A method and system for monitoring and managing power supply of a drone

Abstract

The present application relates to the field of unmanned aerial vehicle monitoring, and more particularly to a method and system for monitoring and managing power supply of unmanned aerial vehicles, which obtains a flight control instruction timing queue and power supply parameters, performs power supply excitation analysis according to power supply super parameters, obtains an excitation instruction set, determines corresponding abnormal viewing angles and power supply parameter samples, simultaneously obtains a predetermined viewing area image mapped in a time domain with the flight control instruction timing queue, determines a scene type based on the predetermined viewing area image, performs instruction rule analysis, determines a post-potential instruction pool, and subsequently matches the post-potential instruction pool according to the determined scene type during unmanned aerial vehicle flight, verifies power supply excitation tendency, performs power supply parameter intervention, and further adjusts a setting strategy using the excitation instruction corresponding to the adaptability of the abnormal viewing angle and verifies the power supply intervention effect. The present application uses the correlation between the flight control instruction combination and the power supply parameter change, combines the flight scene pre-judgment, intervenes in the power supply parameter in advance, and improves the safety and reliability of power supply management.

Description

Technical Field

[0001] This relates to the field of drone monitoring, and in particular to a method and system for monitoring and managing drone power supply. Background Technology

[0002] When a drone is performing a flight mission, its power supply system typically collects and monitors parameters such as battery voltage, current, temperature, and remaining charge in real time through a battery management system or power management unit. When the monitored parameters exceed preset safety thresholds, the system triggers an alarm or executes protective actions to prevent safety accidents caused by power supply anomalies. With the diversification of drone mission scenarios and the increasing complexity of payloads, the sequence of maneuvers during flight and the coordinated operation of multiple payloads can lead to varying degrees of transient changes in power supply parameters. Power supply monitoring has become one of the key technical aspects of ensuring the reliable operation of drones.

[0003] For example, Chinese Patent Publication No. CN111181209A discloses a multi-channel power management system for unmanned aerial vehicles (UAVs), including a power supply, a power management module, an MCU control module, and a wireless module. The power management module's input is connected to the power supply, and its output has several power supply channels. The power management module internally houses several power management units, each with one end connected to the power management module's input and the other end connected to one of the power supply channels. The MCU control module is connected to all power management units on one end and to the wireless module on the other. The power management module converts a single voltage value into multiple voltage values, allowing the UAV to meet different power supply voltage requirements with only one power module, thus increasing its payload capacity. The wireless module enables information interaction with the control terminal, achieving intelligent power management and remote power control.

[0004] However, the following problems still exist in the existing technology. In existing technologies, UAV power supply monitoring methods mainly rely on passive threshold comparisons of parameters such as voltage and current. They do not consider the correlation between transient changes in power supply parameters and flight control commands, nor do they combine flight scenario information for advance prediction. They also lack analysis of power supply parameter changes triggered by specific command combinations. This makes it impossible to identify power supply risks before dangerous commands are executed, and it is difficult to implement adaptive intervention for power supply anomalies caused by different command combinations. As a result, power supply management is lagging and inefficient. Summary of the Invention

[0005] To address this, the present invention provides an intelligent synchronization system based on multi-terminal interaction, which overcomes the problem in the prior art that lacks analysis of changes in power supply parameters stimulated by specific command combinations, resulting in the inability to identify power supply risks before the execution of dangerous commands, difficulty in implementing adaptive intervention for power supply variations caused by different command combinations, and lagging power supply management with poor results.

[0006] To achieve the above objectives, in one aspect, the present invention provides a method for monitoring and managing the power supply of unmanned aerial vehicles (UAVs), comprising: Read the UAV flight control commands, obtain the flight control command timing queue, and synchronously acquire the power supply parameters mapped in the time domain; Power supply excitation analysis is performed on power supply parameters, including analyzing the deviation of power supply parameters from steady-state operation under multiple observation perspectives, identifying power supply excitation phenomena, and locking the set of excitation commands that induce the power supply excitation phenomena. The set of excitation commands consists of a single flight control command or a combination of multiple flight control commands. Determine the variation perspective corresponding to the excitation command set, and construct the power supply parameter sample corresponding to each variation perspective; Obtain a predetermined field-of-view image that is mapped to the time domain of the flight control command timing queue, and determine the scene type based on the predetermined field-of-view image to perform command pattern analysis, including determining the flight control command triggered in the corresponding response window after the scene type is verified, so as to determine the associated potential command pool in each scene type. During the drone's flight, the corresponding scenario type of the drone is continuously verified, and the power supply excitation tendency is verified based on the associated potential command pool matched with the scenario type. Based on the verification results, power supply parameters are intervened, including determining the adjustment strategy based on the anomaly perspective corresponding to the excitation command set in the potential command pool, and verifying the power supply intervention effect based on the power supply parameters corresponding to the anomaly perspective monitored in real time and the sample power supply parameters.

