Flight control and electronic speed controller integrated intelligent control system

Abstract

The invention relates to the technical field of unmanned aerial vehicle control, in particular to a flight control and electronic speed controller integrated intelligent control system which comprises a local data processing module, a data uploading module, a cloud computing and scheduling module, a local execution module and a state feedback and iteration module. According to the system, high-real-time core data are locally processed preferentially, low-real-time and interactive data are uploaded in a classified mode, the cloud end optimizes parameters by means of sufficient computing power and dynamically allocates tasks on the basis of local computing power loads, and according to a closed-loop feedback iteration scheduling strategy, the problem of processing delay caused by insufficient computing power and unreasonable scheduling of a traditional system is solved. Motor overcurrent protection false triggering is avoided, flight attitude stability and motor safety are guaranteed, high-efficiency utilization of computing power is realized, system control precision and adaptability are improved, and high efficiency, stability and reliability of integrated control are achieved.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) control technology, specifically to an integrated intelligent control system for flight control and electronic speed regulation. Background Technology

[0002] The flight control and electronic speed controller (ESC) integrated system needs to process three types of data simultaneously: ① The flight control side needs to perform attitude calculation, path planning, and control law calculation; ② The ESC side needs to perform closed-loop control of motor speed, overcurrent / overheat protection, and phase current sampling; ③ Interactive data, such as command encoding, status decoding, and data buffering.

[0003] Small drone control systems almost 100% use local control chips as the core processing unit. For example, most open-source and commercial drone flight controllers (such as ArduPilot and PX4) are built on MCUs.

[0004] If the CPU computing power is insufficient, or the task scheduling priority is unreasonable (such as flight control path planning occupying too much computing power), it will cause delays in ESC data processing. For example, if the phase current sampling data is not parsed in time, it will cause the motor overcurrent protection to be falsely triggered. Summary of the Invention

[0005] The purpose of this invention is to provide an integrated intelligent control system for flight control and electronic speed regulation to solve the problems mentioned in the background art.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] The integrated intelligent control system for flight control and electronic speed regulation includes:

[0008] The local data processing module is used to start and run continuously after the system is powered on and initialized. It classifies three types of data, prioritizes the processing of flight control attitude calculation, ESC motor speed closed-loop control, overcurrent and overheat protection and phase current sampling, and temporarily stores the interactive data after verification.

[0009] The data upload module is used to start uploading after low real-time data is generated, and to trigger uploading when interactive data reaches the cache threshold or is timed. It transmits flight control path planning, control law calculation related data and temporarily stored interactive data to the cloud via wireless communication.

[0010] The cloud computing and scheduling module is used to receive data, start processing and continuously iterate, complete flight control path planning, control law parameter optimization, and generate local and cloud task allocation and priority strategies based on local computing power load.

[0011] The local execution module is used to receive path results, control parameters and scheduling strategies from the cloud, adjust local task priorities, execute flight control and ESC control, and collect system status data.

[0012] The status feedback and iteration module is used to periodically feed status data from the local machine to the cloud, and the cloud periodically iterates and schedules the strategy, evaluates the rationality of the strategy based on the feedback, and adjusts and distributes it.

[0013] Furthermore, after the system is powered on and initialized, it starts and runs continuously. Specifically, after the system is powered on, it completes the initialization process of self-testing of the local MCU and wireless communication module. Then, the local data processing module starts and runs continuously. This module classifies three types of data in real time: flight control side, ESC side, and interactive data. It prioritizes processing high real-time data such as flight control attitude calculation, ESC side motor speed closed-loop control, overcurrent and overheat protection, and phase current sampling. Interactive data is temporarily stored after format verification, and failed data is directly discarded and marked as invalid data.

[0014] Furthermore, the classification of the three types of data prioritizes the processing of flight control attitude calculation, ESC motor speed closed-loop control, overcurrent and overheat protection, and phase current sampling. Specifically, the system classifies data according to real-time requirements: flight control attitude calculation, ESC phase current sampling, motor speed closed-loop control, and overcurrent and overheat protection are classified as high real-time data; flight control path planning and control law calculation are classified as low real-time data; and interactive data is classified separately. After the local MCU starts, it prioritizes the processing of high real-time data, ensuring timely processing through the highest priority interrupt to guarantee flight stability and motor safety, and avoid false triggering of protection. Low real-time data and interactive data do not occupy core computing power at this time and will be processed later with allocated resources.

