A mountain harvesting system and method based on drone swarm and monorail carrier

Abstract

The present invention discloses a mountain picking system and method based on a drone group and a monorail carrier, the method comprising: S101 scanning the image information and position information of fruit trees in an orchard; S102 extracting fruit information according to the image information and position information of the fruit trees in the orchard; S103 planning the picking area, picking path and picking sequence according to the fruit information; S104 the picking drone picks the fruit according to the planned picking area, picking path and picking sequence; S105 extracting the remaining space of the loading basket according to the image information of the loading basket; S106 calculating the remaining loading quantity of the loading basket; S107 the picking drone delivers to the loading basket according to the remaining loading quantity; S108 after all the loading baskets are fully loaded, the transport vehicle drives to the designated location to unload, and the picking drone delivers to the loading basket in the next transport vehicle. The picking drone classifies and picks the fruits according to the individual size, avoiding classification at a later stage, and shortening the process of processing the fruits at a later stage.

Description

Technical Field

[0001] The present invention belongs to the technical field of unmanned picking, and in particular relates to a mountain picking system based on a drone swarm and a monorail carrier vehicle. Background Art

[0002] China is one of the world's important fruit producing areas, and its annual fruit production ranks among the top in the world. At present, the planting and management of most fruits still rely on manual processing, especially fruit picking requires a lot of labor, and the level of mechanization and automation is low. For example, persimmons usually grow in high and scattered positions, and the peel is easily damaged during the picking process. It is a typical difficult-to-pick fruit. At present, most of its picking forms are manual picking, which is usually divided into two methods. The first method is that workers stand on the ground and hold simple tools made of long bamboo poles to pick, and judge the position of persimmons from bottom to top with the naked eye, but it is inefficient and requires high arm strength and vision of workers; the second method is that workers climb up the persimmon tree to pick, but this picking method is risky, and workers are inconvenient to move in the space where branches are intertwined, which is easy to cause safety accidents. However, pure manual handling is not only labor-intensive and inefficient, but also costly. In particular, fruit trees are planted in uneven mountains, and the labor intensity of pure manual handling is much greater than that of flat land or wide orchards. To this end, the prior art has proposed a monorail transport device that can be used in mountainous areas, etc., as well as various picking systems that utilize drones and transport vehicles to work in coordination.

[0003] For example, the utility model with application number 202321203265.2 discloses a fruit tree picking track device, including a loader, a track, a limit clamp connected and fixed at the lower part of the loader, the limit clamp is rotatably connected to the lower part with two pulleys to form a transmission device, the transmission device is sleeved on the outer periphery of the track, sleeves are provided at the four ends of the lower part of the loader, a buffer is connected inside the sleeve, a support rod is fixedly connected to the lower end of the buffer, a universal joint is connected to the lower end of the support rod, a roller is connected to the other end of the universal joint, and slide grooves are opened on both sides of the upper part of the track, and the rollers are arranged in the slide grooves for rolling.

[0004] In addition, the prior art CN202110271356.9 discloses a task allocation and path planning method for multiple robots for picking and transporting in an orchard. The main steps of this scheme are as follows: 1. Ensure that multiple robots can harvest smoothly under a fault-free state, continuously update the harvesting area, and divide the nature of the robot's working tasks according to the number of robots traveling in the harvesting area; 2. When there is a single robot traveling in the updated harvesting area, perform tightly coupled task allocation, and mark the robot status for planning the shortest path to the target point; 3. When there are more than two robots traveling in the updated harvesting area, perform loosely coupled task allocation, establish a coordination order for multiple robots for picking and transporting, and mark the robot status for planning the shortest path to the target point; 4. Calculate the accurate target point for it based on the robot status and whether the robot's target point is clear, and establish the shortest path from the robot's current position to the target point.

[0005] In the above-mentioned prior art, the fruits are collected in a unified manner, so after collection, they need to be sorted through a sorting process to grade the fruits (the grading standard is mainly the individual size of the fruits). Summary of the invention

[0006] The purpose of the present invention is to provide a mountain picking system based on a drone swarm and a monorail carrier vehicle, which partially solves or alleviates the above-mentioned deficiencies in the prior art and can utilize drones and carrier vehicles to work together to achieve automatic grading and picking of fruits.

