Method and system for automatic delivery of goods by unmanned aerial vehicle

Abstract

The present application discloses a method and system for automatically delivering goods by drones, and relates to the technical field of drone delivery of goods. The present application provides a method for automatically delivering goods by drones, including: obtaining a first target area, and the drone queries the target user who receives the goods in the first target area; obtaining the target delivery area for the delivery of goods according to the gesture information of the target user; judging whether the target delivery area meets the delivery conditions; if so, the drone delivers the goods to the target delivery area; if not, according to the target delivery area, projecting at least one alternative delivery area that meets the delivery conditions at a position adjacent to the target delivery area, for the target user to select and / or the target user guiding the drone to adjust the position of the alternative delivery area through gesture information.

Description

Technical Field

[0001] The present application relates to the technical field of drone cargo delivery, and in particular to a method and system for automated drone cargo delivery. Background Art

[0002] With the rapid development of science and technology, drone technology has been widely used in various fields, especially in logistics and distribution scenarios. Traditional cargo delivery methods often require manual intervention, which is not only inefficient but also difficult to implement in complex or dangerous environments.

[0003] In the existing technology, drones usually deliver goods based on preset coordinates or simple visual recognition technology. However, this method has many limitations. On the one hand, the preset coordinates may cause inaccurate delivery due to environmental factors (such as wind direction, terrain changes) or human factors (such as changes in the target user's position); on the other hand, simple visual recognition technology is difficult to cope with complex and changeable delivery environments, such as obstacles in the target area and unclear gesture information of the target user. Summary of the invention

[0004] The main purpose of this application is to provide a method and system for automated cargo delivery by drones, aiming to solve the technical problem that drones usually deliver cargo based on preset coordinates or simple visual recognition technology, which has many limitations.

[0005] To achieve the above objectives, in a first aspect, the present application provides a method for automatically delivering goods by a drone, comprising:

[0006] A first target area is obtained, and the drone searches for a target user who receives the goods in the first target area;

[0007] According to the gesture information of the target user, a target delivery area for goods delivery is obtained;

[0008] Determining whether the target delivery area meets delivery conditions;

[0009] If yes, the drone will drop the cargo to the target drop area;

[0010] If not, at least one alternative delivery area that meets the delivery conditions is projected at a position adjacent to the target delivery area according to the target delivery area for the target user to select and / or the target user guides the drone to adjust the position of the alternative delivery area through gesture information.

[0011] Optionally, the step of acquiring the first target area and searching for target users who receive the goods in the first target area further includes:

[0012] Acquire a cargo delivery task, and determine the first target area according to the destination location information of the cargo delivery task;

[0013] Acquiring navigation path information according to the starting point location information of the cargo delivery task and the first target area;

[0014] Controlling the UAV to move to the first target area according to the navigation path information;

[0015] Image information of users in the first target area is obtained, and the image information is compared with image information of target users pre-stored in a database to determine the target users.

[0016] Optionally, before acquiring the target delivery area for goods delivery according to the gesture information of the target user, the method further includes:

[0017] Obtaining a perspective transformation matrix of the drone, and transforming the perspective coordinate system of the drone into a consistent coordinate system of the three-dimensional space through the perspective transformation matrix;

[0018] Convert the three-dimensional space into a two-dimensional image through perspective projection.

[0019] Projecting the gesture information and posture information of the target user into a two-dimensional image, extracting the position information of the gesture area in the two-dimensional image, and mapping the gesture area into a three-dimensional space;

[0020] According to the gesture area in the three-dimensional space, a target placement area indicated by the gesture information is obtained.

[0021] Optionally, the step of acquiring the target placement area indicated by the gesture information according to the gesture area in the three-dimensional space further includes:

[0022] Obtain the target user's personal attributes and gesture information, train through deep learning algorithms, and obtain a classification model of the target user and gestures;

[0023] The target person's attributes are converted into digital features and stored, and all valid gesture information of the target person is classified and trained to obtain a gesture category prediction model;

[0024] Use the trained target person attribute and gesture category prediction model to predict the target user's gesture and obtain the meaning of the gesture information.