[0015] Furthermore, the temporary storage of the interactive data after verification specifically involves the following steps: the interactive data includes instruction encoding, status decoding, and data caching. The local data processing module performs format verification on the data, which includes verification bit matching and data length matching. Data that passes the verification is temporarily stored in a local circular buffer using a first-in-first-out mechanism to avoid overflow. Data that fails the verification is directly discarded and marked as invalid.

[0016] Furthermore, the low real-time data is uploaded after it is generated, and the interactive data is uploaded when it reaches the cache threshold or is triggered at a time. Specifically, the low real-time data is the original data of flight control path planning and the control law calculation parameters generated before uploading; the interactive data is temporarily stored locally and uploaded according to the cache threshold or triggered at a time; before uploading, the data is compressed and encrypted and transmitted through the wireless communication module, and the service quality mechanism is enabled to mark low latency priority data frames and carry timestamps to facilitate cloud-based tracking of data timing.

[0017] Furthermore, the process of transmitting flight control path planning, control law calculation-related data, and temporarily stored interactive data to the cloud via wireless communication specifically involves the following: the data to be uploaded includes environmental data related to flight control path planning, original parameters related to control law calculation, and temporarily stored interactive data that has passed verification; the data is compressed and encrypted before being uploaded; it is transmitted through a designated wireless communication module; the communication link uses a quality of service mechanism to prioritize low latency; and the transmitted data frames carry timestamps so that the cloud can trace the data sequence.

[0018] Furthermore, after receiving the data, the process is initiated and continuously iterated to complete the flight control path planning and control law parameter optimization. Specifically, the cloud receives the flight control path planning environment data, the original parameters for control law calculation, and the verified interactive data uploaded locally. Relying on sufficient computing power and algorithm model library, the path and control law parameters are optimized using corresponding algorithms. Based on the system status data fed back locally, the optimization is continuously iterated to ensure that the results are adapted to the UAV's operating status.

[0019] Furthermore, the step of generating local and cloud task allocation and priority strategies based on local computing power load specifically involves the cloud receiving local real-time computing power load related data, formulating task allocation rules based on the real-time requirements of three types of data, prioritizing the processing of high real-time data locally, flexibly allocating low real-time data according to local computing power, generating a task priority strategy with the highest interruption priority for local high real-time tasks, and a local-cloud data interaction cycle. The strategy is updated as the local computing power load changes.

[0020] Furthermore, the process of receiving path results, control parameters, and scheduling strategies from the cloud and adjusting local task priorities involves the local execution module receiving path planning results, control law parameters, and global scheduling strategies from the cloud via the local MCU. Upon receipt, a second-highest priority parsing interrupt is triggered, prioritizing the parsing of control law parameters and scheduling strategies. Data integrity is verified using CRC, and if this fails, a retransmission is requested. After successful parsing, task priorities are adjusted based on the real-time operating system, with overcurrent and overheat protection on the ESC side having the highest priority, followed by attitude calculation on the flight control side. The priority of low-real-time tasks is lowered to ensure that high-real-time tasks preempt local computing power.

[0021] Furthermore, the execution of flight control and electronic speed controller (ESC) control, while simultaneously collecting system status data, specifically involves the local execution module executing control tasks based on the path planning results, optimized control law parameters, and global scheduling strategies sent from the cloud. The flight control module ensures flight stability based on local attitude data, while the ESC adjusts the motors and monitors protection based on phase current data. The real-time operating system maintains task priority and continuously collects status data related to flight control, ESC, and interaction. The collected data is temporarily stored in a local buffer with a dual-buffer mechanism to avoid conflicts.

[0022] Compared with existing technologies, this invention has the following advantages: This application effectively solves the processing delay problems caused by insufficient computing power and unreasonable task scheduling in traditional UAV flight control ESC systems through a task scheduling and data processing mechanism that combines local and cloud collaboration. The system prioritizes processing core high real-time data such as flight attitude calculation and ESC motor speed closed-loop control, ensuring timely response with the highest priority interrupt, avoiding false triggering of motor overcurrent protection caused by phase current sampling delay, and ensuring stable flight attitude and safe motor operation.

