Fruit picking method based on picking robot, picking robot and electronic equipment

Abstract

The present invention provides a fruit picking method based on a picking robot, a picking robot and an electronic device, which relate to the field of fruit picking, and include S1, obtaining map information of an orchard; S2, establishing a spatial model of an orchard map and a picking robot according to the map information and depth information obtained by a depth camera, and obtaining a coordinate set of fruit distribution; S3, performing preliminary path planning covering the coordinate set according to a preset path planning algorithm according to the spatial model and the coordinate set; S4, establishing a kinematic model of the robot; S5, determining a kinematic inverse solution of different picking postures of the robot according to the coordinate set and the kinematic model; S6, optimizing the preliminary path planning into a navigation path according to the kinematic inverse solution; S7, executing the navigation path through a mobile chassis to realize fruit picking, wherein the mobile chassis is connected to the robot arm and drives the robot arm to move, and can complete the picking of fruits in the orchard with high efficiency without collision, thereby saving energy consumption.

Description

Technical Field

[0001] The present invention relates to the field of fruit picking, and in particular to a fruit picking method based on a picking robot, a picking robot and electronic equipment. Background Art

[0002] In existing fruit picking robots, path planning usually focuses more on how to quickly and accurately reach the picking location, but less on the impact of the robot arm's posture on the mobile chassis when performing the picking operation. This design flaw may lead to a series of problems in the actual picking process, thus affecting the efficiency and success rate of picking.

[0003] First, since the influence of the posture of the robot arm on the mobile chassis when performing picking is ignored, the robot may not be able to stably stay on the preset path. When the robot arm performs picking operations, changes in its posture may affect the chassis, causing the robot to deviate from the preset path, which will not only reduce the picking efficiency, but also cause fruit damage or omission.

[0004] Secondly, existing robots have limited adaptability to tree trunks and branches during the picking process. Due to the lack of consideration of the relationship between the robot arm posture and the chassis in path planning, it is difficult for the robot to flexibly adjust its posture according to the actual distribution of tree trunks and branches, which makes it easy for the robot to collide or get stuck when encountering obstacles during the picking process. Summary of the invention

[0005] The technical problem to be solved by the present invention is that the existing fruit picking robots basically do not consider or consider very little the impact of the posture of the robotic arm on the chassis when performing picking in the path planning, and no evaluation index is established to evaluate the quality of the chassis navigation path of the robotic arm during the picking process. The purpose is to provide a fruit picking method, a picking robot and an electronic device based on a picking robot to solve the above problem.

[0006] A fruit picking method based on a picking robot comprises S1, obtaining map information of an orchard, S2, establishing a spatial model of an orchard map and a robotic arm picking according to the map information and depth information obtained by a depth camera, and obtaining a coordinate set of fruit distribution;

[0007] S3, performing preliminary path planning covering the coordinate set according to the spatial model and the coordinate set and a preset path planning algorithm;

[0008] S4, establish the kinematic model of the robot arm;

[0009] S5, determining the inverse kinematic solutions of different picking postures of the robotic arm according to the coordinate set and the kinematic model;

[0010] S6, optimizing the preliminary path planning into a navigation path according to the inverse kinematic solution;

[0011] S7, executing the navigation path through the mobile chassis to achieve the picking of fruits, wherein the mobile chassis is connected to the mechanical arm and drives the mechanical arm to move.

[0012] Furthermore, the map information is formed by the laser radar collecting information in the orchard in advance.

[0013] Furthermore, the S3 further includes:

[0014] S31, the spatial model is a collection of point cloud data, and a grid map for path planning is constructed by matching and calibrating the point cloud data and the coordinate set;

[0015] S32, inputting the grid map into a preset path planning algorithm to complete the preliminary path planning covering the coordinate set.

[0016] Furthermore, in S3, the preset path planning algorithm is an ant colony algorithm.

[0017] Furthermore, in S4, the kinematic model is determined according to DH parameters of the robotic arm.

[0018] Furthermore, the S6 further includes:

[0019] S61, determining a loss function according to a preset multi-objective to determine an optimal solution of the kinematic inverse solution;

[0020] S62, taking the optimal solution as a constraint condition, adjusting the preliminary path planning into a navigation path.