[0007] Furthermore, the process of analyzing the deviation of power supply parameters from steady-state operation under multiple observation perspectives and determining the power supply excitation phenomenon includes: Determine the average power supply parameters for each observation angle during the UAV's historical flight to determine the anchoring parameters corresponding to each observation angle; Determine the offset of the power supply parameters relative to the anchoring parameters for each observation angle; If the power supply parameters meet the excitation conditions, then it is determined that a power supply excitation phenomenon exists. The excitation conditions include that the offset corresponding to any observation angle is greater than a preset offset threshold and the power supply protection is not triggered. The power supply parameters include voltage drop depth, peak current ratio and power change rate. The observation angles include voltage drop angle, current surge angle and power surge angle.

[0008] Furthermore, the process of locking the set of excitation commands that induce the power supply excitation phenomenon includes, Determine the time domain segment in which the power supply excitation phenomenon occurs, and based on the flight control command timing queue, read several flight control commands executed by the UAV within the time domain segment to construct a temporary flight control excitation command set; Based on the flight control command timing queue, when the set of temporary flight control excitation commands appears, the prior probability of the existence of a power supply excitation phenomenon is read; The set of temporary flight control excitation commands that meet the confidence conditions is determined as the excitation command set; The confidence condition is that the prior probability is greater than a preset confidence prior probability threshold.

[0009] Furthermore, the process of determining the variation perspective corresponding to the excitation command set and constructing the power supply parameter samples corresponding to each variation perspective includes, The observation perspective that satisfies the excitation conditions is taken as the mutation perspective; Record the power supply parameters under different perspectives as power supply parameter samples.

[0010] Further, the process of acquiring a predetermined field-of-view image mapped to the time domain of the flight control command timing queue, and determining the scene type based on the predetermined field-of-view image, includes: The scene visual features of each dimension are extracted based on the predetermined field of view image, including texture complexity, depth edge density and obstacle occupancy. Determine the distribution range of scene visual features in each dimension, and determine the combination of distribution ranges; The corresponding scene type is determined based on the combination of distribution ranges; The scene type and distribution range are matched one-to-one, and the predetermined field of view is the field of view of the drone's flight direction.

[0011] Furthermore, the process of determining the flight control commands triggered within the response window corresponding to the verified scenario type, and thus determining the associated pool of potential subsequent commands for each scenario type, includes: Once the same scenario type is determined, the flight control commands within the corresponding time window are statistically analyzed, and the set of excitation commands within them is identified. Confidence verification is performed on each set of stimulus instructions, and the set of stimulus instructions that passes the verification is placed into the post-potential instruction pool corresponding to the scenario type. The confidence verification includes the fact that the probability of the set of excitation instructions appearing within the corresponding time window after several scene types have been verified is greater than a predetermined confidence prior probability threshold, and the post-potential instruction pool is only used to store the set of excitation instructions.

[0012] Furthermore, the process of verifying the power supply excitation tendency based on the associated potential instruction pool associated with scene type matching includes, If the potential instruction pool contains a set of excitation instructions, then a power supply excitation tendency is determined. When it is verified that there is a tendency for power supply excitation, it is determined that power supply parameters need to be intervened.

[0013] Furthermore, the process of determining the adjustment strategy based on the variation perspective corresponding to the set of excitation instructions in the potential instruction pool includes: The adjustment strategy to be triggered is determined based on the perspective of mutation; Each variation perspective is matched with a corresponding adjustment strategy.

[0014] Furthermore, the process of verifying the effectiveness of power supply intervention based on the power supply parameters corresponding to the anomalies observed in real-time monitoring and the sample power supply parameters includes: From the perspective of statistical anomalies, the normal distribution of the sample power supply parameters is used to determine the confidence interval of the sample power supply parameters; If the current power supply parameters deviate from the confidence interval, the power supply intervention is deemed effective.