[0023] Through data classification and layered upload strategies, low real-time data and interactive data are temporarily stored after verification, then compressed and encrypted before being uploaded according to threshold or timed rules. QoS mechanisms ensure low latency and security during transmission. The cloud leverages ample computing power to optimize path planning and control law parameters, uses LSTM to predict local computing load, dynamically allocates tasks, and updates priority strategies, achieving efficient utilization of computing resources.

[0024] The closed-loop feedback mechanism between local and cloud systems enables the cloud to iterate scheduling strategies based on real-time status data, adjust task allocation and computing power ratio, and ensure that system parameters are always adapted to the UAV's operating status, thereby further improving control accuracy and system adaptability, and achieving efficient, stable and reliable integrated flight control and ESC control. Attached Figure Description

[0025] Figure 1 This is a block diagram of the integrated intelligent control system for flight control and electronic speed regulation of the present invention. Detailed Implementation

[0026] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments and accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] This application discloses an integrated intelligent control system for flight control and electronic speed regulation, such as... Figure 1 As shown, it includes:

[0028] The local data processing module is used to start and run continuously after the system is powered on and initialized. It classifies three types of data, prioritizes the processing of flight control attitude calculation, ESC motor speed closed-loop control, overcurrent and overheat protection and phase current sampling, and temporarily stores the interactive data after verification.

[0029] The data upload module is used to start uploading after low real-time data is generated, and to trigger uploading when interactive data reaches the cache threshold or is timed. It transmits flight control path planning, control law calculation related data and temporarily stored interactive data to the cloud via wireless communication.

[0030] The cloud computing and scheduling module is used to receive data, start processing and continuously iterate, complete flight control path planning, control law parameter optimization, and generate local and cloud task allocation and priority strategies based on local computing power load.

[0031] The local execution module is used to receive path results, control parameters and scheduling strategies from the cloud, adjust local task priorities, execute flight control and ESC control, and collect system status data.

[0032] The status feedback and iteration module is used to periodically feed status data from the local machine to the cloud, and the cloud periodically iterates and schedules the strategy, evaluates the rationality of the strategy based on the feedback, and adjusts and distributes it.

[0033] The system starts and runs continuously after power-on initialization. Specifically, upon power-on, the system first completes the initialization process, which includes local MCU self-test and wireless communication module self-test. Within 0-100ms after initialization, the local data processing module starts and runs continuously until the system is powered off. This module performs real-time classification of three types of data: flight control data, ESC data, and interactive data. After classification, high real-time data is processed first, specifically including flight control attitude calculation, and ESC-side motor speed closed-loop control, overcurrent and overheat protection, and phase current sampling. For interactive data, the module first performs format verification, including checksum verification and data length matching. Verified interactive data is temporarily stored in a local 1KB circular buffer, while failed interactive data is discarded and marked as invalid.

[0034] The three data categories prioritize flight control attitude calculation, ESC motor speed closed-loop control, overcurrent and overheat protection, and phase current sampling. Specifically, the system formulates classification rules based on real-time data requirements. Flight control attitude calculation and ESC phase current sampling, requiring a response time of ≤10ms, are classified as high real-time data. ESC motor speed closed-loop control and overcurrent and overheat protection, directly related to motor operation safety, are also included in the high real-time category. Flight control path planning and control law calculation, with a response time requirement of ≤500ms, are classified as low real-time data. Interactive data, including command encoding, status decoding, and data caching, are classified separately. After the local MCU starts, high real-time data is processed first. A high real-time data processing interrupt is triggered using the MCU's built-in timer, with the highest priority, allowing it to preempt other non-real-time tasks. This ensures timely completion of flight control attitude calculation, providing a foundation for UAV attitude stability. Simultaneously, it ensures accurate execution of ESC motor speed closed-loop control, real-time phase current sampling, and overcurrent and overheat protection checks, preventing false triggering of motor overcurrent protection due to data processing delays. Low real-time data and interactive data will not occupy core computing power for the time being, and will be processed during the processing gaps of high real-time data or according to subsequent strategies.