[0021] Furthermore, it also includes: S8, performing the navigation path evaluation on the last mobile chassis, adjusting the kinematic inverse solution according to the evaluation result, obtaining the adjusted kinematic inverse solution, and updating the navigation path according to the kinematic inverse solution.

[0022] Furthermore, the evaluation index includes the number of collisions.

[0023] Another aspect of the present invention provides a harvesting robot, comprising:

[0024] A robotic arm, used to adjust different picking postures;

[0025] A depth camera, disposed on the robotic arm and moving with the robotic arm, for obtaining depth information;

[0026] An actuator, which is disposed on the mechanical arm and moves together with the mechanical arm, and is used to perform the action of picking fruits;

[0027] a mobile chassis, connected to the robotic arm, for executing a navigation path in the orchard;

[0028] A controller is used to implement the above method by controlling the mechanical arm, the depth camera, the picking actuator and the mobile chassis.

[0029] Another aspect of the present invention provides an electronic device, including:

[0030] A processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the method is implemented when the processor executes the program.

[0031] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0032] 1. The present invention proposes a fruit picking method based on a picking robot. According to the posture of the robot arm, a picking position where the chassis can perform path planning is obtained, and the posture of the robot arm during the picking process is added as a constraint to the chassis path planning algorithm, which is helpful to improve the efficiency of fruit picking. Specifically, it can complete the picking of fruits in the orchard with high efficiency without collision, saving energy consumption; a model for path planning of the optimal picking position is built, and then by establishing evaluation indicators, fruit picking based on the picking robot is realized, which is helpful to improve productivity, and can help farmers and scientific researchers better plan resources and liberate excess labor;

[0033] 2. The present invention proposes a picking robot and electronic equipment, which obtain the picking position where the chassis can perform path planning according to the position of the robot arm, and the posture of the robot arm during the picking process is added as a constraint to the chassis path planning algorithm, which is helpful to improve the efficiency of fruit picking, specifically, it can complete the picking of fruits in the orchard with high efficiency without collision, saving energy consumption; build a model for path planning of the optimal picking position, and then realize fruit picking based on the picking robot by establishing evaluation indicators, which is helpful to improve productivity, and can help farmers and scientific researchers better plan resources and liberate excess labor;

[0034] 3. The present invention proposes an electronic device, which obtains the picking position where the chassis can perform path planning according to the posture of the robot arm, and adds the posture of the robot arm during the picking process as a constraint to the chassis path planning algorithm, which helps to improve the efficiency of fruit picking. Specifically, it can complete the picking of fruits in the orchard with high efficiency without collision, saving energy. A model for path planning at the optimal picking position is built, and then evaluation indicators are established to realize fruit picking based on the picking robot, which helps to improve productivity and can help farmers and scientific researchers better plan resources and liberate excess labor. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The accompanying drawings are used to better understand the present invention and do not constitute a limitation of the present invention.

[0036] Figure 1 is a schematic diagram of a grape picking method based on a picking robot according to an embodiment of the present invention;

[0037] Figure 2 is a schematic diagram of the operation process of a picking robot according to an embodiment of the present invention;

[0038] Figure 3 is a schematic diagram of an ant colony algorithm process according to an embodiment of the present invention;

[0039] Figure 4 is a schematic diagram of establishing a DH model according to an embodiment of the present invention;

[0040] Figure 5 is a schematic diagram of a harvesting robot according to an embodiment of the present invention;

[0041] Figure 6 is a schematic diagram of an electronic device according to an embodiment of the present invention.

[0042] Reference numerals:

[0043] 100 - picking robot, 101 - mechanical arm, 102 - depth camera, 103 - actuator, 104 - mobile chassis, 105 - control unit, 200 - electronic device, 201 - processor, 202 - memory. DETAILED DESCRIPTION

[0044] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.