[0015] On the other hand, a system for monitoring and managing the power supply of unmanned aerial vehicles (UAVs) is also provided, comprising: The instruction reading module is used to read the UAV flight control instructions, obtain the flight control instruction timing queue, and synchronously acquire the power supply parameters mapped in the time domain; The excitation analysis module is used to analyze the power supply excitation for power supply parameters, including analyzing the deviation of power supply parameters from steady-state operation from multiple observation perspectives, determining the power supply excitation phenomenon, and locking the set of excitation commands that induce the power supply excitation phenomenon. The sample determination module is used to determine the variation perspective corresponding to the excitation command set and to construct the power supply parameter sample corresponding to each variation perspective. The post-command analysis module is used to obtain a predetermined field-of-view image that is mapped to the time domain of the flight control command timing queue, and to determine the scene type based on the predetermined field-of-view image in order to perform command pattern analysis. This includes determining the flight control command triggered in the corresponding response window after the scene type is verified, so as to determine the post-potential command pool associated with each scene type. The incentive verification module is used to continuously verify the scene type corresponding to the drone during the drone's flight, and verify the power supply incentive tendency based on the associated potential command pool matched with the scene type. The power supply management module is used to intervene in power supply parameters based on the verification results. This includes determining the adjustment strategy based on the anomaly perspective corresponding to the excitation instruction set in the potential instruction pool, and verifying the power supply intervention effect based on the power supply parameters corresponding to the anomaly perspective and the sample power supply parameters monitored in real time.

[0016] Compared with existing technologies, this invention acquires the flight control command timing queue and power supply parameters, performs power supply excitation analysis based on power supply hyperparameters, obtains the excitation command set, and determines the corresponding variable perspective and power supply parameter samples. Simultaneously, it acquires a predetermined field-of-view image mapped to the flight control command timing queue in the time domain, determines the scene type based on the predetermined field-of-view image, performs command pattern analysis, and determines a pool of potential subsequent commands. Subsequently, during UAV flight, the potential subsequent command pool is matched according to the determined scene type to verify the power supply excitation tendency and intervene in the power supply parameters. During the intervention process, the invention utilizes the setting and adjustment strategy of the variable perspective corresponding to the excitation commands to verify the power supply intervention effect. This invention leverages the correlation between the changes in power supply parameters caused by the combination of flight control commands and pre-judgment of flight scenarios to intervene in power supply parameters in advance, improving the safety and reliability of power supply management.

[0017] In particular, this invention performs power supply excitation analysis to reflect the abnormal changes in power supply parameters caused by excitation under specific commands or combinations of commands. In practice, the continuous execution of certain commands during UAV flight may excite significant changes in power supply parameters. During these stages, compared to the normal operation of the UAV, power supply parameters are more data-representative and easier to detect power supply anomalies. Furthermore, this invention monitors power supply parameters from multiple observation perspectives, reflecting transient deviations in power supply parameters from multiple dimensions, accurately identifying the excitation command combinations that induce power supply excitation, and determining the anomaly perspectives corresponding to different excitation command combinations. This provides data support for subsequent power supply parameter intervention and power supply strategy adjustment, improving the reliability of power supply parameter adjustment.

[0018] In particular, this invention determines the scene type based on a predetermined field of view image and analyzes the command patterns. In practice, UAVs often execute similar commands when facing certain typical scenarios. For example, they need to frequently adjust their attitude when traversing, and they frequently perform obstacle avoidance maneuvers when there are multiple obstacles. Therefore, there is a correlation between the scene and the UAV's flight control commands. Based on this, this invention considers using scene visual features to distinguish typical scenes from multiple dimensions and determine the associated potential command pool. The potential command pool reflects the combination of flight control commands that the UAV may be highly inclined to execute and induce power supply excitation under the corresponding scene type. This provides effective data support for the subsequent early intervention of power supply parameters by utilizing scene type during UAV flight, thereby proactively and adaptively intervening in power supply parameters and improving the safety and reliability of power supply management.

[0019] In particular, when intervening in power supply, this invention uses the current scenario type to find the corresponding set of excitation instructions and determine the change perspective, and makes adaptive setting and adjustment strategies. It makes proactive preventive adjustments in the face of different scenario types, and then uses the power supply parameters corresponding to the change perspective to verify the reliability of the adjustment, thereby improving the safety and reliability of power supply management. Attached Figure Description

[0020] Figure 1 A schematic diagram illustrating the steps of the unmanned aerial vehicle (UAV) power supply monitoring and management method according to an embodiment of the invention; Figure 2 This is a logic block diagram for determining the presence of a power supply excitation phenomenon in an embodiment of the invention. Figure 3 A logic block diagram for determining the power supply excitation set in an embodiment of the invention; Figure 4 This is a logic block diagram for verifying the power supply excitation tendency in an embodiment of the invention. Detailed Implementation

