A digital fruit picking intelligent drone

Abstract

The present invention provides a digital-based intelligent fruit-picking drone, which is aimed at many hilly areas with fruit trees. The terrain is relatively rugged, and the fruit trees are higher than the ground. The existing technology cannot complete the picking work and cannot fully protect the outer skin of the fruit. The present invention can realize the functions of automatic identification of fruits, grabbing, and cutting fruit stems; at the same time, the drone is equipped with 5G signal equipment and a digital twin system, which can monitor and predict fruit picking from a remote end, and ensure that the signal is good when it completes the picking work outside; the drone picking mechanical arm adopts visual image processing and pressure sensors, which can complete the drone's autonomous identification of the position and size of the fruit; the digital twin system predicts fruit picking and monitors the health status of the picking equipment in real time. If the equipment is damaged, it can return to the home in time for repair; the fruit picking basket is equipped with a gravity sensor, which can complete the task and return to the home by itself. Fruit picking drones can reduce the labor intensity of manpower and improve the efficiency of picking work.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural unmanned aerial vehicles, and in particular to a digital-based intelligent unmanned aerial vehicle for fruit picking. Background Art

[0002] Fruit picking is an agricultural activity carried out according to the period of fruit ripening. Manual fruit picking during the busy farming season is not only inefficient, but the fruit may also be picked in a hurry and the quality may not be guaranteed.

[0003] Developing and using picking robots to carry out picking activities can effectively increase the speed of harvesting, and the quality of the fruit can also have a unified standard, and the consumer experience is guaranteed. However, since many fruit trees are located in hilly areas and the terrain is relatively rugged, and the fruit trees are higher than the ground, the existing technology cannot complete the picking work. The existing fruit picking equipment mainly picks from the ground, relying on mechanical power transmission to clamp the fruit, and manually picking the fruit. The process not only causes great damage to the fruit and fails to protect the appearance of the fruit, but also due to the rugged terrain, most picking equipment cannot operate, and manual control is still required for picking, which has high labor costs and high labor intensity.

[0004] Most existing drone fruit picking equipment is designed only to solve the problem of not being able to pick fruits at high places, and few of them take into account the protection of the outer skin of the fruit. During the fruit picking process, it is still necessary to manually search for the fruit and operate the drone to pick the fruit, which cannot reduce the labor intensity. Summary of the invention

[0005] The purpose of the present invention is to provide a fruit picking drone device that can automatically search for and pick fruits with rapid response and almost no damage to fruits, in order to address the problems of low efficiency of manual fruit picking, easy damage to the fruit skin, and low degree of digitization and intelligence of drones.

[0006] To achieve the above object, the technical solution adopted by the present invention is:

[0007] A digital-based intelligent drone equipment for fruit picking, comprising a four-wing drone body, a universal device and a power transmission assembly below the drone body, a retractable four-axis robotic arm mounted on both sides of the universal device, the end of the robotic arm consisting of an end effector capable of automatic identification and grasping, and a detachable fruit picking basket equipped below the universal device.

[0008] The four-wing UAV body is equipped with a remote sensing device, a GPS positioning device, a central controller, an equipment detection chip and a small 5G signal receiving device.

[0009] The remote sensing device and the GPS positioning device can enable the drone to quickly and accurately reach the fruit picking location without the need for manual transportation and remote control under the fruit trees.

[0010] The equipment detection chip of the present invention adopts a digital twin system, which can perform behavior prediction and simulation. During the fruit picking process, it can provide the optimal fruit picking angle and orientation by calculating and simulating the grasping action, and reasonably plan the picking path, so as to protect the fruit to the greatest extent; at the same time, the health status of the drone equipment and the robotic arm equipment is detected to avoid damage to the equipment during the picking process. According to the types, sizes and states of different fruits, the digital twin system will conduct a picking evaluation on the fruits, and compare them with the actual fruits after picking to ensure that there are no rotten fruits.

[0011] The small 5G signal receiving device is built into the main body of the drone. Since a large number of fruit trees will shield the signal during the fruit picking process, the 5G signal receiving device can ensure that the drone receives stable signals and can provide feedback on the real-time position and picking situation.

[0012] The universal device is mounted directly below the drone and cooperates with the drone body by means of a circular guide rail. It can rotate to ensure that the mechanical arm can perform multi-directional picking operations. It has a built-in power transmission assembly and a power assembly to provide power to the mechanical arms mounted on both sides of the universal device.

