Unmanned aerial vehicle express delivery system and method based on artificial intelligence

Abstract

The invention discloses an unmanned aerial vehicle express delivery system and an unmanned aerial vehicle express delivery method based on artificial intelligence, and belongs to the technical field of smart city logistics. Whether the battery output power of the unmanned aerial vehicle is abnormal or not is judged according to the linear regression model, so that whether the unmanned aerial vehicle meets the express delivery task or not is judged, the situation that the unmanned aerial vehicle is damaged due to the fact that battery data monitoring is not fine enough is avoided, and the package transportation success rate and user experience are improved; and the unmanned aerial vehicle can complete the express delivery task.

Description

Technical Field

[0001] This invention relates to the field of smart city logistics technology, specifically to an artificial intelligence-based drone delivery system and method. Background Technology

[0002] With the rapid development of the e-commerce industry, the volume of express delivery business has exploded. The traditional ground logistics system has gradually revealed structural pain points such as low efficiency and high cost in the "last mile" delivery, making the innovation of intelligent logistics technology represented by drones an important direction for the industry to break through the deadlock.

[0003] Drone-based intelligent delivery technology, by integrating drone platforms, navigation control, communication networks, and intelligent algorithms, possesses significant advantages in overcoming terrain obstacles and achieving contactless delivery, providing an innovative technological path to solve the traditional last-mile delivery challenges. However, two core technological bottlenecks exist in practical applications: First, the dynamic changes in the package load cause real-time fluctuations in the drone's weight, leading to non-linear changes in power consumption and battery depletion rates, resulting in insufficient endurance; second, continuous high-load operation of the motor can easily cause overheating failures, increasing the risk of crashes. These issues make it difficult for drones to meet the stability and reliability requirements of commercial operation. Therefore, developing precise drone power allocation strategies and real-time power consumption monitoring methods has become a key research direction for promoting the practical application of drone delivery technology. Summary of the Invention

[0004] The purpose of this invention is to provide an artificial intelligence-based drone delivery system and method to solve the problems mentioned in the background section.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: An AI-based drone delivery method, which provides the following solution: Step S1: The package is weighed, the tracking number is bound by the barcode, and the data is uploaded to the cloud database. The package is then assigned to the corresponding drone delivery warehouse according to the destination. Step S1-1: Scan the package information and upload the information to the cloud database; Step S1-2: Distribute the packages to the corresponding drone delivery warehouses according to their destinations.

[0006] Step S2: Collect the battery consumption curve and drone output power curve of the drone under full battery power during the delivery of various packages of different weights along the same route; Step S2-1: Using drones of the same model and parameters, monitor the battery consumption and output power of the drones when performing drone delivery tasks on the same road segment under different weights. Step S2-2: Divide the drone's load weight range, and plot the battery consumption curve and output power curve based on the monitored changes in drone battery power consumption and output power over time.

[0007] By using battery consumption curves and output power curves, the data can be visualized, laying the foundation for analyzing the impact of payload on drone performance.

[0008] Step S3: Divide the collected drone battery power consumption curves and output power curves under different load weights into intervals. Divide the collected drone battery power consumption curves and output power curves into four stages according to the drone's flight state: climb, descent, level flight and hovering. Step S3-1: Divide the battery energy consumption curve and output power curve under different loading weight conditions into loading weight ranges; Step S3-2: Use the 3σ principle to remove abrupt changes in power and energy from the collected battery consumption curve and output power curve, use digital filtering technology to eliminate high-frequency noise from the drone, and align the timestamps using the Dynamic Time Warping (DTW) algorithm. Step S3-3: Identify the UAV's flight status based on the kinematic characteristics of each flight state, specifically as follows: The hovering output power curve is characterized by maintaining a constant output power at a baseline that overcomes the aircraft's weight, with an error range of [-5%, +5%]. The amplitude fluctuation of the power curve does not exceed ±5% of the baseline constant value. The hovering battery consumption curve is characterized by the slope of the battery consumption curve. The error of the value is ≤5%; The output power curve in level flight is characterized by a standard deviation of ≤8% for the curve fluctuation amplitude within the level flight range. The battery consumption curve is characterized by... The error of the value is ≤5%; The output power curve during the climbing phase is characterized by a step-like rise and a value 20% higher than the baseline constant value, with the fluctuation range of the curve within the range of 20% higher than the baseline constant value [-5%, +5%]. At the beginning of the climb, there is a step-like rise, and after the step-like rise is completed, the power is maintained within the range of 20% higher than the baseline constant value [-5%, +5%]. The output power curve in the declining state exhibits a stepped decline and a state where it is 20% below the baseline constant value and the curve fluctuation range is within the range of -3% to +3% below the baseline constant value.

