling windows is used for online reconfiguration, including: Within each scheduling cycle of the task bus, a static scheduling window is defined, and the remaining bandwidth outside the static window within the scheduling cycle constitutes the dynamic scheduling window. The static scheduling window is used to transmit task data streams with strict periodicity and high deterministic requirements; the dynamic scheduling window is used to transmit task data streams that are non-periodic, bursty, or have dynamically changing bandwidth requirements. Within the static scheduling window, a static scheduling table is generated. Based on the static scheduling table, the data packet transmission sequence for the next scheduling period is calculated, and the data packet transmission sequence is transmitted through the task bus controller. The static scheduling table is generated offline based on the scheduling period of the task data stream. Within the dynamic scheduling window, the multi-objective optimization model based on the task bus dynamically calculates the data packet transmission sequence for the next scheduling cycle, and executes the transmission operation of the data packet transmission sequence through the task bus controller; the multi-objective optimization model of the task bus takes ensuring system security and transmission efficiency as its optimization objectives.

8. The heterogeneous bus scheduling method based on an open avionics system for multi-purpose unmanned aerial vehicles according to claim 7, characterized in that, The multi-objective optimization model of the task bus includes primary objective, core performance objective, service quality objective, and auxiliary fairness objective; among them, The primary objective is to maximize the on-time delivery rate of all high-criticality task data streams in the system. The core performance objective is to maximize the total bandwidth utilization of the task bus within the current scheduling cycle. The quality of service objective is to minimize the overall average transmission latency and latency jitter of all task data streams; The auxiliary fairness objective is to ensure fairness in service access among data streams of lower criticality levels, while meeting the primary objective, core performance objective, and quality of service objective.

9. The heterogeneous bus scheduling method based on an open avionics system for multi-purpose unmanned aerial vehicles according to claim 1, characterized in that, The task bus scheduling process also includes: When the system detects that the criticality level of a newly requested task data stream is higher than the current dynamic criticality benchmark, or higher than the criticality level of a task data stream currently being scheduled in the system, it performs online reconfiguration on all current task data streams based on the data characteristic pattern, bandwidth requirement estimate, transmission deadline constraint, and remaining bandwidth status of the current dynamic scheduling window of the newly requested task data stream.

10. The heterogeneous bus scheduling method based on an open avionics system for multi-purpose unmanned aerial vehicles according to claim 1, characterized in that, The task bus scheduling process also includes: When the system detects that a high-criticality task data stream that has requested admission may miss its transmission deadline constraint, it performs online reconfiguration of the task data stream based on its criticality level, data characteristic pattern, bandwidth requirement estimate, and the urgency of the remaining deadline.

Images