[0025] Optionally, the step of acquiring a target delivery area for goods delivery according to the gesture information of the target user further includes:

[0026] When the gesture information indicates that the drone follows the target user, the drone acquires the gesture information of the target user in real time;

[0027] The drone takes the target user as a reference target, plans the flight route of the drone, and keeps the distance between the drone and the target user within a preset range;

[0028] When the drone recognizes that the gesture information of the target user is to stop moving, the drone stops moving and maintains the current state unchanged.

[0029] Optionally, the step of determining whether the target delivery area meets delivery conditions further includes:

[0030] If the target delivery area does not meet the delivery conditions, obtaining weight information and size information of the goods to be delivered;

[0031] If the weight information and the size information are within a preset range, a virtual placement area is set in a preset interval in front of the target user according to the location information of the target user and the height information of the target user, and the distance of the virtual placement area relative to the ground is one third to two thirds of the height of the target user.

[0032] Optionally, if not, the step of projecting at least one candidate delivery area that meets the delivery condition at a position adjacent to the target delivery area according to the target delivery area includes:

[0033] According to the size information of the goods to be delivered, the projection area of ​​the alternative delivery area is obtained.

[0034] Optionally, if not, the step of projecting at least one candidate delivery area that meets the delivery condition at a position adjacent to the target delivery area according to the target delivery area further includes:

[0035] Obtaining an alternative distance of each candidate delivery area relative to the target user, and evaluating the flatness of the candidate delivery area;

[0036] According to the candidate distance and the flatness, evaluating the priority of each candidate delivery area;

[0037] According to the priority, all the candidate delivery areas are sorted and numbered from high to low priority.

[0038] Optionally, the step of providing the target user with a selection and / or the target user guiding the drone to adjust the position of the candidate delivery area through gesture information includes:

[0039] According to the ranking number, the target user selects one of the candidate delivery areas as the final delivery area through voice and / or a first gesture;

[0040] Or the target user selects one of the candidate delivery areas through voice and / or a first gesture, and then guides the drone to adjust the position of the selected candidate delivery area through a second gesture, with the final position of the candidate delivery area being the final delivery area.

[0041] In a second aspect, the present application provides a drone automated cargo delivery system, comprising:

[0042] A first target area acquisition module, which is configured to acquire a first target area, and the drone searches for a target user who receives the goods in the first target area;

[0043] A target delivery area acquisition module, which is configured to acquire a target delivery area for goods delivery according to the gesture information of the target user;

[0044] A condition judgment module, configured to judge whether the target delivery area meets the delivery condition;

[0045] A delivery module, which is configured to cause the drone to deliver the cargo to the target delivery area if the target delivery area meets the delivery conditions;

[0046] The alternative delivery area acquisition module is configured to project at least one alternative delivery area that meets the delivery conditions at a position adjacent to the target delivery area based on the target delivery area if the target delivery area does not meet the delivery conditions, so that the target user can select it and / or the target user can guide the drone to adjust the position of the alternative delivery area through gesture information.

[0047] Beneficial effects that this application can achieve:

[0048] The embodiment of the present application proposes a method and system for automated delivery of goods by drones, which determines the target delivery area for goods delivery by obtaining the gesture information of the target user, which helps to ensure that the goods can be accurately delivered to the user's desired location and reduce the inconvenience or loss caused by inaccurate delivery location. When it is determined that the target delivery area does not meet the delivery conditions, the drone can project an alternative delivery area at an adjacent location. This function increases the flexibility of delivery, allowing the drone to find a suitable delivery location in different environments and conditions. The entire delivery process is automated, from obtaining the target area, querying the target user, determining the delivery area, judging the delivery conditions, and finally delivering the goods, which greatly improves the delivery efficiency and automation. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is a flow chart of the method for automatically delivering goods by drones of the present application;

[0050] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0053] In the present invention, unless otherwise clearly specified and limited, the terms "connection", "fixation", etc. should be understood in a broad sense. For example, "fixation" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0054] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the meaning of "and / or" appearing in the full text includes three parallel schemes. Taking "A and / or B" as an example, it includes scheme A, or scheme B, or a scheme that satisfies both A and B. In addition, the technical solutions between the various embodiments can be combined with each other, but it must be based on the ability of ordinary technicians in the field to implement. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0055] Example 1

[0056] Reference Figure 1 The first embodiment of the present application provides a method for automatically delivering goods by a drone, including the following steps:

[0057] S10: Acquire a first target area, and the drone searches for a target user who is to receive the goods in the first target area.