[0035] The temporary storage of verified interactive data, in practice, involves the following: the interactive data includes instruction encoding, status decoding, and data buffering. The local data processing module performs format verification on the interactive data. Verification includes checking the data length against checksums. Verified interactive data is temporarily stored in a local circular buffer with a capacity of 1KB. The circular buffer employs a first-in, first-out (FIFO) mechanism to prevent data overflow. Interactive data that fails verification is discarded and marked as invalid.

[0036] The low real-time data is uploaded after generation, triggered when interactive data reaches a cache threshold or is periodically uploaded. Specifically, low real-time data refers to the original path planning data and control law calculation parameters on the flight controller side. After this type of data is generated, the upload process will begin within 50 milliseconds. Interactive data is temporarily stored in a local 1KB circular buffer. When the amount of interactive data stored in the circular buffer reaches 800B, an upload will be triggered. Even if the cache threshold is not reached, the system will still trigger an interactive data upload every 30 milliseconds. Before uploading, the data is processed using the LZ4 data compression algorithm, which has a compression rate of no less than 60%. The compressed data is then encrypted using the AES-128 algorithm. The encrypted data is transmitted via an industrial-grade 5G or WiFi 6 wireless communication module. A Quality of Service (QoS) mechanism is enabled on the communication link, and uploaded data is marked as low-latency priority. This ensures that the data transmission latency does not exceed 80 milliseconds, and the data frames carry timestamps for easy cloud-based data timing tracking.

[0037] The process of transmitting flight control path planning, control law calculation, and temporarily stored interactive data to the cloud via wireless communication is implemented as follows: The data to be uploaded includes flight control path planning data, control law calculation data, and temporarily stored interactive data. Flight control path planning data primarily consists of environmental data, including GPS positioning information and obstacle distance data. Control law calculation data mainly consists of raw parameters, including UAV mass and inertia matrix. Interactive data is valid data temporarily stored after passing format verification. Before uploading, the data must be compressed using the LZ4 data compression algorithm, which has a compression rate of no less than 60%. The compressed data must be encrypted using the AES-128 algorithm. The encrypted data is transmitted through a designated wireless communication module, such as an industrial-grade 5G module or a WiFi 6 module. The communication link must enable a Quality of Service (QoS) mechanism, marking uploaded data as low-latency priority. This ensures that data transmission latency does not exceed 80 milliseconds. Transmitted data frames must carry timestamps to facilitate cloud-based data timing tracking.

[0038] After receiving the data, processing is initiated and continuously iterated to complete flight control path planning and control law parameter optimization. Specifically, in implementation, the cloud receives locally uploaded data via a wireless communication link. This data includes environmental data related to flight control path planning, raw parameters related to control law calculation, and verified, temporarily stored interactive data. The cloud initiates the processing flow within 100ms of receiving the data. The cloud relies on the sufficient computing power provided by the GPU / CPU cluster. The cloud internally deploys a task processing algorithm model library including RRT* and MPC algorithms, supporting rapid invocation of relevant algorithms. During processing, the RRT* algorithm is first used to dynamically optimize the flight control path, ensuring the path effectively avoids real-time obstacles. Simultaneously, the flight control law parameters are optimized based on the model predictive control algorithm, focusing on adjusting key parameters such as PID gain. The single processing cycle of the path planning task does not exceed 300ms. The cloud continuously receives locally fed-back system status data and performs iterative optimization of the control law parameters every 200ms based on this data. After each optimization, the parameters are compared with the preset control target. The preset control objectives include a motor speed error of no more than ±5 rpm and a phase current sampling delay of no more than 10 ms. Through continuous iteration, the flight control path planning results and control law parameters are ensured to always adapt to the UAV's operating state.