[0007] In order to solve the above-mentioned technical problems, the present invention specifically adopts the following technical solutions: a mountain picking method based on a drone swarm and a monorail carrier, comprising:

[0008] S101 scans the image and location information of fruit trees in the orchard;

[0009] S102 extracts fruit information based on the image information and position information of the fruit trees in the garden, wherein the fruit information includes the number, position, individual size, and maturity of the fruit;

[0010] S103 plans the picking area, picking path and picking order according to the fruit information;

[0011] The S104 harvesting drone harvests fruits according to the planned harvesting area, harvesting path and harvesting sequence; the harvested fruits are placed into the corresponding loading baskets on the carrier according to the individual classification of the fruits, and after successful placement, a signal is fed back to the upper computer for counting by the upper computer;

[0012] After counting that the number of fruits loaded in a certain loading basket reaches a percentage threshold, S105 retrieves the image information of the loading basket when the last harvesting drone dropped the fruit into the loading basket; and extracts the remaining space of the loading basket according to the image information of the loading basket;

[0013] S106 calculates the remaining loading quantity of the loading basket according to the average individual size of the fruits loaded in the loading basket and the remaining space of the loading basket;

[0014] The S107 picking drone delivers the harvested food to the loading basket according to the remaining loading quantity;

[0015] S108 After all loading baskets are fully loaded, the transport vehicle drives to the designated location to unload, and the picking drone delivers the goods to the loading basket in the next transport vehicle.

[0016] As an improvement, the harvesting drones are organized into teams according to the individual classification of the fruits, and each team of drones picks the fruits of the corresponding individual classification;

[0017] The number of drones in each team is inversely proportional to the size of the corresponding fruit individual classification.

[0018] As an improvement, after counting the number of fruits loaded in the first loading basket of the transport vehicle reaches a percentage threshold, S201 obtains the remaining space of all loading baskets;

[0019] S202 calculates the remaining loading quantity of the loading basket according to the average individual size of the fruits loaded in the loading basket and the remaining space of the loading basket;

[0020] S203: Re-forming the loading drones into teams according to the remaining load quantity of each loading basket, and the number of drones in each team is proportional to the remaining load quantity of the corresponding loading basket.

[0021] As an improvement, the transport vehicle sets a parking position according to the planned path of the harvesting drone, including:

[0022] S301 obtains the positions of the first N fruits that the picking drone needs to pick according to the picking order;

[0023] S302 traverses all parking points on the driving path of the transport vehicle, and calculates the total distance between each parking point and the positions of the N fruits;

[0024] S303 selects the parking point with the shortest total distance as the current parking point of the transport vehicle;

[0025] S304 obtains the positions of the next N fruits and executes steps S302 and S303;

[0026] S305 repeats step S304 until picking is completed.

[0027] As an improvement, the harvesting drone obtains image information of the current fruit to be picked while picking according to the planned path;

[0028] The individual size and maturity of the fruit are extracted based on the image information and checked with the fruit information. If they meet the requirements, the fruit will be picked; if the maturity meets the requirements but the individual size does not, a temporary task is inserted to pick the fruit and put it into the corresponding loading basket; if the maturity does not meet the requirements, the picking of the fruit will be abandoned.

[0029] As an improvement, the maturity of the fruit is judged based on the color and / or individual size of the fruit.

[0030] The present invention also provides a mountain harvesting system based on a drone swarm and a monorail carrier, comprising:

[0031] Monitoring drones are used to scan image information and location information of fruit trees in orchards;

[0032] The host computer is used to extract fruit information based on the image information and location information of the fruit trees in the garden obtained by the monitoring drone, wherein the fruit information includes the number, location, individual size and maturity of the fruit; and plan the picking area and picking path based on the fruit information;

[0033] The harvesting drone is used to harvest fruits according to the harvesting area and harvesting path planned by the host computer;

[0034] A transport vehicle is used to load the fruits picked by the harvesting drone and transport the fruits to a designated location; the transport vehicle is provided with a plurality of loading baskets for loading different individual categories;

[0035] The harvesting drone will place the harvested fruits into corresponding loading baskets according to the individual classification of the fruits, and after successful placement, it will feedback a signal to the upper computer to facilitate counting by the upper computer;

[0036] After the upper computer counts that the number of fruits loaded in a certain loading basket reaches the percentage threshold, it retrieves the image information of the loading basket when the last harvesting drone dropped into the loading basket; and extracts the remaining space of the loading basket based on the image information of the loading basket;

[0037] Calculate the remaining loading quantity of the loading basket according to the average individual size of the fruits loaded in the loading basket and the remaining space of the loading basket;

[0038] The picking drone delivers the harvest to the loading basket according to the remaining loading quantity;

[0039] After all loading baskets are fully loaded, the transport vehicle drives to the designated location to unload, and the picking drone delivers the goods to the loading basket in the next transport vehicle.