[0009] Based on two dimensions—different payloads and different flight states—this study delves into the changes in battery consumption and output power of drones under different payloads and flight states, providing more reliable data support for performance optimization and practical application of drone delivery.

[0010] Step S4: Establish a standard battery consumption rate and standard output power for multiple drone battery consumption curves and output power curves within each interval; Step S4-1: Perform big data model analysis on the battery consumption curves and output power curves of multiple drones in each interval; Step S4-2: Based on the battery consumption curve analyzed by the big data model, determine the standard battery consumption rate for different flight states within each payload weight range of the drone: ; The average rate in each interval reflects the power consumption characteristics of the drone under specific flight conditions, and is a direct quantification of actual flight behavior, where V... avg Q represents the average battery power consumption rate under different flight conditions. initial Q represents the initial battery level when the aircraft enters flight mode. end The current battery level indicates the end of flight. Step S4-3: Based on the well-analyzed output power curve of the big data model, determine the standard output power of the UAV under different flight conditions in each load weight range; The standard output power calculation formula is: ; ; In the formula P max P represents the peak power under a specific flight condition. avg P represents average power. t P represents the power at each moment within the interval, n is the number of data points, i is the start time of a specific flight state, m is the number of continuous power data points in the interval from i to i+m-1, m=n, t This represents the power at each moment.

[0011] Step S5: Install a monitor inside the drone to monitor the drone's flight status, weight, and battery level in real time, and calculate the battery consumption rate and output power of the drone in the current state based on the monitored flight status. Step S5-1: Set up a monitor for the drone. The monitor will monitor the drone's flight status, weight, and battery level in real time. Step S5-2: Based on the maximum load mass of the drone, the integrated strain gauge weighing sensor module in the monitor sets a weight range with 10kg as the boundary. The integrated strain gauge weighing sensor determines the weight range based on the weight of the drone it monitors. Step S5-3: After the IMU inertial unit module in the monitor detects a change in the drone's flight status, the IMU inertial unit module sends information to the BMS system module in the monitor. After receiving the information, the BMS system monitors the drone's battery consumption rate and output power.

[0012] By monitoring relevant data from various modules of the drone, a systematic improvement has been achieved in the drone's safety, reliability, energy efficiency, and mission execution capabilities, providing crucial support for the large-scale application of drones in scenarios such as express delivery.

[0013] Step S6: When the monitor detects abnormal output power and the abnormality occurs multiple times in a row, the task is terminated and the device is returned to the station for maintenance. Step S6-1: The BMS system in the monitor monitors the output power of the UAV in real time. If the output power detected by the monitor exceeds the standard output power by 20% and the time exceeds 5 seconds, the output power is determined to be abnormal. Step S6-2: After the monitor detects an abnormal output power, the drone automatically returns to the station and sends a message to remind the station staff to carry out maintenance.

[0014] Step S7: Set a lower limit threshold for the drone battery. When the battery reaches the threshold, determine whether the battery level is sufficient for the drone's delivery and return-to-station tasks. If it is sufficient, perform the delivery task; otherwise, return to the station to recharge. Step S7-1: Divide the drone battery power into available battery power and reserved battery power, and set a lower limit threshold for battery power; Step S7-2: When the BMS system in the monitor detects that the drone's battery level has reached the lower limit threshold, it triggers the calculation of reserved battery power. The distance is calculated based on the remaining battery power and the current battery consumption rate. The distance calculation formula is as follows: ; ; In the formula t endurance For battery life, Q remain v represents the remaining battery power. Q The rate of energy consumption is represented by s, where s represents distance traveled and v represents speed. The drone determines the mission distance using its inertial navigation system. If the calculated distance is greater than the mission distance, the delivery mission continues. If the calculated distance is less than the mission distance, the drone returns to the station for charging.