[0058] Specifically, the scope and boundaries of the first target area are clearly defined by GPS coordinates, map data, or preset geographic fences. Ensure that the drone has the ability to access and parse these area definition information. For example, according to the user's positioning information, set the first target area to a radius of 100 meters around the positioning location. The drone flies above or near the first target area according to the preset flight plan or real-time instructions, and is ready to start querying the target user.

[0059] When querying the target user, query through the user identification system on the drone, which may include cameras, infrared sensors, wireless communication devices, etc. Ensure that all sensors and communication devices related to user query are in normal working condition. Use the camera for visual scanning, combined with image processing algorithms, to identify people in the first target area. You can try to establish a connection with potential target users through wireless communication devices (such as Bluetooth, Wi-Fi, etc.) to obtain more accurate user information. If possible, use infrared sensors to detect the heat signature of people to assist in identification when visual conditions are poor. Verify the identified people to ensure that they are the intended target users to receive the goods. This can be achieved by comparing preset user information (such as facial features, identity tags, etc.). If necessary, the user can be asked to perform additional confirmation operations through the display screen on the drone, voice prompts or mobile applications.

[0060] S20: Acquire a target delivery area for goods delivery according to the gesture information of the target user.

[0061] Specifically, the camera or other sensors on the drone should continuously monitor the target user to capture their gesture movements. Ensure that the sensor has a high enough resolution and sensitivity to accurately identify the user's gestures. Use machine learning algorithms or preset gesture recognition models to parse and recognize the captured gesture information. The diversity, speed, direction and other factors of gestures should be considered during the recognition process to improve the accuracy of recognition. Based on the recognized gesture information, determine the target delivery area for cargo delivery. This usually involves matching the gesture with the preset delivery area mapping relationship. If the gesture information is unclear or ambiguous, the drone should be able to request further confirmation or additional instructions from the user.

[0062] S30, determining whether the target delivery area meets delivery conditions;

[0063] Specifically, use the sensors on the drone (such as cameras, lidar, etc.) to scan the target delivery area in real time to obtain detailed environmental information. Analyze environmental information, including terrain, obstacles, crowd density, weather conditions, etc., to evaluate the feasibility of delivery. Ensure that the target delivery area maintains a sufficient safety distance from the drone, users and other obstacles to avoid collision or injury. Calculate and set a reasonable safety distance threshold based on the drone's flight altitude, speed, and the weight and volume of the cargo. Analyze the drone's positioning accuracy, flight stability, and cargo delivery mechanism to ensure that the cargo can be accurately delivered to the target delivery area. If necessary, a simulated delivery test can be performed to verify delivery accuracy and stability.

[0064] S40: If the target delivery area meets the delivery conditions, the drone delivers the cargo to the target delivery area.

[0065] Specifically, after determining that the target delivery area meets the delivery conditions, the drone should receive or generate clear delivery instructions. The delivery instruction should contain specific information about the cargo (such as type, weight, volume, etc.) and the precise location and time of delivery. The drone adjusts its flight attitude and position according to the delivery instruction to ensure that the cargo can be accurately delivered to the target delivery area. This may involve hovering, diving, climbing and other actions of the drone, as well as precise position control. Activate the cargo delivery mechanism of the drone, such as opening the delivery hatch, adjusting the delivery angle, etc. Ensure that the delivery mechanism is in good working condition and can release the cargo accurately and stably. After confirming that everything is ready, the drone performs the delivery operation and delivers the cargo to the target delivery area. The status of the cargo should be continuously monitored during the delivery process to ensure that the cargo reaches the target location safely and accurately. If an abnormal situation is encountered during the delivery process (such as cargo jamming, delivery mechanism failure, etc.), the drone should be able to respond in time and take appropriate treatment measures. This may include stopping delivery, returning to the origin, requesting manual intervention, etc.