[0040] The number of the transport vehicles is two or more, and the transport vehicles are both rail- and land-based.

[0041] As an improvement, the individual fruits are classified into three categories: large, medium and small, and the types of loading baskets used to load the fruits correspond to the classification of the individual fruits.

[0042] As an improvement, the monitoring drone and the harvesting drone are both equipped with a 3D point cloud camera and a 2D color camera.

[0043] The invention is beneficial in that: the invention completes the orchard picking task through the collaborative work between the drone, the transport vehicle and the host computer. First, the monitoring drone is used to scan the image information and position information of the fruit trees in the orchard. Then, the fruit information is extracted based on the image information and position information of the fruit trees in the orchard, and the fruit information includes the number, location, individual size and maturity of the fruit. Then, the picking area, picking path and picking order are planned based on the fruit information. The picking drone picks the fruit according to the planned picking area, picking path and picking order.

[0044] In the present invention, the harvesting drone classifies and harvests the fruits according to their individual sizes, thereby avoiding classification at a later stage and shortening the process of processing the fruits at a later stage.

[0045] In addition, since it is a monorail transport vehicle, only one transport vehicle can run on the monorail at the same time. Therefore, in order to reduce the flight distance of the drone, not only the picking path of the drone is planned, but the transport vehicle also sets a parking position according to the planned path of the picking drone. The parking position of the transport vehicle is used to accommodate the drone and is adjusted according to the picking situation. In this way, the flight distance of the drone is greatly shortened during the entire picking process, thereby improving the picking efficiency.

[0046] In order to solve the problem of uneven loading speed of various loading baskets, the present invention organizes harvesting drones into teams according to individual fruit classifications, and each team of drones harvests fruits of the corresponding individual classification; the number of drones in each team is inversely proportional to the size of the corresponding individual fruit classification.

[0047] In addition, if the drones matched or associated with each loading box are notified after each loading box is fully loaded, it is possible that some drones have already picked the fruits and are on the way to the loading box. If the notification is made after the loading box is fully loaded, a large number of drones may have to travel in vain and then be transferred to the loading box that is coordinated / allocated to it again. If the loading box that is allocated again is far away (for example, when the loading box of the same vehicle is also full, it will be considered to allocate a transport vehicle in other areas, that is, a carrier vehicle), thereby reducing work efficiency to a certain extent. On the other hand, the actual size of each fruit is uneven. Although the size of the fruit is obtained in advance by scanning, there are still differences in actual application. Therefore, when the actual loading quantity planned in advance is reached, the loading box may not be actually full, or it may be fully loaded without reaching the number of devices planned in advance. Based on this, the present application determines the remaining loading space through advance detection, and then timely recalculates the remaining loading quantity of the loading basket according to the remaining space, and then adjusts the number of transportations of the drone to the loading box. Specifically, after counting that the number of fruits loaded in a certain loading basket reaches a percentage threshold, the image information of the loading basket when the last harvesting drone dropped into the loading basket is retrieved; and the remaining space of the loading basket is extracted based on the image information of the loading basket; the remaining loading quantity of the loading basket is calculated based on the average individual size of the fruits loaded in the loading basket and the remaining space of the loading basket; thereby more accurately obtaining the quantity that the loading basket can load, avoiding the problem of being not full or unable to load when loading according to the preset plan.

[0048] In order to further balance the loading speed of various loading baskets, in the present invention, after the number of fruits loaded in the first loading basket of the transport vehicle reaches the percentage threshold, the remaining space of all loading baskets is obtained; the remaining loading quantity of the loading basket is calculated according to the average individual size of the fruits loaded in the loading basket and the remaining space of the loading basket; the loading drones are re-formed according to the remaining loading quantity of each loading basket, and the number of drones in each team is proportional to the remaining loading quantity of the corresponding loading basket. Through the above method, not only can the problem caused by the difference between the theoretical loading capacity and the actual loading capacity be solved, but also the loading speed between various loading baskets can be balanced, so that the time when various loading baskets are fully loaded is close to the same, avoiding idle drones.