[0015] Step S8: When the drone arrives at the delivery destination, the delivery receiving device set up at the window will automatically verify the drone's position. After verification, the drone will carry out the delivery work. After the drone completes the delivery task, it will send a signature information to the customer to remind the customer to sign for the package. Step S8-1: After the drone arrives at the delivery destination, the delivery receiving device sends a signal to the drone. After receiving the signal, the drone begins to verify with the delivery receiving device. After successful verification, the delivery is carried out. Step S8-2: After the drone completes the delivery, it sends a receipt notification to the customer via the network to remind them to sign for the package.

[0016] The AI-based drone delivery system includes a delivery information processing module, a data acquisition and analysis module, a real-time monitoring module, a fault handling module, a power management module, and a delivery module. The express delivery information processing module is responsible for detecting the weight of the express delivery, binding the logistics tracking number through barcode recognition and uploading it to the cloud database, and then allocating the express delivery to the corresponding drone express warehouse according to the destination. The data acquisition and analysis module is responsible for collecting the battery power consumption curve and output power curve of the drone when delivering express packages of different weights over the same distance with a full charge. It divides these curves into intervals according to different load weights and flight states (climb, descent, level flight, hovering), and also sets the standard battery consumption rate and standard output power for each interval. The real-time monitoring module is responsible for monitoring flight status, weight, and battery level in real time, and calculating the current battery consumption rate and output power based on the flight status. The fault handling module is responsible for terminating the mission and allowing the drone to return to the station for repair when abnormal output power is detected and there are multiple consecutive abnormalities. The power management module is responsible for setting a lower limit threshold for the drone battery. When the battery reaches the threshold, it determines whether the battery is sufficient for the delivery and return mission, and then decides whether to continue delivery or return to charge. The express delivery module is responsible for automatically verifying the location of the express receiving device at the window after the drone arrives at the destination. Once the verification is successful, the delivery is carried out, and a receipt reminder is sent to the customer after the delivery is completed.

[0017] Compared with the prior art, the beneficial effects of the present invention are: 1. By collecting battery power consumption curves and outputting power curves for different flight states within different load ranges, a refined analysis of energy consumption is achieved. This enables more accurate matching of the energy requirements of drones under different working states, monitoring of drone operating conditions, reducing drone failure rates, and improving express delivery efficiency.

[0018] 2. The drone monitor collects flight status, weight, battery level, and speed in real time, dynamically calculates the remaining range, and determines whether to return to base, thus preventing the drone from crashing due to insufficient battery power. Attached Figure Description

[0019] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with the implementation of the invention to explain the invention, but do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart illustrating the artificial intelligence-based drone delivery method of the present invention. Figure 2 This is a module distribution diagram of the artificial intelligence-based drone delivery system of this invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Example: Figure 1 , Figure 2 As shown, the present invention provides a technical solution; The AI-based drone delivery method includes the following steps: Step S1: The package is weighed, the tracking number is bound by the barcode, and the data is uploaded to the cloud database. The package is then assigned to the corresponding drone delivery warehouse according to the destination. Step S1-1: Scan the package information and upload the information to the cloud database; Step S1-2: Distribute the packages to the corresponding drone delivery warehouses according to their destinations.

[0022] Step S2: Collect the battery consumption curve and drone output power curve of the drone under full battery power during the delivery of various packages of different weights along the same route; Step S2-1: Using drones of the same model and parameters, monitor the battery consumption and output power of the drones when performing drone delivery tasks on the same road segment under different weights. Step S2-2: Divide the drone's load weight range, and plot the battery consumption curve and output power curve based on the monitored changes in drone battery power consumption and output power over time.

[0023] By using battery consumption curves and output power curves, the data can be visualized, laying the foundation for analyzing the impact of payload on drone performance.