[0021] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0022] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0023] Please see Figure 1 The diagram shown illustrates the steps of a drone power supply monitoring and management method according to an embodiment of the invention. The drone power supply monitoring and management method according to this embodiment includes: Step S1: Read the UAV flight control commands, obtain the flight control command timing queue, and synchronously acquire the power supply parameters mapped in the time domain; Step S2: Perform power supply excitation analysis on the power supply parameters, including analyzing the deviation of power supply parameters from steady-state operation under multiple observation perspectives, determining the power supply excitation phenomenon, and locking the excitation command set that induces the power supply excitation phenomenon. The excitation command set consists of a single flight control command or a combination of multiple flight control commands. Step S3: Determine the variation perspective corresponding to the excitation command set, and construct the power supply parameter sample corresponding to each variation perspective; Step S4: Obtain a predetermined field of view image that is mapped to the time domain of the flight control command timing queue, and determine the scene type based on the predetermined field of view image to perform command pattern analysis, including determining the flight control command triggered in the corresponding response window after the scene type is verified, so as to determine the associated potential command pool in each scene type. Step S5: Continuously verify the scene type corresponding to the drone during the drone flight, and verify the power supply excitation tendency based on the associated potential command pool matched with the scene type. Step S6 involves intervening in the power supply parameters based on the verification results. This includes determining the adjustment strategy based on the anomaly perspective corresponding to the excitation command set in the potential command pool, and verifying the power supply intervention effect based on the power supply parameters corresponding to the anomaly perspective monitored in real time and the sample power supply parameters.

[0024] Specifically, this invention does not limit the model of the UAV. The flight control commands can be obtained by listening to and parsing the command messages output by the flight control system in real time through the UAV flight control system bus, or by obtaining them from the output end of the flight control commands, or in other forms.

[0025] Specifically, in implementation, the form of flight control commands is not limited. To facilitate statistical calculations, flight control commands can be pre-classified into several types, for example... Flight attitude commands: including climb, descent, left roll, right roll, dive, emergency braking and hovering, etc.; heading and maneuver commands: including sharp left turn, sharp right turn, yaw in place, U-turn and return, etc.; speed adjustment commands: including full speed forward, deceleration cruise, stationary hovering, etc.; mission payload commands: including gimbal pitch lock, cargo release mechanism trigger, high-power image transmission, etc.; flight mode commands: including attitude mode switching, altitude hold mode switching, return to home trigger, etc. Of course, the above instruction classification methods are merely examples. Those skilled in the art can adopt other instruction classification standards according to actual needs, such as classification based on instruction function, actuator, or action range. In specific implementations, it is also possible not to merge instruction types, but to directly construct a flight control instruction timing queue based on the unique identifier of each instruction, such as instruction ID, instruction code, or action descriptor, which will not be elaborated further.

[0026] Specifically, please refer to Figure 2 As shown, it is a logic block diagram for determining the existence of a power supply excitation phenomenon according to an embodiment of the invention. The process of analyzing the deviation of power supply parameters from steady-state operation under multiple observation perspectives and determining the power supply excitation phenomenon includes, Determine the average power supply parameters for each observation angle during the UAV's historical flight to determine the anchoring parameters corresponding to each observation angle; Determine the offset of the power supply parameters relative to the anchoring parameters for each observation angle; If the power supply parameters meet the excitation conditions, then it is determined that a power supply excitation phenomenon exists. The excitation conditions include that the offset corresponding to any observation angle is greater than a preset offset threshold and the power supply protection is not triggered. The power supply parameters include voltage drop depth, peak current ratio and power change rate. The observation angles include voltage drop angle, current surge angle and power surge angle.

[0027] Specifically, the purpose of setting observation perspectives is to capture power supply parameters from different angles: observing the voltage drop depth from the voltage drop perspective, observing the peak current ratio from the current surge perspective, and observing the power change rate from the power surge perspective. The anchoring parameters reflect the average level of power supply parameters under the corresponding observation viewpoint. The corresponding anchoring parameters need to be determined in advance for each observation viewpoint. During implementation, Set the mean value of voltage drop depth under the voltage drop perspective as the anchoring parameter corresponding to the voltage drop perspective; Set the mean of the peak current ratio under the current impact perspective as the anchoring parameter corresponding to the current impact perspective. Set the mean of the rate of change of power under the power mutation perspective as the anchoring parameter corresponding to the power mutation perspective; Voltage sag depth is the difference between the lowest voltage and the nominal voltage within a unit time window; The peak current ratio is the ratio of the peak current to the average current within a unit time window. The rate of change of power is the ratio of the peak rate of change of power to the average rate of change of power within a unit time window. The unit time window is 1 second.