[0039] In another preferred embodiment, this application proposes a hybrid algorithm combining dynamic potential energy field and adaptive fuzzy PID, which integrates a potential energy field model predicted by the dynamic environment with fuzzy logic-driven PID parameter self-tuning, focusing on UAV flight control and ESC integrated scenarios. The specific process is as follows:

[0040] Prepare the input data, including environmental data, status data, and control parameters;

[0041] Environmental data includes the planar coordinates of the obstacle at time k. The velocity of the obstacle at time k GPS planar coordinates of the UAV at time k The flight altitude of the drone at time k ;

[0042] State data includes the drone's flight speed at time k. The roll angle of the drone at time k Pitch angle The actual speed of the motor at time k ;

[0043] Control parameters include the initial proportional gain of the PID controller. ,integral Differential gain System control cycle The reference speed n of the motor ref (Control target value);

[0044] Then, path planning is performed using a dynamic potential field algorithm, including:

[0045] Constructing a dynamic gravitational potential energy field for UAV target points The formula for calculating the gravitational potential energy generated by a drone is:

[0046] ;

[0047] In the formula, Let be the gravitational potential energy experienced by the drone at time k; The gravitational coefficient; The target point's planar coordinates for the UAV; Let be the planar coordinates of the UAV at time k.

[0048] The gravitational force is along the negative direction of the potential energy gradient, and the corresponding formula for calculating the gravitational vector is:

[0049] ;

[0050] In the formula, Let $k$ be the gravitational vector acting on the UAV at time $k$ (driving the UAV toward the target point). This is the gradient operator, used to solve for the rate of change of the potential energy field.

[0051] A dynamic repulsive potential energy field is constructed, considering the motion characteristics of obstacles, and the repulsive potential energy of the obstacle velocity prediction term is introduced. The calculation formula is as follows:

[0052] ;

[0053] The corresponding formula for calculating the repulsive force vector is:

[0054] ;

[0055] The total potential energy field and the path point are combined, and the total force is the vector sum of the attractive and repulsive force vectors, that is:

[0056] ;

[0057] In the formula, Let K be the repulsive force vector experienced by the UAV at time k. Let k be the repulsive potential energy experienced by the drone at time k; The repulsion coefficient is... The radius of the repulsive force. Let k be the velocity of the obstacle at time k; Time for predicting obstacle location Let k be the real-time straight-line distance between the drone and the obstacle. The formula for calculation is: ; Let K be the total virtual force acting on the drone at time k.

[0058] After normalizing the total force, the target position of the UAV at time k+1 is obtained as follows:

[0059] ;

[0060] In the formula, Let K be the target plane coordinates of the UAV at time k+1; Let K be the target plane coordinates of the UAV at time k+1; Maximum speed for safe flight of drones; The magnitude of the total force vector; It is the unit vector of the total force.

[0061] Parameter optimization is achieved through an adaptive fuzzy PID algorithm, including:

[0062] Input variables include motor speed error This is the difference between the actual motor speed and the reference speed at time k; the rate of change of speed error. That is, the rate of change of the rotational speed error at time k relative to the previous time (k1);

[0063] Output variables include , , , where is the correction value used at time k to adjust the proportional, integral, and derivative gains of the PID controller.

[0064] Input variables , and output variables , , Each set is divided into 5 fuzzy subsets: {Negative Large (NB), Negative Small (NS), Zero (Z), Positive Small (PS), Positive Large (PB)}, and the membership function adopts a triangular distribution.

[0065] With speed error For example, the formula for calculating its membership function is:

[0066] ;

[0067] In the formula, Speed ​​error The membership degree of the “Negative Large (NB)” fuzzy subset (value range [0,1]); Speed ​​error The membership degree of the “zero (Z)” fuzzy subset (value range [0,1]).

[0068] Fuzzy rules cover all input combinations in the scenario, as shown in the following example: If For NB, If it is NB, then For PB, For NB, For PS; if For PS, If it is NS, then For PS, For Z, For NS.

[0069] The centroid method is used to clarify the fuzzy output, resulting in... , , The exact value of the PID gain is obtained, and the updated PID gain calculation formula is as follows:

[0070] ;

[0071] In the formula, , , The PID controller's proportional, integral, and derivative gains (dimensionless) are updated at time k. , , This is the initial gain of the PID controller.

[0072] Output optimized real-time PID gain ( , , ), used for closed-loop control of motor speed.

[0073] The cloud receives system status data (motor speed error |e(k)|, phase current sampling delay t_delay) from the local system every 200ms. When either the motor speed error |e(k)|>5rpm or the phase current sampling delay t_delay>10ms is met, the algorithm parameters are automatically adjusted.