[0024] Step S3: Divide the collected drone battery power consumption curves and output power curves under different load weights into intervals. Divide the collected drone battery power consumption curves and output power curves into four stages according to the drone's flight state: climb, descent, level flight and hovering. Step S3-1: Divide the battery energy consumption curve and output power curve under different loading weight conditions into loading weight ranges; Step S3-2: Use the 3σ principle to remove abrupt changes in power and energy from the collected battery consumption curve and output power curve, use digital filtering technology to eliminate high-frequency noise from the drone, and align the timestamps using the Dynamic Time Warping (DTW) algorithm. Step S3-3: Identify the UAV's flight status based on the kinematic characteristics of each flight state, specifically as follows: The hovering output power curve is characterized by maintaining a constant output power at a baseline that overcomes the aircraft's weight, with an error range of [-5%, +5%]. The amplitude fluctuation of the power curve does not exceed ±5% of the baseline constant value. The hovering battery consumption curve is characterized by the slope of the battery consumption curve. The error of the value is ≤5%; The output power curve in level flight is characterized by a standard deviation of ≤8% for the curve fluctuation amplitude within the level flight range. The battery consumption curve is characterized by... The error of the value is ≤5%; The output power curve during the climbing phase is characterized by a step-like increase and a fluctuation range of 20% above the baseline constant value [-5%, +5%]. At the beginning of the climb, there is a step-like increase, and after the step-like increase is completed, the power is maintained in the range of 20% above the baseline constant value [-5%, +5%].

[0025] The output power curve in the declining state exhibits a stepped decline and a state where it is 20% below the baseline constant value and the curve fluctuation range is within the range of -3% to +3% below the baseline constant value.

[0026] Based on two dimensions—different payloads and different flight states—this study delves into the changes in battery consumption and output power of drones under different payloads and flight states, providing more reliable data support for performance optimization and practical application of drone delivery.

[0027] Step S4: Establish a standard battery consumption rate and standard output power for multiple drone battery consumption curves and output power curves within each interval; Step S4-1: Perform big data model analysis on the battery consumption curves and output power curves of multiple drones in each interval; Step S4-2: Based on the battery consumption curve analyzed by the big data model, determine the standard battery consumption rate for different flight states within each payload weight range of the drone: ; The average rate in each interval reflects the power consumption characteristics of the drone under specific flight conditions, and is a direct quantification of actual flight behavior, where V... avg Q represents the average battery power consumption rate under different flight conditions. initial Q represents the initial battery level when the aircraft enters flight mode. end The current battery level indicates the end of flight. Step S4-3: Based on the well-analyzed output power curve of the big data model, determine the standard output power of the UAV under different flight conditions in each load weight range; The standard output power calculation formula is: ; ; In the formula P max P represents the peak power under a specific flight condition. avg P represents average power. t P represents the power at each moment within the interval, n is the number of data points, i is the start time of a specific flight state, m is the number of continuous power data points in the interval from i to i+m-1, m=n, t This represents the power at each moment.

[0028] Step S5: Install a monitor inside the drone to monitor the drone's flight status, weight, and battery level in real time, and calculate the battery consumption rate and output power of the drone in the current state based on the monitored flight status. Step S5-1: Set up a monitor for the drone. The monitor will monitor the drone's flight status, weight, and battery level in real time. Step S5-2: Based on the maximum load mass of the drone, the integrated strain gauge weighing sensor module in the monitor sets a weight range with 10kg as the boundary. The integrated strain gauge weighing sensor determines the weight range based on the weight of the drone it monitors. Step S5-3: After the IMU inertial unit module in the monitor detects a change in the drone's flight status, the IMU inertial unit module sends information to the BMS system module in the monitor. After receiving the information, the BMS system monitors the drone's battery consumption rate and output power.

[0029] By monitoring relevant data from various modules of the drone, a systematic improvement has been achieved in the drone's safety, reliability, energy efficiency, and mission execution capabilities, providing crucial support for the large-scale application of drones in scenarios such as express delivery.

[0030] Step S6: When the monitor detects abnormal output power and the abnormality occurs multiple times in a row, the task is terminated and the device is returned to the station for maintenance. Step S6-1: The BMS system in the monitor monitors the output power of the UAV in real time. If the output power detected by the monitor exceeds the standard output power by 20% and the time exceeds 5 seconds, the output power is determined to be abnormal. Step S6-2: After the monitor detects an abnormal output power, the drone automatically returns to the station and sends a message to remind the station staff to carry out maintenance.