[0066] S50: If the target delivery area does not meet the delivery conditions, at least one alternative delivery area that meets the delivery conditions is projected at a position adjacent to the target delivery area according to the target delivery area, for the target user to select and / or the target user guides the drone to adjust the position of the alternative delivery area through gesture information.

[0067] Specifically, the projection equipment or display screen on the drone is used to display the alternative delivery areas to the target users in an intuitive manner.

[0068] The projection should be clear and accurate, easy for users to understand and choose. Provide users with the option of selecting alternative delivery areas, which they can select through mobile applications, gesture information or other means. If the user wants to adjust the location of the alternative delivery area, the drone should be able to make corresponding adjustments based on the user's gesture information or instructions. After the user selects or adjusts the alternative delivery area, the drone should confirm again whether the area meets the delivery conditions. If the conditions are met, the drone will execute the delivery of the goods to the selected alternative delivery area. This ensures that when the target delivery area does not meet the delivery conditions, the drone can quickly and accurately search and provide alternative delivery areas, while fully considering the needs and preferences of users to ensure that the goods can be safely and accurately delivered to a suitable location. This helps to improve the flexibility of drone delivery of goods and user satisfaction.

[0069] Example 2

[0070] Based on Example 1, this example provides a method for automatically delivering goods by a drone, including the following steps:

[0071] S10: Acquire a first target area, and the drone searches for a target user who is to receive the goods in the first target area.

[0072] Optionally, the step of acquiring the first target area and the drone searching for target users who receive the goods within the first target area further includes:

[0073] S11, obtaining a cargo delivery task, and determining the first target area according to the destination location information of the cargo delivery task;

[0074] Specifically, relevant information of the cargo delivery task is received, including the starting point, end point, weight, volume, etc. of the cargo. According to the end point location information (such as GPS coordinates, address, etc.), the scope and boundary of the first target area are determined on the map.

[0075] S12, acquiring navigation path information according to the starting position information of the cargo delivery task and the first target area;

[0076] Specifically, the best navigation path is planned based on the starting point location information and the first target area using the navigation system of the drone or an external map service. The navigation path should take into account factors such as the drone's flight altitude, speed, avoiding obstacles and no-fly zones.

[0077] S13, controlling the drone to move to the first target area according to the navigation path information;

[0078] Specifically, the drone automatically flies to the first target area according to the navigation path information. During the flight, the drone should continuously monitor the flight status and environment to ensure safe flight.

[0079] S14: Obtain image information of users in the first target area, compare the image information with image information of target users pre-stored in a database, and determine the target users.

[0080] Specifically, after the drone arrives at the first target area, it uses a camera or other image acquisition equipment to obtain image information of users in the area. The collected image information is compared with the image information of the target user pre-stored in the database, and the target user is identified through the image processing algorithm. If multiple potential target users are identified, the drone can further confirm them through other means (such as wireless communication, voice prompts, etc.).

[0081] S20: Acquire a target delivery area for goods delivery according to the gesture information of the target user.

[0082] Optionally, before acquiring the target delivery area for goods delivery according to the gesture information of the target user, the method further includes:

[0083] S21, obtaining a perspective transformation matrix of the drone, and transforming the perspective coordinate system of the drone into a consistent coordinate system of the three-dimensional space through the perspective transformation matrix;

[0084] Specifically, when the drone is flying and observing the target user, its view coordinate system may be inconsistent with the standard three-dimensional space coordinate system. The drone's view transformation matrix is ​​a mathematical tool used to transform the drone's view coordinate system (usually based on the drone's camera view) to a standard three-dimensional space coordinate system (such as the world coordinate system). By obtaining the view transformation matrix, the drone's view coordinate system can be transformed to be consistent with the three-dimensional space coordinate system, thereby more accurately understanding the target user's gestures and postures.

[0085] S22, converting the three-dimensional space into a two-dimensional image by perspective projection;

[0086] Specifically, in order to process and analyze the gesture information of the target user on the drone, the scene in the three-dimensional space needs to be converted into a two-dimensional image. Using the principle of perspective projection, points in the three-dimensional space can be mapped onto a two-dimensional plane to form a two-dimensional image.