[0045] like Figure 1 As shown, the present invention provides a fruit picking method based on a picking robot 100, including S1, obtaining map information of an orchard, S2, establishing a spatial model of an orchard map and a picking robot 101 according to the map information and depth information obtained by a depth camera 102, and obtaining a coordinate set of fruit distribution;

[0046] S3, performing preliminary path planning covering the coordinate set according to the spatial model and the coordinate set and a preset path planning algorithm;

[0047] S4, establishing a kinematic model of the robot arm 101;

[0048] S5, determining the inverse kinematic solutions of different picking postures of the robot arm 101 according to the coordinate set and the kinematic model;

[0049] S6, optimizing the preliminary path planning into a navigation path according to the inverse kinematic solution;

[0050] S7, executing the navigation path through the mobile chassis 104 to achieve the picking of fruits, wherein the mobile chassis 104 is connected to the mechanical arm 101 and drives the mechanical arm 101 to move.

[0051] In some embodiments, Figure 2 As shown, the operation process of the picking robot 100 may include:

[0052] Including S1, obtaining map information corresponding to the orchard through lidar mapping.

[0053] S2, extracting the picking location of the picking robot 100 based on the collected map information and in combination with the depth information acquired by the depth camera 102.

[0054] S3, performing coverage path planning according to the extracted coordinates.

[0055] S4, establishing a kinematic model of the robot arm 101 according to the DH parameters of the robot arm 101.

[0056] S5, finding the inverse kinematic solution for different picking postures of the robot arm 101 at each point.

[0057] S6, optimize the planned path according to the result of the inverse kinematics solution. Inverse kinematics aims to solve the joint angle according to the target posture, so as to drive the robot arm 101 to achieve the required movement. For different picking postures, we need to iteratively calculate the inverse kinematics solution according to the specific task requirements and constraints to find a joint angle combination that meets the conditions.

[0058] S7, after initialization, the picking robot 100 navigates in the orchard according to the optimized path.

[0059] The contribution of the present invention is that it can obtain the picking position where the chassis can perform path planning based on the posture of the robot arm 101; and add the posture of the robot arm 101 during the picking process as a constraint to the chassis path planning algorithm and optimize the existing path; and also evaluate the quality of the path of the robot arm 101 during the picking process by establishing evaluation indicators.

[0060] The map information is formed by the laser radar collecting information in the orchard in advance.

[0061] In this embodiment, the formation of map information may include the steps of equipment selection, data collection, data processing and map update. According to the orchard environment and navigation requirements, a suitable laser radar device is selected, and a multi-line laser radar is used to obtain more three-dimensional spatial information and improve the accuracy of the map. At the same time, a stable and reliable device should be selected to adapt to the complex environment in the orchard. Data collection refers to data collection in the orchard using the selected laser radar device. The entire orchard should be covered as much as possible during the collection process, including features such as trees, roads, and buildings. Data processing refers to processing the collected laser radar data, including filtering, denoising, and registration steps. Through data processing, feature points, obstacles, roads and other information in the orchard are extracted to construct a map model of the orchard. Map update refers to the use of laser radar equipment for re-collection or local update. At the same time, the accuracy and real-time nature of the map should be maintained to meet navigation requirements.

[0062] The S3 further includes:

[0063] S31, the spatial model is a collection of point cloud data, and a grid map for path planning is constructed by matching and calibrating the point cloud data and the coordinate set;

[0064] S32, inputting the grid map into a preset path planning algorithm to complete the preliminary path planning covering the coordinate set.

[0065] In some embodiments, the orchard map and the picking space model of the robot arm 101 are specifically included. Taking grape fruit picking as an example, the distribution position of the grape fruit is determined based on the depth information collected by the depth camera 102, and the laser radar obtains the point cloud data of the grape orchard, and the coordinates of the fruit distribution obtained by the depth camera 102 are combined to establish a grid map for path planning. Specifically, the model uses the depth camera 102 and the laser radar to collect and process data to achieve accurate fruit positioning and path planning. First, the depth camera 102 can obtain the three-dimensional position data of the distribution of grape fruits by capturing the depth information of the orchard scene. After processing, these data can accurately identify the position of each grape in space, providing key positioning information for subsequent picking operations. At the same time, the laser radar, as an active sensor, can obtain detailed point cloud data of objects in the orchard by emitting lasers to the orchard and receiving the reflected signals. These point cloud data contain rich three-dimensional structural information, which is helpful to further analyze the layout and obstacle distribution of the orchard. After obtaining the data of the depth camera 102 and the laser radar, a series of data processing and fusion operations need to be performed. By matching and calibrating the coordinates of the fruit distribution with the point cloud data acquired by the lidar, a grid map for path planning can be constructed.