[0028] Specifically, the offset threshold is set based on the anchoring parameters, with the aim of capturing situations where power supply parameters change significantly from various observation angles. In implementation... The offset of the power supply parameters relative to the anchoring parameters under each observation view during the historical flight of the UAV is pre-statistically calculated, and the average offset is calculated to reflect the average level of the corresponding offset under each observation view. The offset threshold corresponding to the observation view is set as the average offset and the preset margin coefficient. The preset margin coefficient is selected in the range [1.15, 1.3], preferably 1.2, to filter out cases where the offset is at a high level and the power supply protection is not triggered. The purpose of selecting this range is to preset the margin coefficient. If it is less than 1.15, normal fluctuations will be misjudged. If it is too large, many anomalies caused by power supply excitation phenomena will be missed.

[0029] Specifically, please refer to Figure 3 As shown, it is a logic block diagram for determining the power supply excitation set according to an embodiment of the invention. The process of locking the set of excitation commands that induce the power supply excitation phenomenon includes, Determine the time domain segment in which the power supply excitation phenomenon occurs, and based on the flight control command timing queue, read several flight control commands executed by the UAV within the time domain segment to construct a temporary flight control excitation command set; Based on the flight control command timing queue, when the set of temporary flight control excitation commands appears, the prior probability of the existence of a power supply excitation phenomenon is read; The set of temporary flight control excitation commands that meet the confidence conditions is determined as the excitation command set; The confidence condition is that the prior probability is greater than a preset confidence prior probability threshold.

[0030] Specifically, the purpose of setting a confidence prior probability threshold is to reflect that the set of temporary flight control excitation commands is not random. Therefore, the confidence prior probability threshold is in the range [0.3, 0.6], preferably 0.5. The set of temporary flight control excitation commands that occurs occasionally is filtered out with 0.3 as the lower limit, and the confidence prior probability threshold is set too high so that the excitation command set cannot be identified.

[0031] Specifically, the process of determining the variation perspective corresponding to the excitation command set and constructing the power supply parameter samples corresponding to each variation perspective includes: The observation perspective that satisfies the excitation conditions is taken as the mutation perspective; Record the power supply parameters under different perspectives as power supply parameter samples.

[0032] This invention analyzes power supply excitation to reflect the abnormal changes in power supply parameters caused by specific commands or combinations of commands. In practice, the continuous execution of certain commands during UAV flight may cause significant changes in power supply parameters. During these phases, power supply parameters are more representative of data and easier to detect power supply anomalies compared to normal UAV operation. Furthermore, this invention monitors power supply parameters from multiple observation perspectives, reflecting transient deviations in power supply parameters from multiple dimensions, accurately identifying the combinations of excitation commands that induce power supply excitation, and determining the anomaly perspectives corresponding to different combinations of excitation commands. This provides data support for subsequent power supply parameter intervention and power supply strategy adjustments, improving the reliability of power supply parameter adjustments.

[0033] Specifically, the process of acquiring a predetermined field-of-view image mapped to the time domain of the flight control command timing queue, and determining the scene type based on the predetermined field-of-view image, includes: The scene visual features of each dimension are extracted based on the predetermined field of view image, including texture complexity, depth edge density and obstacle occupancy. Determine the distribution range of scene visual features in each dimension, and determine the combination of distribution ranges; The corresponding scene type is determined based on the combination of distribution ranges; The scene type and distribution range are matched one-to-one, and the predetermined field of view is the field of view of the drone's flight direction.

[0034] Specifically, there is no limitation on the method of calculating texture complexity. In practice, the gray-level co-occurrence matrix can be determined after the predetermined field image is converted into a grayscale image, and the entropy value can be used as the texture complexity. Of course, those skilled in the art can also use other methods that can reflect texture complexity, which will not be elaborated here.

[0035] Specifically, when acquiring depth edge density, an airborne binocular stereo vision or monocular depth estimation network is used to obtain a depth map aligned with the pixels of a predetermined field of view image. The value of each pixel in the depth map represents the distance from that point to the UAV. The Sobel gradient operator is used to calculate the vertical and horizontal gradient values ​​of the depth map in the vertical and horizontal directions, respectively. The arithmetic square root of the sum of the squares of the vertical and horizontal gradient values ​​is then calculated to obtain the depth gradient magnitude. The depth gradient magnitude reflects the degree of change of the depth value along the spatial direction in the neighborhood of the pixel position. The depth edge density is calculated as the ratio of the number of pixels with a depth gradient magnitude greater than a predetermined depth gradient threshold to the total number of pixels. A higher depth edge density indicates the presence of numerous abrupt depth boundaries within the viewport, meaning that the distances between objects and between objects and the background change frequently and drastically. Typical high-value scenarios include traversing forests, flying close to building clusters, and indoor or narrow alley environments.