[0087] S23, projecting the gesture information and posture information of the target user into a two-dimensional image, extracting the position information of the gesture area in the two-dimensional image, and mapping the gesture area into a three-dimensional space;

[0088] Specifically, in the two-dimensional image, the gesture area of ​​the target user is identified and extracted, including information such as the shape, position and direction of the gesture. Using the principle of inverse perspective projection, the gesture area in the two-dimensional image is mapped back to the three-dimensional space to obtain the actual position and posture of the gesture in the three-dimensional space.

[0089] S24. Acquire a target placement area indicated by the gesture information according to the gesture area in the three-dimensional space.

[0090] Specifically, in three-dimensional space, the position, shape, and posture of the gesture area are analyzed to determine the target placement area for the cargo indicated by the target user. By calculating the relative relationship between the gesture area and the current position of the drone, the specific position and range of the target placement area can be determined.

[0091] Optionally, the step of acquiring the target placement area indicated by the gesture information according to the gesture area in the three-dimensional space further includes:

[0092] S241, obtaining the character attributes and gesture information of the target user, and training through a deep learning algorithm to obtain a classification model of the target user and gesture;

[0093] Specifically, first, collect the target user's personal attributes (such as age, gender, height, body shape, etc.) and the corresponding gesture information (such as gesture shape, motion trajectory, etc.). Then, use deep learning algorithms (such as convolutional neural network CNN, recurrent neural network RNN, etc.) to train these data to learn feature representations that can distinguish different users and gestures.

[0094] S242, converting the target person's attributes into digital features and storing them, performing classification training on all valid gesture information of the target person to obtain a gesture category prediction model;

[0095] Specifically, the target user's personal attributes are converted into digital features (such as vectors, matrices, etc.) and stored in the database. Then, these digital features and the corresponding gesture information are classified and trained, and a gesture category prediction model is constructed using machine learning algorithms (such as support vector machines (SVMs), decision trees (DTs, etc.).

[0096] S243: Use the trained target person attribute and gesture category prediction model to predict the target user's gesture and obtain the meaning of the gesture information.

[0097] Specifically, when the drone captures the target user's gesture information, it is input into the trained prediction model. The model will output the meaning of the gesture information (such as indicating the location and direction of the goods) based on the input gesture information and the target user's personal attributes.

[0098] Optionally, the step of acquiring a target delivery area for goods delivery according to the gesture information of the target user further includes:

[0099] S201, when the gesture information indicates that the drone follows the target user, the drone acquires the gesture information of the target user in real time;

[0100] Specifically, the drone uses a camera or other sensors to capture the gesture information of the target user in real time. The gesture information here may include the shape, motion trajectory, speed, etc. of the gesture.

[0101] S202, the drone plans a flight route of the drone with the target user as a reference target, and keeps the distance between the drone and the target user within a preset range;

[0102] Specifically, the drone plans a flight route to follow the target user based on the real-time location information and gesture information of the target user using a path planning algorithm (such as the Dijkstra algorithm). At the same time, the drone adjusts the flight speed and direction to ensure that the distance between the drone and the target user remains within a preset safety range.

[0103] S203: When the drone recognizes that the gesture information of the target user is to stop moving, the drone stops moving and maintains the current state unchanged.

[0104] Specifically, the drone recognizes the gesture information of the target user through image processing algorithms or machine learning models. When a gesture of stopping movement is recognized (such as a gesture shape of a specific pattern, a gesture speed slowing down, etc.), the drone stops moving immediately and maintains the current altitude, speed and direction unchanged. After the drone stops moving, it can perform other operations as needed, such as waiting for further instructions from the target user, delivering goods, etc. If the target user continues to move or changes the gesture information, the drone will re-plan the flight route and follow the target user.

[0105] S30: Determine whether the target delivery area meets delivery conditions.

[0106] Specifically, the step of determining whether the target delivery area meets the delivery conditions further includes:

[0107] S31. If the target delivery area does not meet the delivery conditions, obtain the weight information and size information of the goods to be delivered;

[0108] Specifically, the drone obtains the weight and size information of the goods to be delivered through internal sensors or external input devices (such as barcode scanners, RFID readers, etc.). This information is crucial to determine whether the goods are suitable for delivery at the current location. By obtaining the weight and size information of the goods, the suitability of the target delivery area can be further evaluated, and a basis can be provided for the subsequent setting of the virtual placement area.