[0074] For example, the configuration parameter adjustment rule is: gravity coefficient The adjustment range is 4~6 (the farther away from the target, the larger the value); repulsion coefficient The adjustment range is 80~120 (the higher the obstacle density, the larger the value); the threshold for the fuzzy rule error interval will be... The NB interval threshold has been adjusted from 50rpm to 40rpm to improve error response sensitivity.

[0075] The processing cycle for a single path planning operation does not exceed 250ms, ensuring that the system's real-time requirements are met.

[0076] The final output includes path planning results and control law parameters.

[0077] For example, the path planning result is a sequence of continuous waypoint coordinates for the UAV. and corresponding speed commands;

[0078] The control law parameter is the real-time optimized PID gain. .

[0079] The process of generating local and cloud task allocation and priority strategies based on local computing load is implemented as follows: The cloud receives real-time computing load data from the local system. This data includes the CPU utilization of the local MCU and the execution time of each task. The cloud uses an LSTM neural network to predict the computing load of the local MCU for the next 100ms, with a prediction accuracy of at least 90%. The cloud formulates task allocation rules based on the real-time requirements of three types of data. High real-time data, requiring a response time of no more than 10ms, is prioritized for processing by the local MCU. Low real-time data, requiring a response time of no more than 500ms, is flexibly allocated based on the local computing load. When the CPU utilization of the local MCU reaches or exceeds 70%, the cloud automatically transfers some control law calculation subtasks to the cloud for processing. Simultaneously, the cloud generates a local task priority strategy, ensuring that high real-time tasks maintain the highest interrupt priority. The strategy includes the data interaction cycle between the local and cloud systems. This task allocation and priority strategy is updated every 200ms to ensure real-time matching between the strategy and changes in local computing load.

[0080] The process of receiving path results, control parameters, and scheduling strategies from the cloud and adjusting local task priorities involves the local execution module receiving data from the cloud via its local MCU. The data includes packaged path planning results, control law parameters, and a global scheduling strategy. The path planning results include waypoint coordinate sequences and velocity curves. Control law parameters include PID gain values ​​and control cycles. The global scheduling strategy encompasses local task priority ranking and data interaction cycle optimization instructions. Upon receiving data, the local MCU triggers a result parsing interrupt. This interrupt has the second-highest priority, only lower than the high real-time data processing interrupt. The MCU prioritizes parsing the control law parameters and scheduling strategy before parsing the path planning results. The data carries a CRC checksum, which the MCU uses to verify data integrity. If parsing fails, the local MCU immediately requests a retransmission from the cloud. Retransmissions are limited to a maximum of three times, with a 10ms interval between each retransmission. Upon successful parsing, the local MCU dynamically adjusts task priorities based on the FreeRTOS real-time operating system. The adjustment follows fixed rules, with the ESC-side overcurrent and overheat protection task maintaining the highest priority. The flight control-side attitude calculation task has the next highest priority. The priority of low real-time tasks such as flight control path planning and control law calculation has been reduced. By adjusting the priority, it is ensured that high real-time tasks can preempt local computing resources, and low real-time tasks are prevented from interfering with ESC data processing and flight control attitude calculation.

[0081] The implementation involves executing flight control and ESC control while simultaneously collecting system status data. Specifically, the local execution module executes flight control and ESC control tasks based on path planning results, optimized control law parameters, and global scheduling strategies sent from the cloud. For flight control, based on locally calculated attitude data and optimized PID gain values ​​and control cycles, the module drives the UAV's flight control actuators to ensure stable flight attitude. For ESC control, relying on real-time collected phase current data, the module adjusts the motor's operating state according to the speed closed-loop control logic, while continuously monitoring the motor phase current values ​​and MOSFET temperatures to trigger overcurrent and overheat protection mechanisms. The local FreeRTOS real-time operating system ensures task execution, maintaining the ESC-side overcurrent and overheat protection task as the highest priority, followed by the flight control-side attitude calculation task. System status data acquisition continues continuously at a sampling frequency of no less than 1kHz. The acquired data includes flight control-side gyroscope and accelerometer counts, ESC-side phase current values ​​and MOSFET temperatures, and the command completion rate of the interactive data. The collected status data is temporarily stored in a local status cache. This cache uses a dual-buffer mechanism: one buffer is used for data collection, and the other is used for data feedback, to avoid conflicts between the collection and feedback processes.