[0036] The predetermined depth gradient threshold is set to 0.5m, with the aim of capturing the boundaries of obstacles that have real significance in physical space, while avoiding capturing boundaries that are too small. The obstacle occupancy rate is the ratio of the area of ​​the obstacle in the predetermined field of view to the area of ​​the predetermined field of view.

[0037] It is understandable that the texture complexity, depth edge density, and obstacle occupancy will be in different ranges under different scene types. Therefore, scene types can be distinguished based on the distribution range of scene visual features in each dimension.

[0038] In practice, the visual features of each scene can be divided into three ranges to avoid excessive division leading to fragmentation of scene types. The range boundaries are determined based on the maximum and minimum values ​​of the visual features of each scene. The interval formed by the maximum and minimum values ​​is divided into three continuous sub-intervals, and each sub-interval is the range after division.

[0039] This invention determines scene types based on a predetermined field-of-view image and analyzes command patterns. In practice, UAVs typically execute similar commands when facing certain typical scenarios. For example, they need to frequently adjust their attitude when traversing obstacles and frequently perform obstacle avoidance maneuvers when there are multiple obstacles. Therefore, there is a correlation between the scene and the UAV's flight control commands. Based on this, this invention considers using scene visual features to distinguish typical scenes from multiple dimensions and determine the associated potential command pool. The potential command pool reflects the combination of flight control commands that the UAV is likely to execute and induce power supply excitation under the corresponding scene type. This provides effective data support for proactively intervening in power supply parameters during UAV flight by utilizing scene types, thereby proactively and adaptively intervening in power supply parameters and improving the safety and reliability of power supply management.

[0040] Specifically, the process of determining the flight control commands triggered within the response window corresponding to the verified scenario type, and thus determining the associated pool of potential subsequent commands for each scenario type, includes: Once the same scenario type is determined, the flight control commands within the corresponding time window are statistically analyzed, and the set of excitation commands within them is identified. Confidence verification is performed on each set of stimulus instructions, and the set of stimulus instructions that passes the verification is placed into the post-potential instruction pool corresponding to the scenario type. The confidence verification includes the fact that the probability of the set of excitation instructions appearing within the corresponding time window after several scene types have been verified is greater than a predetermined confidence prior probability threshold, and the post-potential instruction pool is only used to store the set of excitation instructions.

[0041] Specifically, the time window can be set to 5 seconds to extract the drone's response within a short period of time after the scene type is determined.

[0042] Specifically, please refer to Figure 4 As shown, this is a logic block diagram of the verification of power supply excitation tendency according to an embodiment of the invention. Based on the downstream potential instruction pool associated with scene type matching, the process of verifying power supply excitation tendency includes, If the potential instruction pool contains a set of excitation instructions, then a power supply excitation tendency is determined. When it is verified that there is a tendency for power supply excitation, it is determined that power supply parameters need to be intervened.

[0043] Specifically, the process of determining the adjustment strategy based on the anomaly perspective corresponding to the set of excitation instructions in the potential instruction pool includes the following steps: The adjustment strategy to be triggered is determined based on the perspective of mutation; Each variation perspective is matched with a corresponding adjustment strategy.

[0044] Specifically, no specific limitations are made on the adjustment strategies for matching different perspectives. Those skilled in the art can set corresponding adjustment strategies according to the actual situation. In implementation, optional... For the perspective of voltage drop, the voltage of the buoyancy system can be increased without affecting the operation of the avionics system, so as to buffer the voltage drop. The voltage after buoyancy must be lower than the maximum rated voltage of the power system. For the perspective of current impact, the operation of redundant loads should be restricted. Redundant loads are electrical loads of the non-powered system of the UAV, such as LED fill lights, gimbal drive motors, and image transmission modules. The fill light intensity should be appropriately reduced, the rotation speed of the gimbal drive motors should be appropriately limited, and the power of the image transmission module should be reduced.

[0045] For a power surge perspective, the power rise rate of the propulsion system is limited. For example, the original power rise rate is limited to 90%, so that the power transitions to the target value at a gentler slope. In special circumstances, this does not affect the command response time and has little impact on flight performance. By using a subtle power limit, safety is improved.

[0046] Specifically, the process of verifying the effectiveness of power supply intervention based on real-time monitoring of anomalies and corresponding power supply parameters versus sample power supply parameters includes: From the perspective of statistical anomalies, the normal distribution of the sample power supply parameters is used to determine the confidence interval of the sample power supply parameters; If the current power supply parameters deviate from the confidence interval, the power supply intervention is deemed effective.

[0047] In practice, the power supply parameters should be reduced accordingly after intervention. Therefore, the lower limit of the confidence interval is considered as the confidence interval.