[0049] In addition, in order to solve the problem of individual fruit and maturity errors caused by occlusion when extracting fruit information, the harvesting drone in the present invention obtains image information of the current fruit to be picked during the picking process according to the planned path; the individual size and maturity of the fruit are extracted according to the image information and checked with the fruit information, and the fruit is picked if they meet the requirements; if the maturity meets the requirements but the individual size does not meet the requirements, a temporary task is inserted to pick the fruit and put it into the corresponding loading basket; if the maturity does not meet the requirements, the picking of the fruit is abandoned. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, each element or part is not necessarily drawn according to the actual scale. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings without paying creative labor.

[0051] Figure 1 These are photos of manual picking and transportation in mountain orchards in a certain area;

[0052] Figure 2 A flowchart of a mountain harvesting method based on a drone swarm and a monorail carrier according to an embodiment of the present invention;

[0053] Figure 3 It is a schematic diagram for extracting fruit information;

[0054] Figure 4 This is a structural diagram of a mountain harvesting system based on the collaborative operation of a drone swarm and a monorail carrier vehicle in the present invention. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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 making creative work are within the scope of protection of the present invention.

[0056] Herein, suffixes such as "module", "component" or "unit" used to represent elements are only used to facilitate the description of the present invention, and have no specific meanings by themselves. Therefore, "module", "component" or "unit" can be used mixedly.

[0057] In this document, the terms "upper", "lower", "inner", "outer", "front", "back", "one end", "the other end" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance.

[0058] In this document, unless otherwise clearly specified and limited, the terms "installed", "provided with", "connected", etc. should be understood in a broad sense. For example, "connected" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection, a direct connection, or an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0059] Herein "and / or" includes any and all combinations of one or more of the associated listed items.

[0060] Herein, "plurality" means two or more than two, ie, it includes two, three, four, five, etc.

[0061] Embodiment 1

[0062] In order to solve the above problems, Figure 2 As shown, the present invention provides a mountain picking method based on the coordinated operation of a drone group and a monorail carrier transport, and the fruits in the orchard are graded and picked according to the individual size. The specific steps include:

[0063] S101 scans the image information and location information of the fruit trees in the orchard.

[0064] The orchard is scanned by a monitoring drone to obtain image information and location information of the fruit trees in the orchard. It is foreseeable that the image information may include 3D point cloud information and 2D color image information, and the 3D point cloud information and the 2D color image information are combined.

[0065] S102 extracts fruit information based on the image information and position information of the fruit trees in the garden, wherein the fruit information includes the quantity, position, individual size, and maturity of the fruits.

[0066] like Figure 3 As shown in the figure, by using the combined 3D point cloud information and 2D color image information, the fruit in the image can be segmented through instance segmentation technology, and then input into the trained model to judge the maturity of the fruit. Generally speaking, the maturity of the fruit can be judged by the individual size and / or color. Of course, the judgment method can be adaptively adjusted between different fruits.

[0067] Of course, due to the occlusion between fruits and between fruits and branches during scanning, it is possible to misjudge the individual fruit and maturity. In order to solve this problem, further solutions will be provided later, which will not be elaborated here.

[0068] In addition, the number and location of fruits can be counted after instance segmentation. The fruits are divided into three groups: large, medium and small according to their individual sizes, which will serve as the basis for the subsequent drone grouping.

[0069] In actual operation, the fruits can also be grouped according to their own standards, and adaptive adjustments can be made in subsequent steps.

[0070] It should be noted that the difference between the "monitoring drone" in this implementation and the "harvesting drone" used in the subsequent steps is only in the operation, and the drones themselves can be the same in structure. That is, when the drone performs the monitoring task, it is called a "monitoring drone", and when it performs the picking task, it is called a "harvesting drone". Of course, special drones can also be equipped according to functional requirements, and this is not limited in the present invention.

[0071] S103 plans the picking area, picking path and picking order according to the fruit information.

[0072] After obtaining the number and location of all fruits in the orchard, the picking area, picking path and picking order are planned for picking.

[0073] In order to improve efficiency and avoid long-distance round trips, the orchard is divided into several picking areas, and each drone picks within the assigned picking area.