[0082] The system periodically feeds back status data to the cloud, and the cloud iterates the scheduling strategy periodically, evaluating its rationality based on the feedback and adjusting its deployment accordingly. Specifically, the local status buffer stores the collected system status data. This data includes gyroscope values ​​from the flight controller, accelerometer counts from the flight controller, phase current values ​​from the electronic speed controller (ESC), MOSFET temperatures from the ESC, completion rates of interactive data commands, local task execution time, and communication link latency. Every 50ms, the local MCU packages this data from the status buffer according to a preset format. The packaged data is encrypted using the AES-128 algorithm. After encryption, it is fed back to the cloud via an industrial-grade 5G or WiFi 6 wireless communication module. The cloud receives the local feedback data every 200ms and initiates scheduling strategy iteration. An internal system operation status evaluation model is established in the cloud. During evaluation, the received status data is compared with preset control targets. These preset control targets include a motor speed error of no more than ±5 rpm and a phase current sampling delay of no more than 10ms. If there is a deviation between the status data and the preset control targets, such as a phase current sampling delay greater than 10ms, the cloud dynamically adjusts the task allocation ratio between the local and cloud systems. The adjustments include increasing the proportion of computing power allocated to local high real-time tasks. After the adjustments are complete, a new global scheduling policy is generated. This new policy is sent to the local MCU via the original communication link. The sent data carries a CRC checksum to ensure integrity. The local MCU receives the data, parses and executes it according to the established process, forming a closed-loop optimization. This process continues until the system is powered off.

[0083] This application also protects an electronic device deployed via the aforementioned integrated flight control and electronic speed control system.

[0084] This application also protects a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the functions of the flight control and electronic speed control integrated intelligent control system provided in the embodiments of the present invention.

[0085] The computer storage medium of this invention can be any combination of one or more computer-readable media. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, system, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, system, or device.

[0086] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit programs for use by or in conjunction with an instruction execution system, system, or device.

[0087] Program code contained on a computer-readable medium may be transmitted using any suitable medium, including—but not limited to—wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0088] The above description is merely one specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. An integrated intelligent control system for flight control and electronic speed regulation, characterized in that, include: The local data processing module is used to start and run continuously after the system is powered on and initialized. It classifies three types of data, prioritizes the processing of flight control attitude calculation, ESC motor speed closed-loop control, overcurrent and overheat protection and phase current sampling, and temporarily stores the interactive data after verification. The data upload module is used to start uploading after low real-time data is generated, and to trigger uploading when interactive data reaches the cache threshold or is timed. It transmits flight control path planning, control law calculation related data and temporarily stored interactive data to the cloud via wireless communication. The cloud computing and scheduling module is used to receive data, start processing and continuously iterate, complete flight control path planning, control law parameter optimization, and generate local and cloud task allocation and priority strategies based on local computing power load. The local execution module is used to receive path results, control parameters and scheduling strategies from the cloud, adjust local task priorities, execute flight control and ESC control, and collect system status data. The status feedback and iteration module is used to periodically feed status data from the local machine to the cloud, and the cloud periodically iterates and schedules the strategy, evaluates the rationality of the strategy based on the feedback, and adjusts and distributes it.

2. The integrated intelligent control system for flight control and electronic speed regulation according to claim 1, characterized in that, After power-on initialization, the system starts and runs continuously. Specifically, after power-on, the system completes the initialization process of self-testing of the local MCU and wireless communication module. Subsequently, the local data processing module starts and runs continuously. This module classifies three types of data in real time: flight control side, ESC side, and interactive data. It processes high real-time data such as flight control attitude calculation, ESC side motor speed closed-loop control, overcurrent and overheat protection, and phase current sampling. After the interactive data is format-verified, the successful data is temporarily stored, and the failed data is directly discarded and marked as invalid data.

3. The integrated intelligent control system for flight control and electronic speed regulation according to claim 2, characterized in that, The system categorizes data into three types for processing flight control attitude calculation, ESC motor speed closed-loop control, overcurrent and overheat protection, and phase current sampling. Specifically, the system categorizes data based on real-time requirements: flight control attitude calculation, ESC phase current sampling, motor speed closed-loop control, and overcurrent and overheat protection are classified as high real-time data; flight control path planning and control law calculation are classified as low real-time data; and interactive data is categorized separately. The local MCU processes high real-time data upon startup, ensuring timely processing through the highest priority interrupt to guarantee flight stability and motor safety, and to prevent false protection triggers. Low real-time data and interactive data do not currently occupy core computing power and will be processed later with allocated resources.