[0048] An early warning can be issued if the intervention is deemed ineffective.

[0049] When intervening in power supply, this invention uses the current scenario type to find the corresponding set of excitation instructions and determine the perspective of change, and then makes adaptive adjustment strategies. It makes proactive preventive adjustments in the face of different scenario types, and then uses the power supply parameters corresponding to the perspective of change to verify the reliability of the adjustment, thereby improving the safety and reliability of power supply management.

[0050] This embodiment also provides a system for applying the unmanned aerial vehicle (UAV) power supply monitoring and management method, including... The instruction reading module is used to read the UAV flight control instructions, obtain the flight control instruction timing queue, and synchronously acquire the power supply parameters mapped in the time domain; The excitation analysis module is used to analyze the power supply excitation for power supply parameters, including analyzing the deviation of power supply parameters from steady-state operation from multiple observation perspectives, determining the power supply excitation phenomenon, and locking the set of excitation commands that induce the power supply excitation phenomenon. The sample determination module is used to determine the variation perspective corresponding to the excitation command set and to construct the power supply parameter sample corresponding to each variation perspective. The post-command analysis module is used to obtain a predetermined field-of-view image that is mapped to the time domain of the flight control command timing queue, and to determine the scene type based on the predetermined field-of-view image in order to perform command pattern analysis. This includes determining the flight control command triggered in the corresponding response window after the scene type is verified, so as to determine the post-potential command pool associated with each scene type. The incentive verification module is used to continuously verify the scene type corresponding to the drone during the drone's flight, and verify the power supply incentive tendency based on the associated potential command pool matched with the scene type. The power supply management module is used to intervene in power supply parameters based on the verification results. This includes determining the adjustment strategy based on the anomaly perspective corresponding to the excitation instruction set in the potential instruction pool, and verifying the power supply intervention effect based on the power supply parameters corresponding to the anomaly perspective and the sample power supply parameters monitored in real time.

[0051] It should be noted that the multiple functional modules involved in this application are only a logical division based on the functions implemented according to the present invention, and are not a strict limitation on the physical structure; in practical applications, the above functional modules can be implemented by one or more integrated circuits, a processor executing program code in memory, or a combination of the above devices.

[0052] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A method for monitoring and managing the power supply of unmanned aerial vehicles (UAVs), characterized in that, include: Read the UAV flight control commands, obtain the flight control command timing queue, and synchronously acquire the power supply parameters mapped in the time domain; Power supply excitation analysis is performed on power supply parameters, including analyzing the deviation of power supply parameters from steady-state operation under multiple observation perspectives, identifying power supply excitation phenomena, and locking the set of excitation commands that induce the power supply excitation phenomena. The set of excitation commands consists of a single flight control command or a combination of multiple flight control commands. Determine the variation perspective corresponding to the excitation command set, and construct the power supply parameter sample corresponding to each variation perspective; A predetermined field-of-view image mapped to the time domain of the flight control command timing queue is acquired. The scene type is determined based on the predetermined field-of-view image to perform command pattern analysis. This includes determining the flight control commands triggered within the corresponding response window after the scenario type is verified, in order to determine the pool of potential subsequent commands associated with each scenario type; During the drone's flight, the corresponding scenario type of the drone is continuously verified, and the power supply excitation tendency is verified based on the associated potential command pool matched with the scenario type. Based on the verification results, power supply parameters are intervened, including determining the adjustment strategy based on the anomaly perspective corresponding to the excitation command set in the potential command pool, and verifying the power supply intervention effect based on the power supply parameters corresponding to the anomaly perspective monitored in real time and the sample power supply parameters.

2. The UAV power supply monitoring and management method according to claim 1, characterized in that, The process of analyzing deviations of power supply parameters from steady-state operation under multiple observation perspectives and determining power supply excitation phenomena includes: Determine the average power supply parameters for each observation angle during the UAV's historical flight to determine the anchoring parameters corresponding to each observation angle; Determine the offset of the power supply parameters relative to the anchoring parameters for each observation angle; If the power supply parameters meet the excitation conditions, then it is determined that a power supply excitation phenomenon exists. The excitation conditions include that the offset corresponding to any observation angle is greater than a preset offset threshold and the power supply protection is not triggered. The power supply parameters include voltage drop depth, peak current ratio and power change rate. The observation angles include voltage drop angle, current surge angle and power surge angle.