[0074] In addition, a parking spot for the transport vehicle is set, and the picking path is planned according to the parking spot and the location of the fruit. In the present invention, the drone and the transport vehicle work in collaboration, which means that not only the drone is required to find the location of the transport vehicle, but also the transport vehicle is required to cater to the drone and provide a most convenient parking spot. Specifically, in this embodiment, the transport vehicle sets a parking position according to the planned path of the picking drone, including:

[0075] S301 obtains the positions of the first N fruits that the picking drone needs to pick according to the picking order.

[0076] The specific number of N can be adjusted according to the actual situation. That is to say, after picking N fruits, the current parking point may no longer meet the requirement of reducing the distance of the drone, so it needs to be adjusted. It can be adaptively adjusted according to the area of ​​the orchard, the location of the fruit, etc., which will not be described in detail in this embodiment.

[0077] S302 traverses all parking points in the driving path of the transport vehicle, and calculates the total distance between each parking point and the positions of the N fruits.

[0078] In this implementation, in order to cater to the special topography of mountain orchards, the transport vehicles used are both rail and land. In the mountain orchards, tracks are set up for the transport vehicles to run smoothly, and after going down the mountain, the land mode is adopted to save costs. In practice, in order to further save costs, the tracks are all single-track, that is, there is only one round-trip track. Therefore, a loop line can be set up around the orchard to connect with the round-trip track. First, it is convenient to set up multiple parking spots to facilitate the delivery of fruits by picking drones, and second, when multiple transport vehicles are running at the same time, it avoids congestion.

[0079] In this step, any point on the loop can be used as a parking point. These parking points are traversed, and then the total distance between each parking point and the location of N fruits is calculated, that is, the total distance picked by the drone.

[0080] S303 selects the parking point with the shortest total distance as the current parking point of the transport vehicle.

[0081] The parking point with the shortest distance is the most convenient parking point, so it is set as the current parking point.

[0082] S304 obtains the positions of the next N fruits and executes steps S302 and S303.

[0083] After the current N fruits are picked, the positions of the next N fruits are obtained, and steps S302 and S303 are executed.

[0084] S305 repeats step S304 until picking is completed.

[0085] After the current N fruits are picked, the positions of the next N fruits are obtained, and so on until all the fruits are picked.

[0086] The S104 harvesting drone picks fruits according to the planned picking area, picking path and picking order; the picked fruits are placed into the corresponding loading baskets on the transport vehicle according to individual fruit classification. After successful placement, a signal is fed back to the upper computer for counting by the upper computer.

[0087] Before picking, the harvesting drones are organized into teams according to the individual classification of the fruits, and each team of drones picks the fruits of the corresponding individual classification; the number of drones in each team is inversely proportional to the size of the corresponding individual classification of fruits.

[0088] In this embodiment, the fruits are divided into three categories according to their individual sizes: large, medium and small, so the loading baskets on the transport vehicle are also divided into three categories: A, B and C. Large fruits are placed in the A-category basket, medium fruits are placed in the B-category basket, and small fruits are placed in the C-category basket.

[0089] The harvesting drones are also divided into three teams: A, B, and C. Team A's harvesting drones pick large fruits and put them into Class A boxes, Team B's harvesting drones pick medium-sized fruits and put them into Class B boxes, and Team C's harvesting drones pick small fruits and put them into Class C boxes.

[0090] Of course, the harvesting drones of Team A are also grouped according to the harvesting areas. Similarly, the harvesting drones of Teams B and C are also grouped as above.

[0091] Of course, it is foreseeable that, since the size of the ABC boxes is the same, the number of fruits required for full load is different. For example, the A box may be fully loaded with 100 large individual fruits, while the C box may need 120 small individual fruits to be fully loaded. In order to solve the problem of imbalanced quantity, the number of drones in each team of the present invention is inversely proportional to the size of the corresponding fruit individual classification, that is, the number of drones in team A is smaller than that of team C, and the difference in the number of fruits is balanced by the number of drones.

[0092] The host computer sends the location of each fruit, the location of the transport vehicle parking point, the picking route planned according to the fruit location and the transport vehicle parking point, the picking order, and the working area of ​​each picking drone to the picking drone, and the picking drone can pick according to the plan.