[0031] Step S7: Set a lower limit threshold for the drone battery. When the battery reaches the threshold, determine whether the battery level is sufficient for the drone's delivery and return-to-station tasks. If it is sufficient, perform the delivery task; otherwise, return to the station to recharge. Step S7-1: Divide the drone battery power into available battery power and reserved battery power, and set a lower limit threshold for battery power; Step S7-2: When the BMS system in the monitor detects that the drone's battery level has reached the lower limit threshold, it triggers the calculation of reserved battery power. The distance is calculated based on the remaining battery power and the current battery consumption rate. The distance calculation formula is as follows: ; ; In the formula t endurance For battery life, Q remain v represents the remaining battery power. Q The power consumption rate is represented by s, where s represents distance traveled and v represents speed.

[0032] The drone determines the mission distance using its inertial navigation system. If the calculated distance is greater than the mission distance, the delivery mission continues. If the calculated distance is less than the mission distance, the drone returns to the station for charging.

[0033] Step S8: When the drone arrives at the delivery destination, the delivery receiving device set up at the window will automatically verify the drone's position. After verification, the drone will carry out the delivery work. After the drone completes the delivery task, it will send a signature information to the customer to remind the customer to sign for the package.

[0034] Step S8-1: After the drone arrives at the delivery destination, the delivery receiving device sends a signal to the drone. After receiving the signal, the drone begins to verify with the delivery receiving device. After successful verification, the delivery is carried out. Step S8-2: After the drone completes the delivery, it sends a receipt notification to the customer via the network to remind them to sign for the package.

[0035] The AI-based drone delivery system includes a delivery information processing module, a data acquisition and analysis module, a real-time monitoring module, a fault handling module, a power management module, and a delivery module. The express delivery information processing module is responsible for detecting the weight of the express delivery, binding the logistics tracking number through barcode recognition and uploading it to the cloud database, and then allocating the express delivery to the corresponding drone express warehouse according to the destination. The data acquisition and analysis module is responsible for collecting the battery power consumption curve and output power curve of the drone when delivering express packages of different weights over the same distance with a full charge. It divides these curves into intervals according to different load weights and flight states (climb, descent, level flight, hovering), and also sets the standard battery consumption rate and standard output power for each interval. The real-time monitoring module is responsible for monitoring flight status, weight, and battery level in real time, and calculating the current battery consumption rate and output power based on the flight status. The fault handling module is responsible for terminating the mission and allowing the drone to return to the station for repair when abnormal output power is detected and there are multiple consecutive abnormalities. The power management module is responsible for setting a lower limit threshold for the drone battery. When the battery reaches the threshold, it determines whether the battery is sufficient for the delivery and return mission, and then decides whether to continue delivery or return to charge. The express delivery module is responsible for automatically verifying the location of the express receiving device at the window after the drone arrives at the destination. Once the verification is successful, the delivery is carried out, and a receipt reminder is sent to the customer after the delivery is completed.

[0036] Example 1: A certain model of drone has a maximum payload of 10kg and a battery capacity of 5000mAh. The payload ranges are divided into: light payload (0kg-3kg), medium payload (3kg-6kg), and heavy payload (6kg-10kg). 50 full-charge flight tests are conducted at a fixed distance of 2.5km according to the payload range. Data is collected on the level flight phase of the medium payload (3kg-6kg) from 2 to 10 minutes. The initial battery level was 4720mAh, and the final battery level was 3520mAh. ; Simultaneously monitor the output power fluctuation between 850W and 950W, and calculate the average power through 200 time points:

[0037] Calculations show that the standard battery consumption rate for a drone flying at level speed under medium load is 150mAh / min, and the standard output power is 900W. Example 2: The lower limit of battery capacity is set to 1500mAh. When the drone is still some distance from its destination, the BMS system detects that the remaining battery capacity Qremain = 1600mAh, the current battery consumption rate vQ = 100mAh / min, and the flight speed is 10m / s. According to the calculation formula: ; ; The inertial navigation system determined the mission distance to be 8000m. Since 9600m > 8000m, the drone's battery power was sufficient for the delivery mission, so the delivery mission continued.