[0109] S32. If the weight information and the size information are within a preset range, a virtual placement area is set in a preset interval in front of the target user according to the location information of the target user and the height information of the target user, and the distance of the virtual placement area relative to the ground is one third to two thirds of the height of the target user.

[0110] Specifically, the drone obtains the location information and height information of the target user in real time through the onboard camera or other sensors. Then, based on this information and the preset algorithm, a virtual placement area is set within a certain distance (such as 1-2 meters) in front of the target user. The height of this area is set to one-third to two-thirds of the height of the target user to ensure that the goods will not be too high or too low when they are delivered, so that the target user can use them conveniently. By combining the location and height information of the target user and the characteristics of the goods, the present invention can intelligently set a suitable virtual placement area, which not only ensures the safe delivery of the goods, but also improves the convenience of users to use them.

[0111] The setting of the virtual placement area can take into account a variety of factors, such as the fragility of the goods, the mobility of the target user, etc. If the goods are fragile or the target user has difficulty in moving, the virtual placement area can be set lower to reduce the risk of damage to the goods and facilitate user access.

[0112] After setting the virtual placement area, the drone can perform corresponding delivery operations, such as adjusting the flight altitude, speed, direction, etc., to ensure that the goods are accurately and safely delivered to the virtual placement area. At the same time, the drone can also inform the target user through voice prompts, display screens, etc. that the goods are about to be delivered, and please pay attention to receive them.

[0113] S40: If the target delivery area meets the delivery conditions, the drone delivers the cargo to the target delivery area.

[0114] S50: If the target delivery area does not meet the delivery conditions, at least one alternative delivery area that meets the delivery conditions is projected at a position adjacent to the target delivery area according to the target delivery area, for the target user to select and / or the target user guides the drone to adjust the position of the alternative delivery area through gesture information.

[0115] Optionally, if the target delivery area does not meet the delivery condition, the step of projecting at least one candidate delivery area that meets the delivery condition at a position adjacent to the target delivery area according to the target delivery area includes:

[0116] S51. Obtain the projection area of ​​the candidate delivery area according to the size information of the goods to be delivered.

[0117] Specifically, the drone first obtains the length, width, height and other dimensions of the goods to be delivered through internal sensors or external input devices. Then, based on these dimensions and the placement of the goods (such as flat, upright, etc.), the projected area of ​​the goods on the ground is calculated. This projected area will serve as an important basis for determining the size of the alternative delivery area. By obtaining the projected area of ​​the goods, it can be ensured that the alternative delivery area is large enough to accommodate the goods to be delivered, avoiding the goods from exceeding the area during the delivery process, causing safety hazards or delivery failure.

[0118] Optionally, if the target delivery area does not meet the delivery condition, the step of projecting at least one candidate delivery area that meets the delivery condition at a position adjacent to the target delivery area according to the target delivery area further includes:

[0119] S52, obtaining an alternative distance of each candidate delivery area relative to the target user, and evaluating the flatness of the candidate delivery area;

[0120] Specifically, the drone uses the sensors it carries to obtain the relative distance between the candidate delivery area and the target user in real time. At the same time, the drone uses a camera or other sensors to scan the ground of the candidate delivery area to evaluate its flatness, such as whether there is a slope or unevenness. The candidate distance and flatness are important factors affecting the accuracy and safety of cargo delivery. By obtaining this information, it can provide a basis for the subsequent evaluation of the priority of the candidate delivery area.

[0121] S53, evaluating the priority of each candidate delivery area according to the candidate distance and the flatness;

[0122] Specifically, each candidate delivery area is scored or ranked according to the preset evaluation criteria (e.g., the closer the candidate distance and the higher the flatness, the higher the priority). Different weights can be assigned to the candidate distance and flatness to comprehensively evaluate the priority of the candidate delivery area. By evaluating the priority, the most suitable candidate delivery area can be selected for the drone to ensure that the cargo can be delivered safely and accurately.

[0123] S54: According to the priority, all the candidate delivery areas are sorted and numbered from high to low priority.