4. The integrated intelligent control system for flight control and electronic speed regulation according to claim 1, characterized in that, The temporary storage of the verified interactive data is specifically as follows: the interactive data includes instruction encoding, status decoding, and data caching. The local data processing module performs format verification on it, which includes verification bit matching and data length matching. Data that passes the verification is temporarily stored locally. A circular buffer with a first-in-first-out mechanism is used to avoid overflow. Data that fails the verification is directly discarded and marked as invalid data.

5. The integrated intelligent control system for flight control and electronic speed regulation according to claim 1, characterized in that, The low real-time data is uploaded after it is generated, and uploaded when the interactive data reaches the cache threshold or is triggered at a time. Specifically, the low real-time data is the original data of flight control path planning and the control law calculation parameters, which are uploaded after they are generated; the interactive data is temporarily stored locally and uploaded according to the cache threshold or triggered at a time; the data is compressed and encrypted before being uploaded and transmitted through the wireless communication module.

6. The integrated intelligent control system for flight control and electronic speed regulation according to claim 5, characterized in that, The process of transmitting flight control path planning, control law calculation related data, and temporarily stored interactive data to the cloud via wireless communication specifically includes flight control path planning related environmental data, control law calculation related raw parameters, and verified temporarily stored interactive data; the data is compressed and encrypted before being uploaded. Transmission is carried out through a designated wireless communication module; the communication link enables a quality of service mechanism to prioritize low latency; and the transmitted data frames carry timestamps so that the cloud can trace the data sequence.

7. The integrated intelligent control system for flight control and electronic speed regulation according to claim 1, characterized in that, After receiving the data, the process is initiated and continuously iterated to complete the flight control path planning and control law parameter optimization. Specifically, the cloud receives the flight control path planning environment data, the original parameters for control law calculation, and the verified interactive data uploaded locally. Relying on sufficient computing power and algorithm model library, the path and control law parameters are optimized using the corresponding algorithms, and continuous iterative optimization is performed based on the system status data fed back locally.

8. The integrated intelligent control system for flight control and electronic speed regulation according to claim 1, characterized in that, The process of generating local and cloud task allocation and priority strategies based on local computing power load involves the cloud receiving local real-time computing power load data and formulating task allocation rules based on the real-time requirements of three types of data. High real-time data is processed locally first, while low real-time data is flexibly allocated according to local computing power. A task priority strategy with the highest interruption priority for local high real-time tasks and a local-cloud data interaction cycle are generated, and the strategy is updated as the local computing power load changes.

9. The integrated intelligent control system for flight control and electronic speed regulation according to claim 1, characterized in that, The process of receiving path results, control parameters, and scheduling strategies from the cloud and adjusting local task priorities involves the local execution module receiving path planning results, control law parameters, and global scheduling strategies from the cloud via the local MCU. Upon receipt, a second-highest priority parsing interrupt is triggered, prioritizing the parsing of control law parameters and scheduling strategies. Data integrity is verified using CRC, and if this fails, a retransmission is requested. After successful parsing, task priorities are adjusted based on the real-time operating system, with overcurrent and overheat protection on the ESC side having the highest priority, followed by attitude calculation on the flight control side. The priority of low-real-time tasks is reduced to ensure that high-real-time tasks preempt local computing power.

10. The intelligent control system integrating flight control and electronic speed regulation according to claim 1, characterized in that, The process involves executing flight control and electronic speed controller (ESC) control while simultaneously collecting system status data. Specifically, the local execution module executes control tasks based on the path planning results, optimized control law parameters, and global scheduling strategies sent from the cloud. The flight control system ensures flight stability based on local attitude data, and the ESC adjusts the motors and monitors protection based on phase current data. The real-time operating system maintains task priority and continuously collects status data related to flight control, ESC, and interaction. The collected data is temporarily stored in a local buffer with a dual-buffer mechanism to avoid conflicts.

Images