[0093] In addition, since there are many cases of overlapping fruits and overlapping fruits and branches during scanning in step S101, errors may occur when extracting fruit size and maturity information in step S102, so a review is required during the specific picking process.

[0094] Specifically, when the harvesting drone is picking fruits according to the planned path, it obtains image information of the fruits to be picked; it extracts the individual size and maturity of the fruits based on the image information and checks them with the fruit information, and picks the fruits if they meet the requirements; if the maturity meets the requirements but the individual size does not, it inserts a temporary task to pick the fruits and put them into the corresponding loading basket; if the maturity does not meet the requirements, it abandons the picking of the fruits.

[0095] When the harvesting drone is picking fruits according to the planned path, it can have close and shielded contact with the picked fruits (because it is sequential picking, the outer fruits will be picked first and then the inner fruits during planning), so that the fruits can be fully exposed to the lens of the harvesting drone, thereby obtaining a complete and clear image of the fruit.

[0096] After obtaining a complete and clear image of the fruit, the model is input again to judge the individual and maturity, and compared with the fruit information stored on the host computer before. If it is verified, the original picking plan is executed. If the maturity meets the requirements but the individual is different, the original picking task for the fruit is deleted, and a new task is inserted for the drone to put the fruit into the corresponding loading basket after picking. For example, a certain fruit is judged as a medium individual when it is blocked, but when picking, it is found that the individual should be a large individual, so it is necessary to delete the task of putting it into the medium individual loading basket and create a new task of loading it into the large individual loading basket.

[0097] Of course, if the maturity of the fruit is incorrect, it means that the maturity does not meet the requirements. Regardless of the individual, the current fruit picking task will be deleted and the next picking task will be executed.

[0098] In addition, after each harvesting drone successfully drops a fruit, it will send a successful drop signal to the host computer, which will facilitate the host computer to count the number of fruits loaded in each loading basket.

[0099] After counting that the number of fruits loaded in a certain loading basket reaches a percentage threshold, S105 retrieves the image information of the loading basket when the last harvesting drone dropped the fruit into the loading basket; and extracts the remaining space of the loading basket based on the image information of the loading basket.

[0100] Since the present invention adopts the method of classified picking, although the problem of quantity imbalance has been solved as much as possible by matching different numbers of drones in the formation, in the actual picking process, even individuals of the same species have differences in body shape, and the irregular shapes make it difficult to accurately predict the number of items that can be loaded in a loading basket.

[0101] In order to solve the problem that a certain type of loading basket has been loaded, but the other loading baskets are still loading, making the drone empty, the present invention sets a percentage threshold according to the expected loading quantity. For example, under normal circumstances, the loading basket can load 100 large individual fruits, so an 80% percentage threshold is set, that is, when 80 fruits are loaded, the image information of the loading basket is retrieved when the harvesting drone that puts the 80th fruit in it drops the 80th fruit. Through the image information, the accumulation of fruits in the loading basket and the remaining space can be observed.

[0102] S106 calculates the remaining loading quantity of the loading basket according to the average individual size of the fruits loaded in the loading basket and the remaining space of the loading basket.

[0103] Even though they are the same type of fruit, there are still differences in individual fruits. For a loading basket that ideally loads 100 large individual fruits, if 80 fruits are loaded, ideally it should occupy 80% of the total volume, but in reality it may occupy more space or less space.

[0104] Therefore, this step will be more accurate in calculating the amount that can be loaded in the remaining space according to the average volume of the existing fruits.

[0105] For example, out of a total of 100 units of space, 80 fruits have already occupied 90 units of space. Therefore, we can roughly deduce that on average each fruit picked this time will take up 9 / 8 units of space. In the remaining 10 units of space, the number that can be loaded is 10 / (9 / 8), which is approximately equal to 8 fruits, rather than the originally planned 20 fruits.

[0106] The S107 picking drone delivers the harvested food to the loading basket according to the remaining loading quantity.

[0107] Continuing with the previous example, the harvesting drone only needs to drop 8 fruits into the loading basket to fully load it, rather than the originally planned 20.

[0108] In addition, this embodiment also provides a further solution to solve the problem of uneven loading speed of various loading baskets, specifically including:

[0109] After counting that the number of fruits loaded in the first loading basket of the transport vehicle reaches a percentage threshold, S201 obtains the remaining space of all loading baskets.

[0110] The method for obtaining the remaining space of the loading basket refers to step S105 and will not be repeated here.