[0038] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. An artificial intelligence-based drone delivery method, characterized by: The method includes the following steps: Step S1: The package is weighed, the tracking number is bound by the barcode, and the data is uploaded to the cloud database. The package is then assigned to the corresponding drone delivery warehouse according to the destination. Step S2: Collect the battery consumption curve and drone output power curve of the drone under full battery power during the delivery of various packages of different weights along the same route; Step S3: Divide the collected drone battery power consumption curves and output power curves under different load weights into intervals. Divide the collected drone battery power consumption curves and output power curves into four stages according to the drone's flight state: climb, descent, level flight and hovering. Step S4: Establish a standard battery consumption rate and standard output power for multiple drone battery consumption curves and output power curves within each interval; Step S5: Install a monitor inside the drone to monitor the drone's flight status, weight, and battery level in real time, and calculate the drone's current battery consumption rate and output power based on the monitored flight status. Step S6: When the monitor detects abnormal output power and the abnormality occurs multiple times in a row, the task is terminated and the device is returned to the station for maintenance. Step S7: Set a lower limit threshold for the drone battery. When the battery reaches the threshold, determine whether the battery level is sufficient for the drone's delivery and return-to-station tasks. If it is sufficient, perform the delivery task; otherwise, return to the station to recharge. Step S8: When the drone arrives at the delivery destination, the delivery receiving device set up at the window will automatically verify the drone's position. After verification, the drone will carry out the delivery work. After the drone completes the delivery task, it will send a signature information to the customer to remind the customer to sign for the package.

2. The drone delivery method based on artificial intelligence according to claim 1, characterized in that: The specific steps of step S1 are as follows: Step S1-1: Scan the package information and upload the information to the cloud database; Step S1-2: Distribute the packages to the corresponding drone delivery warehouses according to their destinations.

3. The artificial intelligence-based drone delivery method according to claim 2, characterized in that: The specific steps of step S2 are as follows: Step S2-1: Using drones of the same model and parameters, monitor the battery consumption and output power of the drones when performing drone delivery tasks on the same road segment under different weights. Step S2-2: Divide the drone's load weight range, and plot the battery consumption curve and output power curve based on the monitored changes in drone battery power consumption and output power over time.

4. The artificial intelligence-based drone delivery method according to claim 3, characterized in that: The specific steps of step S3 are as follows: Step S3-1: Divide the battery energy consumption curve and output power curve under different loading weight conditions into loading weight ranges; Step S3-2: Use the 3σ principle to remove abrupt changes in power and energy from the collected battery consumption curve and output power curve, use digital filtering technology to eliminate high-frequency noise from the drone, and align the timestamps using the Dynamic Time Warping (DTW) algorithm. Step S3-3: Identify the UAV's flight status based on the kinematic characteristics of each flight state, specifically as follows: The hovering output power curve is characterized by maintaining a constant output power at a baseline that overcomes the aircraft's weight, with an error range of [-5%, +5%]. The amplitude fluctuation of the power curve does not exceed ±5% of the baseline constant value. The hovering battery consumption curve is characterized by the slope of the battery consumption curve. The error of the value is ≤5%; The output power curve in level flight is characterized by a standard deviation of ≤8% for the curve fluctuation amplitude within the level flight range. The battery consumption curve is characterized by... The error of the value is ≤5%; The output power curve during the climbing phase is characterized by a step-like rise and a value 20% higher than the baseline constant value, with the fluctuation range of the curve within the range of 20% higher than the baseline constant value [-5%, +5%]. At the beginning of the climb, there is a step-like rise, and after the step-like rise is completed, the power is maintained within the range of 20% higher than the baseline constant value [-5%, +5%]. The output power curve in the declining state exhibits a stepped decline and a state where it is 20% below the baseline constant value and the curve fluctuation range is within the range of -3% to +3% below the baseline constant value.

5. The artificial intelligence-based drone delivery method according to claim 4, characterized in that: The specific steps of step S4 are as follows: Step S4-1: Perform big data model analysis on the battery consumption curves and output power curves of multiple drones in each interval; Step S4-2: Based on the battery consumption curve analyzed by the big data model, determine the standard battery consumption rate for different flight states within each payload weight range of the drone: ； The average rate in each interval reflects the power consumption characteristics of the drone under specific flight conditions, and is a direct quantification of actual flight behavior, where V... avg Q represents the average battery power consumption rate under different flight conditions. initial Q represents the initial battery level when the aircraft enters flight mode. end The current battery level indicates the end of flight. Step S4-3: Based on the well-analyzed output power curve of the big data model, determine the standard output power of the UAV under different flight conditions in each load weight range; The standard output power calculation formula is: ； ； In the formula P max P represents the peak power under a specific flight condition. avg P represents average power. t P represents the power at each moment within the interval, n is the number of data points, i is the start time of a specific flight state, m is the number of continuous power data points in the interval from i to i+m-1, m=n, t This represents the power at each moment.