[0124] Specifically, according to the evaluation results, all candidate delivery areas are sorted and assigned a unique number. This number can be used in subsequent control instructions to instruct the drone to select which candidate delivery area to deliver. The sorting number can facilitate the drone or the target user to make a choice among multiple candidate delivery areas and perform delivery operations in order of priority.

[0125] Optionally, the step of providing the target user with a selection and / or the target user guiding the drone to adjust the position of the candidate delivery area through gesture information includes:

[0126] S501: According to the sorting number, the target user selects one of the candidate delivery areas as the final delivery area through voice and / or a first gesture;

[0127] Specifically, the drone displays the ranking numbers of the candidate delivery areas to the target user through display screens, voice prompts, etc. The target user can select one of the candidate delivery areas as the final delivery area according to his or her needs and preferences through voice commands (such as "select area number 1") or first gestures (such as a gesture pointing to a certain number). By allowing the target user to select the candidate delivery area, it can be ensured that the location of the cargo delivery meets the user's expectations and needs, thereby improving user satisfaction.

[0128] S502, or the target user selects one of the candidate delivery areas through voice and / or a first gesture, and then guides the drone to adjust the position of the selected candidate delivery area through a second gesture, with the final position of the candidate delivery area being the final delivery area.

[0129] Specifically, the target user first selects an alternative delivery area through a voice command or a first gesture. Then, if the user is not completely satisfied with the location of the selected alternative delivery area, the second gesture (such as a gesture indicating movement up, down, left, right, etc.) can be used to guide the drone to fine-tune the location of the alternative delivery area. The drone adjusts the location of the alternative delivery area in real time according to the user's gesture instructions until the user is satisfied. Finally, the adjusted location of the alternative delivery area is used as the final delivery area. By allowing the target user to fine-tune the location of the alternative delivery area, the accuracy of cargo delivery and user satisfaction can be further improved. This function is particularly important in some scenarios that require precise delivery (such as delivering cargo into a narrow space).

[0130] After the final delivery area is determined, the drone can perform corresponding delivery operations, such as adjusting the flight altitude, speed, direction, etc., to ensure that the goods are accurately and safely delivered to the final delivery area. At the same time, the drone can also inform the target user through voice prompts, display screens, etc. that the goods are about to be delivered, and please pay attention to receive them.

[0131] Example 3

[0132] This embodiment provides a drone automated cargo delivery system, including:

[0133] A first target area acquisition module, which is configured to acquire a first target area, and the drone searches for a target user who receives the goods in the first target area;

[0134] A target delivery area acquisition module, which is configured to acquire a target delivery area for goods delivery according to the gesture information of the target user;

[0135] A condition judgment module, configured to judge whether the target delivery area meets the delivery condition;

[0136] A delivery module, which is configured to cause the drone to deliver the cargo to the target delivery area if the target delivery area meets the delivery conditions;

[0137] The alternative delivery area acquisition module is configured to project at least one alternative delivery area that meets the delivery conditions at a position adjacent to the target delivery area based on the target delivery area if the target delivery area does not meet the delivery conditions, so that the target user can select it and / or the target user can guide the drone to adjust the position of the alternative delivery area through gesture information.

[0138] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for automatically delivering goods by drone, characterized in that: include: A first target area is obtained, and the drone searches for a target user who receives the goods in the first target area; According to the gesture information of the target user, a target delivery area for goods delivery is obtained; Determining whether the target delivery area meets delivery conditions; If yes, the drone will drop the cargo to the target drop area; If not, then, based on the target delivery area, at least one candidate delivery area that meets the delivery conditions is projected at a position adjacent to the target delivery area, for the target user to select and / or the target user to guide the drone to adjust the position of the candidate delivery area through gesture information; The step of determining whether the target delivery area meets the delivery conditions includes: If the target delivery area does not meet the delivery conditions, obtaining weight information and size information of the goods to be delivered; If the weight information and the size information are within a preset range, a virtual placement area is set in a preset interval in front of the target user according to the location information of the target user and the height information of the target user, and the distance of the virtual placement area relative to the ground is one third to two thirds of the height of the target user.