[0066] The grid map divides the orchard into a series of grids, each grid representing a harvestable area. When performing a harvesting task, the robot arm 101 can select the optimal path to move according to the information in the grid map to achieve efficient and accurate harvesting operations.

[0067] In S3, the preset path planning algorithm is an ant colony algorithm.

[0068] In some embodiments, the fruit picking coverage path planning algorithm considers the robot picking posture, and constructs the grape picking robot 100 chassis coverage path planning algorithm based on the ACO (Ant Colony Optimization) algorithm to realize the function of covering the picking of the picking robot 100 in the orchard. That is, the path planning algorithm adopts the ant colony algorithm. The ant colony algorithm is a bionic algorithm derived from the simulation of the path-finding method of ants in nature: during the movement, ants can leave pheromones on the path they pass through to transmit information, and ants can sense this substance during the movement and use it to guide their own movement direction. The collective behavior of an ant colony composed of a large number of ants shows a positive feedback phenomenon of information: the more ants that walk on a certain path, the greater the probability that the later ones will choose this path.

[0069] Specifically, the walking paths of ants represent feasible solutions to the problem to be optimized, and all paths of the entire ant colony constitute the solution space of the problem to be optimized. Ants with shorter paths release more pheromones. As time goes by, the accumulated pheromone concentration on the shorter path gradually increases, and the number of ants choosing this path also increases. Eventually, all the ants will converge on the best path under the effect of positive feedback, and this corresponds to the optimal solution to the problem to be optimized. The basic flow chart of the algorithm is as follows: Figure 3 shown.

[0070] The mathematical model of the ant colony algorithm is:

[0071]

[0072] In the above formula, τ ij (t) is the information density on the search path from position i to position j at time t, η ij (t) is the heuristic function of the algorithm. Its value is usually not fixed but expressed as the inverse of the path distance from position i to position j, that is: η ij (t) = 1 / d ij ; α is the pheromone factor of the algorithm, usually in the range of [1,4]; β is the heuristic function factor of the algorithm, usually in the range of [0,5]; allowed k It represents the set of spatial nodes that have not been visited in the problem space, which can be used as the target point set for the next move.

[0073] Since the pheromones left by ants are constantly accumulating during the iteration process, the residual pheromones produced as the running time increases and the number of ants increases will cause the pheromone heuristic factor to be unable to play its role. Therefore, every time a path optimization is completed, the ant pheromones must be updated in time. At time t+1, the update strategy of pheromones from position i to position j is:

[0074] τij(t+1)=τij(t)*(1-ρ)+Δτij(t)(2)

[0075]

[0076] In the above formula (2), ρ is the pheromone volatility coefficient.

[0077] In S4 , the kinematic model is determined according to the DH parameters of the robot arm 101 .

[0078] Furthermore, if Figure 4 As shown, based on the known DH parameters of the robot arm 101, a rigorous kinematic model can be constructed to provide theoretical support for subsequent inverse kinematics solutions and path optimization. DH parameters are a parameterized method for describing the position and posture of the robot link, through which a mathematical relationship can be established between the joint angle of the robot arm 101 and the position and posture of the end effector. Based on these parameters, the kinematic model of the robot arm 101 can be derived to further analyze its motion characteristics and potential constraints.

[0079] The S6 further includes:

[0080] S61, determining a loss function according to a preset multi-objective to determine an optimal solution of the kinematic inverse solution;

[0081] S62, taking the optimal solution as a constraint condition, adjusting the preliminary path planning into a navigation path.

[0082] In addition, path optimization aims to enable the robot arm 101 to complete the picking task more efficiently and smoothly by adjusting the joint angles and motion trajectories. This process usually involves optimizing the joint angles and motion time to achieve the best picking effect and the minimum energy consumption.