6. The artificial intelligence-based drone delivery method according to claim 5, characterized in that: The specific steps of step S5 are as follows: Step S5-1: Set up a monitor for the drone. The monitor will monitor the drone's flight status, weight, and battery level in real time. Step S5-2: Based on the maximum load mass of the drone, the integrated strain gauge weighing sensor module in the monitor sets a weight range with 10kg as the boundary. The integrated strain gauge weighing sensor determines the weight range based on the weight of the drone it monitors. Step S5-3: After the IMU inertial unit module in the monitor detects a change in the drone's flight status, the IMU inertial unit module sends information to the BMS system module in the monitor. After receiving the information, the BMS system monitors the drone's battery consumption rate and output power.

7. The artificial intelligence-based drone delivery method according to claim 6, characterized in that: The specific steps of step S6 are as follows: Step S6-1: The BMS system in the monitor monitors the output power of the UAV in real time. If the output power detected by the monitor exceeds the standard output power by 20% and the time exceeds 5 seconds, the output power is determined to be abnormal. Step S6-2: After the monitor detects an abnormal output power, the drone automatically returns to the station and sends a message to remind the station staff to carry out maintenance.

8. The drone delivery method based on artificial intelligence according to claim 7, characterized in that: The specific steps of step S6 are as follows: Step S7-1: Divide the drone battery power into available battery power and reserved battery power, and set a lower limit threshold for battery power; Step S7-2: When the BMS system in the monitor detects that the drone's battery level has reached the lower limit threshold, it triggers the calculation of reserved battery power. The distance is calculated based on the remaining battery power and the current battery consumption rate. The distance calculation formula is as follows: ； ； In the formula t endurance For battery life, Q remain v represents the remaining battery power. Q The rate of energy consumption is represented by s, where s represents distance traveled and v represents speed. The drone determines the mission distance using its inertial navigation system. If the calculated distance is greater than the mission distance, the delivery mission continues. If the calculated distance is less than the mission distance, the drone returns to the station for charging.

9. The artificial intelligence-based drone delivery method according to claim 8, characterized in that: The specific steps of step S7 are as follows: Step S8-1: After the drone arrives at the delivery destination, the delivery receiving device sends a signal to the drone. After receiving the signal, the drone begins to verify with the delivery receiving device. After successful verification, the delivery is carried out. Step S8-2: After the drone completes the delivery, it sends a receipt notification to the customer via the network to remind them to sign for the package.

10. An artificial intelligence-based drone delivery system, characterized in that: The AI-based drone delivery system includes a delivery information processing module, a data acquisition and analysis module, a real-time monitoring module, a fault handling module, a power management module, and a delivery module. The express delivery information processing module is responsible for detecting the weight of the express delivery, binding the logistics tracking number through barcode recognition and uploading it to the cloud database, and then allocating the express delivery to the corresponding drone express warehouse according to the destination. The data acquisition and analysis module is responsible for collecting the battery power consumption curve and output power curve of the drone when delivering express packages of different weights over the same distance with a full charge. It divides these curves into intervals according to different load weights and flight states (climb, descent, level flight, hovering), and also sets the standard battery consumption rate and standard output power for each interval. The real-time monitoring module is responsible for monitoring flight status, weight, and battery level in real time, and calculating the current battery consumption rate and output power based on the flight status. The fault handling module is responsible for terminating the mission and allowing the drone to return to the station for repair when abnormal output power is detected and there are multiple consecutive abnormalities. The power management module is responsible for setting a lower limit threshold for the drone battery. When the battery reaches the threshold, it determines whether the battery is sufficient for the delivery and return mission, and then decides whether to continue delivery or return to charge. The express delivery module is responsible for automatically verifying the location of the express receiving device at the window after the drone arrives at the destination. Once the verification is successful, the delivery is carried out, and a receipt reminder is sent to the customer after the delivery is completed.

Images