[0111] S202 calculates the remaining loading quantity of the loading basket according to the average individual size of the fruits loaded in the loading basket and the remaining space of the loading basket.

[0112] The method for calculating the remaining load capacity refers to step S106 and will not be repeated here.

[0113] S203: Re-forming the loading drones into teams according to the remaining load quantity of each loading basket, and the number of drones in each team is proportional to the remaining load quantity of the corresponding loading basket.

[0114] After obtaining the actual remaining load quantity of the three types of loading baskets, the drones can be rearranged to allocate an appropriate number of drones so that the speeds at which the three types of loading baskets are fully loaded are ultimately close to the same.

[0115] S108 After all loading baskets are fully loaded, the transport vehicle drives to the designated location to unload, and the picking drone delivers the goods to the loading basket in the next transport vehicle.

[0116] In order to avoid the problem of the picking drone waiting in vain, in this embodiment, when the loading basket of the current transport vehicle reaches a certain threshold, the next empty transport vehicle can be called to wait in the orchard. When the current transport vehicle is fully loaded, the picking drone will drop the fruit into the loading basket of the next transport vehicle, and the cycle will continue until the picking task is completed.

[0117] Embodiment 2

[0118] like Figure 4 As shown, the present invention also provides a mountain picking system based on a drone swarm and a monorail carrier, comprising:

[0119] Monitoring drones are used to scan image information and location information of fruit trees in orchards;

[0120] The host computer is used to extract fruit information based on the image information and location information of the fruit trees in the garden obtained by the monitoring drone, wherein the fruit information includes the number, location, individual size and maturity of the fruit; and plan the picking area and picking path based on the fruit information;

[0121] The harvesting drone is used to harvest fruits according to the harvesting area and harvesting path planned by the host computer;

[0122] A carrier (i.e., a transport vehicle) is used to load the fruits picked by the harvesting drone and transport the fruits to a designated location; the transport vehicle is provided with a number of loading baskets for loading different individual categories;

[0123] The harvesting drone will place the harvested fruits into corresponding loading baskets according to the individual classification of the fruits, and after successful placement, it will feedback a signal to the upper computer to facilitate counting by the upper computer;

[0124] After the upper computer counts that the number of fruits loaded in a certain loading basket reaches the percentage threshold, it retrieves the image information of the loading basket when the last harvesting drone dropped into the loading basket; and extracts the remaining space of the loading basket based on the image information of the loading basket;

[0125] Calculate the remaining loading quantity of the loading basket according to the average individual size of the fruits loaded in the loading basket and the remaining space of the loading basket;

[0126] The picking drone delivers the harvest to the loading basket according to the remaining loading quantity;

[0127] After all loading baskets are fully loaded, the transport vehicle drives to the designated location to unload, and the picking drone delivers the goods to the loading basket in the next transport vehicle.

[0128] The number of the transport vehicles is two or more, and the transport vehicles are both rail- and land-based.

[0129] The individual fruits are classified into three categories: large, medium and small, and the types of loading baskets used to load the fruits correspond to the individual fruit classifications.

[0130] The monitoring drone and the harvesting drone are both equipped with a 3D point cloud camera and a 2D color camera.

[0131] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0132] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are within the protection of the present invention.

Claims

1. A mountain picking method based on drone swarm and monorail carrier, characterized in that include: S101 scans the image and location information of fruit trees in the orchard; S102 extracts fruit information based on the image information and position information of the fruit trees in the garden, wherein the fruit information includes the number, position, individual size, and maturity of the fruit; S103 plans the picking area, picking path and picking order according to the fruit information; S104 The harvesting drone harvests fruits according to the planned harvesting area, harvesting path and harvesting order; the harvested fruits are placed into corresponding loading baskets on the carrier according to the individual classification of the fruits, and after successful placement, a signal is fed back to the upper computer for counting by the upper computer; the individual fruits are classified into three categories: large, medium and small, and the types of loading baskets used to load the fruits correspond to the individual classification of the fruits; After counting that the number of fruits loaded in a certain loading basket reaches a percentage threshold, S105 retrieves the image information of the loading basket when the last harvesting drone dropped the fruit into the loading basket; and extracts the remaining space of the loading basket according to the image information of the loading basket; S106 calculates the remaining loading quantity of the loading basket according to the average individual size of the fruits loaded in the loading basket and the remaining space of the loading basket; The S107 picking drone delivers the harvested food to the loading basket according to the remaining loading quantity; S108 After all loading baskets are fully loaded, the transport vehicle drives to the designated location to unload, and the picking drone delivers the goods to the loading basket in the next transport vehicle.