[0083] In the process of path optimization, multi-objective optimization is a common problem, which aims to meet multiple performance indicators at the same time. In order to achieve this goal, it is necessary to first define a loss function that can comprehensively evaluate the impact of various objectives. The choice of loss function should be determined according to the needs of the actual application, such as time, energy, smoothness, picking effect, etc. In one embodiment of the present invention, collision avoidance and energy consumption are selected as the criteria for determining the loss function. After the loss function is determined, the inverse kinematics solution method will be used to find the joint angle combination that minimizes the loss function value, that is, the optimal solution.

[0084] After obtaining the optimal solution, we need to use it as a constraint to adjust the preliminary path planning. The preliminary path planning is usually based on certain assumptions and simplified conditions, and may not fully conform to the actual situation. By using the optimal solution as a constraint, we can correct and optimize the preliminary path planning to make it more in line with the actual environment and task requirements. The optimized path will be used as a navigation path to guide the robot arm 101 to complete the picking task, which can effectively avoid collisions and save energy.

[0085] The method further includes: S8, performing the navigation path evaluation on the last movement of the mobile chassis 104, adjusting the kinematic inverse solution according to the evaluation result, obtaining the adjusted kinematic inverse solution, and updating the navigation path according to the kinematic inverse solution.

[0086] Specifically, when evaluating the navigation path executed by the mobile chassis 104, it is first necessary to clarify the evaluation indicators and standards. These indicators may include the length, smoothness, energy consumption and other aspects of the path. For one embodiment of the present invention, the number of collisions and energy consumption are determined as evaluation indicators. Through the comprehensive evaluation of these indicators, the performance of the navigation path can be fully understood.

[0087] Based on the evaluation results, we further analyze the problems and deficiencies in the navigation path. These may involve problems such as a long path, excessive energy consumption, or insufficient smoothness. In order to improve these problems, the inverse kinematic solution needs to be adjusted. The process of adjusting the inverse kinematic solution involves optimizing the joint angles and motion trajectories. Through iterative calculations and optimization algorithms, a new combination of inverse solutions can be found to improve the performance of the navigation path. This process needs to fully consider the actual environment and task requirements to ensure that the adjusted inverse solution is feasible and practical.

[0088] Once the adjusted inverse kinematic solutions are obtained, the navigation path can be updated based on these solutions. The updated path is more in line with actual needs, can improve picking efficiency, reduce energy consumption and enhance the smoothness of movement. By continuously adjusting and optimizing the inverse kinematic solutions, the picking path of the robot arm 101 can be gradually improved to achieve more efficient and accurate picking operations.

[0089] The evaluation index includes the number of collisions.

[0090] In an exemplary technical solution, for the grape picking scene, each collision of the picking robot 100 may mean the loss of a grape fruit. Therefore, compared with evaluation indicators such as time and path length, collision-free is more important. Reducing the number of collisions as an evaluation indicator means better economy and reliability, which is more beneficial to improving productivity and helping farmers and researchers to better plan resources and liberate excess labor. Generally speaking, tree trunks, branches and leaves, uneven roads and other reasons are the main causes of collisions of the picking robot 100. Using the number of collisions as an evaluation indicator, these problems can be avoided through multiple iterations.

[0091] The present invention further provides a picking robot 100, comprising:

[0092] The mechanical arm 101 is used to adjust different picking postures;

[0093] A depth camera 102 is provided on the mechanical arm 101 and moves together with the mechanical arm 101 to obtain depth information;

[0094] An actuator 103 is provided on the mechanical arm 101 and moves together with the mechanical arm 101 to perform the action of picking fruits;

[0095] A mobile chassis 104 connected to the mechanical arm 101 for executing a navigation path in the orchard;

[0096] The controller is used to implement the above method by controlling the robotic arm 101, the depth camera 102, the picking actuator 103 and the mobile chassis 104.

[0097] In some implementations, the specific description can be found in the above embodiments, which will not be repeated here. Among them, the control unit 105 can be implemented by a processor 201 and a memory 202, which will be described in detail below, but is not limited to this, and can also be implemented using other technical routes such as integrated IC, FPGA, ASIC, GPU and their combination.

[0098] Figure 5 An exemplary harvesting robot 100 suitable for implementing an embodiment of the present invention is shown. The harvesting robot 100 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiment of the present invention.