2. The method for automatically delivering goods by drone according to claim 1, characterized in that: The step of obtaining the first target area and the drone searching for target users who receive the goods in the first target area also includes: Acquire a cargo delivery task, and determine the first target area according to the destination location information of the cargo delivery task; Acquiring navigation path information according to the starting point location information of the cargo delivery task and the first target area; Controlling the UAV to move to the first target area according to the navigation path information; Image information of users in the first target area is obtained, and the image information is compared with image information of target users pre-stored in a database to determine the target users.

3. The method for automatically delivering goods by drone according to claim 1, characterized in that: Before acquiring the target delivery area for goods delivery according to the gesture information of the target user, the method further includes: Obtaining a perspective transformation matrix of the drone, and transforming the perspective coordinate system of the drone into a consistent coordinate system of the three-dimensional space through the perspective transformation matrix; Convert the three-dimensional space into a two-dimensional image through perspective projection. Projecting the gesture information and posture information of the target user into a two-dimensional image, extracting the position information of the gesture area in the two-dimensional image, and mapping the gesture area into a three-dimensional space; According to the gesture area in the three-dimensional space, a target placement area indicated by the gesture information is obtained.

4. The method for automatically delivering goods by drone as claimed in claim 3, characterized in that: The step of acquiring the target placement area indicated by the gesture information according to the gesture area in the three-dimensional space also includes: Obtain the target user's personal attributes and gesture information, train through deep learning algorithms, and obtain a classification model of the target user and gestures; The target person's attributes are converted into digital features and stored, and all valid gesture information of the target person is classified and trained to obtain a gesture category prediction model; The target user's gestures are predicted through the target person's attributes and the trained gesture category prediction model to obtain the meaning of the gesture information.

5. The method for automatically delivering goods by drone according to claim 1, characterized in that: The step of acquiring a target delivery area for goods delivery according to the gesture information of the target user also includes: When the gesture information indicates that the drone follows the target user, the drone acquires the gesture information of the target user in real time; The drone takes the target user as a reference target, plans the flight route of the drone, and keeps the distance between the drone and the target user within a preset range; When the drone recognizes that the gesture information of the target user is to stop moving, the drone stops moving and maintains the current state unchanged.

6. The method for automatically delivering goods by drone according to claim 1, characterized in that: If not, then according to the target delivery area, the step of projecting at least one candidate delivery area that meets the delivery condition at a position adjacent to the target delivery area comprises: According to the size information of the goods to be delivered, the projection area of ​​the alternative delivery area is obtained.

7. The method for automatically delivering goods by drone as claimed in claim 1, characterized in that: If not, then according to the target delivery area, projecting at least one candidate delivery area that meets the delivery conditions at a position adjacent to the target delivery area, further comprises: Obtaining an alternative distance of each candidate delivery area relative to the target user, and evaluating the flatness of the candidate delivery area; According to the candidate distance and the flatness, evaluating the priority of each candidate delivery area; According to the priority, all the candidate delivery areas are sorted and numbered from high to low priority.

8. The method for automatically delivering goods by drone as claimed in claim 7, characterized in that: The step of providing the target user with a selection and / or the target user guiding the drone to adjust the position of the candidate delivery area through gesture information includes: According to the ranking number, the target user selects one of the candidate delivery areas as the final delivery area through voice and / or a first gesture; Or the target user selects one of the candidate delivery areas through voice and / or a first gesture, and then guides the drone to adjust the position of the selected candidate delivery area through a second gesture, with the final position of the candidate delivery area being the final delivery area.

9. An automatic cargo delivery system of a drone applied to the method described in any one of claims 1 to 8, characterized in that: include: A first target area acquisition module, which is configured to acquire a first target area, and the drone searches for a target user who receives the goods in the first target area; A target delivery area acquisition module, which is configured to acquire a target delivery area for goods delivery according to the gesture information of the target user; A condition judgment module, configured to judge whether the target delivery area meets the delivery condition; A delivery module, which is configured to cause the drone to deliver the cargo to the target delivery area if the target delivery area meets the delivery conditions; The alternative delivery area acquisition module is configured to project at least one alternative delivery area that meets the delivery conditions at a position adjacent to the target delivery area based on the target delivery area if the target delivery area does not meet the delivery conditions, so that the target user can select it and / or the target user can guide the drone to adjust the position of the alternative delivery area through gesture information.

Images