2. A mountain picking method based on drone swarm and monorail carrier according to claim 1, characterized in that: The harvesting drones are organized into teams according to the individual classification of the fruits, and each team of drones picks the fruits of the corresponding individual classification; The number of drones in each team is inversely proportional to the size of the corresponding fruit individual classification.

3. A mountain picking method based on drone swarm and monorail carrier according to claim 2, characterized in that Also includes: After counting that the number of fruits loaded in the first loading basket of the transport vehicle reaches a percentage threshold, S201 obtains the remaining space of all loading baskets; S202 calculates the remaining loading quantity of the loading basket according to the average individual size of the fruits loaded in the loading basket and the remaining space of the loading basket; S203: Re-forming the loading drones into teams according to the remaining load quantity of each loading basket, and the number of drones in each team is proportional to the remaining load quantity of the corresponding loading basket.

4. The mountain picking method based on drone swarm and monorail carrier according to claim 1 is characterized by: The transport vehicle sets a parking position according to the planned path of the harvesting drone, including: S301 obtains the positions of the first N fruits that the picking drone needs to pick according to the picking order; S302 traverses all parking points on the driving path of the transport vehicle, and calculates the total distance between each parking point and the positions of the N fruits; S303 selects the parking point with the shortest total distance as the current parking point of the transport vehicle; S304 obtains the positions of the next N fruits and executes steps S302 and S303; S305 repeats step S304 until picking is completed.

5. The mountain picking method based on drone swarm and monorail carrier according to claim 1 is characterized in that: The harvesting drone obtains image information of the fruit to be picked while picking according to the planned path; The individual size and maturity of the fruit are extracted based on the image information and checked with the fruit information. If they meet the requirements, the fruit will be picked; if the maturity meets the requirements but the individual size does not, a temporary task is inserted to pick the fruit and put it into the corresponding loading basket; if the maturity does not meet the requirements, the picking of the fruit will be abandoned.

6. The mountain picking method based on drone swarm and monorail carrier according to claim 1 is characterized by: Determine the maturity of the fruit based on its color and / or individual size.

7. A mountain harvesting system based on drone swarms and monorail carriers, characterized in that include: Monitoring drones are used to scan image information and location information of fruit trees in orchards; The host computer is used to extract fruit information based on the image information and location information of the fruit trees in the garden obtained by the monitoring drone, wherein the fruit information includes the number, location, individual size and maturity of the fruit; and plan the picking area and picking path based on the fruit information; The harvesting drone is used to harvest fruits according to the harvesting area and harvesting path planned by the host computer; A transport vehicle is used to load the fruits picked by the harvesting drone and transport the fruits to a designated location; the transport vehicle is provided with a plurality of loading baskets for loading fruits of different categories; wherein the fruits are classified into three categories: large, medium and small, and the types of loading baskets for loading fruits correspond to the categories of fruits; The harvesting drone will place the harvested fruits into corresponding loading baskets according to the individual classification of the fruits, and after successful placement, it will feedback a signal to the upper computer to facilitate counting by the upper computer; After the upper computer counts that the number of fruits loaded in a certain loading basket reaches the percentage threshold, it retrieves the image information of the loading basket when the last harvesting drone dropped into the loading basket; and extracts the remaining space of the loading basket based on the image information of the loading basket; Calculate the remaining loading quantity of the loading basket according to the average individual size of the fruits loaded in the loading basket and the remaining space of the loading basket; The picking drone delivers the harvest to the loading basket according to the remaining loading quantity; After all loading baskets are fully loaded, the transport vehicle drives to the designated location to unload, and the picking drone delivers the goods to the loading basket in the next transport vehicle.

8. A mountain harvesting system based on a drone swarm and a monorail carrier according to claim 7, characterized in that: The number of the transport vehicles is two or more, and the transport vehicles are both rail- and land-based.

9. The mountain harvesting system based on drone swarm and monorail carrier according to claim 7, characterized in that: The monitoring drone and the harvesting drone are both equipped with a 3D point cloud camera and a 2D color camera.

Images