[0099] The present invention further provides an electronic device 200, comprising:

[0100] The processor 201, the memory 202, and the computer program stored in the memory 202 and executable on the processor 201, when the processor 201 executes the program, the above method is implemented.

[0101] It should be noted that, Figure 6 As shown, the electronic device 200 in this example is in the form of a general computing device. The components of the electronic device 200 may include but are not limited to: one or more processors 201 or processing units, a system memory 202, and a bus connecting different system components (including the system memory 202 and the processing unit).

[0102] The electronic device 200 includes a computing unit, which can perform various appropriate actions and processes according to a computer program stored in a ROM (Read-Only Memory, read-only memory 202) or a computer program loaded from a storage unit to a RAM (Random Access Memory, random access / access memory 202). In the RAM, various programs and data required for the operation of the device can also be stored. The computing unit, ROM and RAM are connected to each other through a bus. The I / O (Input / Output, input / output) interface is also connected to the bus.

[0103] Multiple components in the electronic device 200 are connected to the I / O interface, including: input units, such as keyboards, mice, etc.; output units, such as various types of displays, speakers, etc.; storage units, such as disks, optical disks, etc.; and communication units, such as network cards, modems, wireless communication transceivers, etc. The communication unit allows the device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunication networks.

[0104] The electronic device 200 used to implement the example of the embodiment of the present invention is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device 200 can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown in the present invention, their connections and relationships, and their functions are only examples and are not intended to limit the implementation of the present invention described and / or required in the present invention.

[0105] The processor 201 is a computing unit, which may be various general and / or special processing components with processing and computing capabilities. Some examples of the processor 201 include, but are not limited to, a CPU (Central Processing Unit), a GPU (Graphic Processing Units), various dedicated AI (Artificial Intelligence) computing chips, various computing units running machine learning model algorithms, a DSP (Digital Signal Processor, Digital Signal Processor 201), and any appropriate processor 201, controller, microcontroller, etc. The computing unit performs the various methods and processes described above, such as an imaging method based on focal plane fitting. For example, in some embodiments, the imaging method based on focal plane fitting may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit. In some embodiments, part or all of the computer program may be loaded and / or installed on the device via a ROM and / or a communication unit. When the computer program is loaded into the RAM and executed by the computing unit, one or more steps of the method described above may be performed. Alternatively, in other embodiments, the processor 201 may be configured to execute the aforementioned method in any other appropriate manner (for example, by means of firmware).

[0106] In the context of the present invention, memory 202 may be a machine-readable medium, which may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or apparatus. A machine-readable medium may also be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a RAM, a ROM, an EPROM (Electrically Programmable Read-Only-Memory, Erasable Programmable Read-Only Memory 202) or a flash memory 202, an optical fiber, a CD-ROM (CompactDisc Read-Only Memory, Portable Compact Disk Read-Only Memory 202), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0107] The program code for implementing the method of the present invention can be written in any combination of one or more programming languages. These program codes can be provided to a processor 201 or controller of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that the program code, when executed by the processor 201 or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code can be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0108] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0109] 1. The present invention provides a fruit picking method based on a picking robot. According to the posture of the robot arm, a picking position where the chassis can perform path planning is obtained, and the posture of the robot arm during the picking process is added as a constraint to the chassis path planning algorithm, which is helpful to improve the efficiency of fruit picking. Specifically, it can complete the picking of fruits in the orchard with high efficiency without collision, saving energy consumption; a model for path planning at the optimal picking position is built, and then by establishing evaluation indicators, fruit picking based on the picking robot is realized, which is helpful to improve productivity, and can help farmers and scientific researchers better plan resources and liberate excess labor;

[0110] 2. A picking robot provided by the present invention obtains a picking position where the chassis can perform path planning according to the position of the robotic arm, and the posture of the robotic arm during the picking process is added as a constraint to the chassis path planning algorithm, which helps to improve the efficiency of fruit picking, specifically, it can complete the picking of fruits in the orchard with high efficiency without collision, saving energy consumption; a model for path planning at the optimal picking position is built, and then by establishing evaluation indicators, fruit picking based on the picking robot is realized, which helps to improve productivity, and can help farmers and scientific researchers better plan resources and liberate excess labor;