3. The UAV power supply monitoring and management method according to claim 2, characterized in that, The process of locking the set of excitation commands that induce the power supply excitation phenomenon includes, Determine the time domain segment in which the power supply excitation phenomenon occurs, and based on the flight control command timing queue, read several flight control commands executed by the UAV within the time domain segment to construct a temporary flight control excitation command set; Based on the flight control command timing queue, when the set of temporary flight control excitation commands appears, the prior probability of the existence of a power supply excitation phenomenon is read; The set of temporary flight control excitation commands that meet the confidence conditions is determined as the excitation command set; The confidence condition is that the prior probability is greater than a preset confidence prior probability threshold.

4. The UAV power supply monitoring and management method according to claim 3, characterized in that, The process of determining the variation perspective corresponding to the excitation command set and constructing the power supply parameter samples corresponding to each variation perspective includes: The observation perspective that satisfies the excitation conditions is taken as the mutation perspective; Record the power supply parameters under different perspectives as power supply parameter samples.

5. The method for monitoring and managing the power supply of unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The process of acquiring a predetermined field-of-view image mapped to the time domain of the flight control command timing queue, and determining the scene type based on the predetermined field-of-view image, includes: The scene visual features of each dimension are extracted based on the predetermined field of view image, including texture complexity, depth edge density and obstacle occupancy. Determine the distribution range of scene visual features in each dimension, and determine the combination of distribution ranges; The corresponding scene type is determined based on the combination of distribution ranges; The scene type and distribution range are matched one-to-one, and the predetermined field of view is the field of view of the drone's flight direction.

6. The UAV power supply monitoring and management method according to claim 1, characterized in that, The process of determining the flight control commands triggered within the response window after the scenario type is verified, and determining the associated potential command pool for each scenario type, includes... Once the same scenario type is determined, the flight control commands within the corresponding time window are statistically analyzed, and the set of excitation commands within them is identified. Confidence verification is performed on each set of stimulus instructions, and the set of stimulus instructions that passes the verification is placed into the post-potential instruction pool corresponding to the scenario type. The confidence verification includes the fact that the probability of the set of excitation instructions appearing within the corresponding time window after several scene types have been verified is greater than a predetermined confidence prior probability threshold, and the post-potential instruction pool is only used to store the set of excitation instructions.

7. The method for monitoring and managing the power supply of unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The process of verifying the power supply excitation tendency based on the downstream potential instruction pool associated with scene type matching includes: If the potential instruction pool contains a set of excitation instructions, then a power supply excitation tendency is determined. When it is verified that there is a tendency for power supply excitation, it is determined that power supply parameters need to be intervened.

8. The method for monitoring and managing the power supply of unmanned aerial vehicles according to claim 1, characterized in that, The process of determining the adjustment strategy based on the anomaly perspective corresponding to the set of excitation instructions in the potential instruction pool includes: The adjustment strategy to be triggered is determined based on the perspective of mutation; Each variation perspective is matched with a corresponding adjustment strategy.

9. The method for monitoring and managing the power supply of unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The process of verifying the effectiveness of power supply intervention based on real-time monitoring of anomalies and corresponding power supply parameters versus sample power supply parameters includes: From the perspective of statistical anomalies, the normal distribution of the sample power supply parameters is used to determine the confidence interval of the sample power supply parameters; If the current power supply parameters deviate from the confidence interval, the power supply intervention is deemed effective.

10. A system applying the UAV power supply monitoring and management method according to any one of claims 1-9, characterized in that, include The instruction reading module is used to read the UAV flight control instructions, obtain the flight control instruction timing queue, and synchronously acquire the power supply parameters mapped in the time domain; The excitation analysis module is used to analyze the power supply excitation for power supply parameters, including analyzing the deviation of power supply parameters from steady-state operation from multiple observation perspectives, determining the power supply excitation phenomenon, and locking the set of excitation commands that induce the power supply excitation phenomenon. The sample determination module is used to determine the variation perspective corresponding to the excitation command set and to construct the power supply parameter sample corresponding to each variation perspective. The post-command analysis module is used to obtain a predetermined field-of-view image that is mapped to the time domain of the flight control command timing queue, and to determine the scene type based on the predetermined field-of-view image in order to perform command pattern analysis. This includes determining the flight control command triggered in the corresponding response window after the scene type is verified, so as to determine the post-potential command pool associated with each scene type. The incentive verification module is used to continuously verify the scene type corresponding to the drone during the drone's flight, and verify the power supply incentive tendency based on the associated potential command pool matched with the scene type. The power supply management module is used to intervene in power supply parameters based on the verification results. This includes determining the adjustment strategy based on the anomaly perspective corresponding to the excitation instruction set in the potential instruction pool, and verifying the power supply intervention effect based on the power supply parameters corresponding to the anomaly perspective and the sample power supply parameters monitored in real time.

Images