[0111] 3. An electronic device provided by the present invention obtains a picking position where the chassis can perform path planning based on the posture of the robotic arm, and the posture of the robotic arm during the picking process is added as a constraint to the chassis path planning algorithm, which helps to improve the efficiency of fruit picking. Specifically, it can complete the picking of fruits in the orchard with high efficiency without collision, saving energy. A model for path planning at the optimal picking position is built, and then evaluation indicators are established to achieve fruit picking based on a picking robot, which helps to improve productivity and can help farmers and scientific researchers better plan resources and liberate excess labor.

[0112] In the several embodiments provided by the present invention, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0113] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0114] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

Claims

1. A fruit picking method based on a picking robot, comprising S1, obtaining map information of an orchard, characterized in that: S2, establishing an orchard map and a spatial model of picking by the robotic arm (101) based on the map information and the depth information obtained by the depth camera (102), and obtaining a coordinate set of fruit distribution; S3, performing preliminary path planning covering the coordinate set according to the spatial model and the coordinate set and a preset path planning algorithm; S4, establishing a kinematic model of the robot arm (101); S5, determining the inverse kinematic solutions of different picking postures of the robot arm (101) according to the coordinate set and the kinematic model; S6, optimizing the preliminary path planning into a navigation path according to the inverse kinematic solution; wherein the method for optimizing the preliminary path planning into the navigation path is to use the inverse kinematic solution method to find a joint angle combination that minimizes the loss function value, obtain an optimal solution, and adjust the preliminary path planning into a navigation path according to the optimal solution as a constraint condition; S7, executing the navigation path by means of a mobile chassis (104) to achieve the picking of fruits, wherein the mobile chassis (104) is connected to the mechanical arm (101) and drives the mechanical arm (101) to move; S8, performing the navigation path evaluation on the last movement of the mobile chassis (104), adjusting the inverse kinematic solution according to the evaluation result, obtaining an adjusted inverse kinematic solution, and updating the navigation path according to the inverse kinematic solution, wherein the evaluation index includes the number of collisions.

2. A fruit picking method based on a picking robot according to claim 1, characterized in that: The map information is formed by the laser radar collecting information in the orchard in advance.

3. A fruit picking method based on a picking robot according to claim 2, characterized in that: The S3 further includes: S31, the spatial model is a collection of point cloud data, and a grid map for path planning is constructed by matching and calibrating the point cloud data and the coordinate set; S32, inputting the grid map into a preset path planning algorithm to complete the preliminary path planning covering the coordinate set.

4. A fruit picking method based on a picking robot according to any one of claims 1 to 3, characterized in that: In S3, the preset path planning algorithm is an ant colony algorithm.

5. The fruit picking method based on a picking robot according to claim 1, characterized in that: In S4, the kinematic model is determined according to the DH parameters of the robot arm (101).

6. The fruit picking method based on a picking robot according to claim 1, characterized in that: The S6 further includes: S61, determining a loss function according to preset multiple objectives to determine an optimal solution for the kinematic inverse solution; wherein the preset multiple objectives include collision avoidance and energy consumption; S62, taking the optimal solution as a constraint condition, adjusting the preliminary path planning into a navigation path.

7. A picking robot (100), characterized in that: include: A mechanical arm (101) for adjusting different picking postures; A depth camera (102), which is disposed on the mechanical arm (101) and moves together with the mechanical arm (101), and is used to obtain depth information; A picking execution mechanism (103) is disposed on the mechanical arm (101) and moves together with the mechanical arm (101) and is used to execute the action of picking fruits; A mobile chassis (104), connected to the mechanical arm (101), for executing a navigation path in an orchard; A controller, used for implementing the method according to any one of claims 1 to 6 by controlling the mechanical arm (101), the depth camera (102), the picking actuator (103) and the mobile chassis (104).

8. An electronic device (200), characterized in that: include: A processor (201), a memory (202), and a computer program stored in the memory (202) and executable on the processor (201), wherein the processor (201) implements the method according to any one of claims 1 to 6 when executing the program.

Images