[0013] The power transmission assembly includes a universal joint power transmission device and a mechanical arm power transmission device.

[0014] The mechanical arm power transmission device is built into the universal device, and transmits power to the mechanical arm according to instructions to complete the grasping action.

[0015] The four-axis robot arm can complete XYZ axis translation, X axis rotation and end effector grasping. It cooperates with the universal device through a circular guide rail to enable it to complete X axis rotation. The middle section of the robot arm is connected by a special-shaped plate. The two sides of the special-shaped plate are connected with hydraulic rods and robot arm rods. The hydraulic rods and robot arm rods on both sides are connected to the special-shaped plate by shafts to ensure that they can perform rotation operations. The end of the hydraulic rod is connected to the end effector.

[0016] The end effector adopts a pressure sensor in combination with visual image processing, and can complete the autonomous identification of the position and size of the fruit by the picking end effector, and detect the contact with the fruit through the pressure sensor on the picking finger, and the cutting is completed by the automatic cutting device at the top.

[0017] The end effector mainly consists of two parts: a control part and a mechanical working part.

[0018] The control part is composed of a motor controller, a central control panel and a switching power supply.

[0019] The control part is mainly responsible for controlling the signal of the single-chip microcomputer. When the central control panel receives the transmission signal from the single-chip microcomputer, it can respond quickly. The central control panel is connected to the battery pack and the motor controller to realize the forward and reverse rotation of the motor, thereby completing the grabbing and releasing actions of the picking fingers.

[0020] The control part is combined with the power assembly in the universal device. The single chip transmits the signal to the central controller and at the same time transmits the signal to the central controller in the drone. The central controller in the drone transmits the signal to the power assembly in the universal transmission device according to the signal, and the motor controls the mechanical arm to adjust the orientation.

[0021] The mechanical working part is composed of a motor, a cutting device, a push rod and picking fingers.

[0022] The mechanical working part is mainly the power output and transmission of the end effector. The torque generated by the motor is transmitted to the picking fingers through the push rod, etc. According to the instructions of the control part, the motor is controlled to perform forward and reverse rotation and the cutting device cutting movement.

[0023] The end effector is mainly composed of a picking finger, a push rod, a clamping ring, a connecting rod, a frame, a pin, a motor driver, a pressure sensor, a visual sensor, a single-chip computer, etc.

[0024] The picking fingers, push rods, clamping rings, connecting rods, racks, etc. are controlled by motors to perform fruit picking and cutting actions. The picking fingers protect the fruit skin from being damaged by their arc-shaped appearance.

[0025] The visual sensor is installed on the frame, performs visual recognition of the fruit, and transmits the signal to the single-chip microcomputer. The single-chip microcomputer recognizes the signal and transmits it to the central controller in the drone and the central controller in the end effector. The central controller in the drone will search for the position and control the robotic arm. The central controller in the end effector will transmit the picking finger closing signal to the motor driver to control the picking finger closing.

[0026] The pressure sensor is installed inside the picking finger, and the picking finger completes the closing action. When the picking finger contacts the fruit, the pressure sensor receives pressure that reaches a threshold value, and sends a signal to the single-chip microcomputer. After the single-chip microcomputer recognizes the signal, it sends a stop command to the motor driver, and the cutting device at the upper end of the picking finger cuts the fruit stem part of the fruit. After the picking finger is completed, it slowly opens to complete the picking process.

[0027] These are visual sensors and pressure sensors. The visual sensor identifies the size, type and position of the fruit and transmits the signal to the microcontroller. The pressure sensor adjusts the threshold accordingly according to the different types and sizes of the fruit. At the same time, the digital twin system will plan the optimal picking path according to the different types and sizes of the fruit.

[0028] The single chip microcomputer identifies the signal transmitted by the pressure sensor when it reaches a threshold value and the image signal transmitted by the visual recognition sensor, and controls the motor driver to complete related actions.

[0029] The fruit picking basket is connected to the bottom end of the universal transmission device through a detachable hook and a hanging rope, and a gravity sensor is installed at the connection position between the hanging rope and the universal device. After the mechanical arm completes the fruit picking work, the single-chip microcomputer transmits a signal to the central controller in the drone, which issues a command to the mechanical arm to place the picked fruits in the picking basket below. When the fruits in the fruit basket reach the threshold of the gravity sensor, the gravity sensor issues a command to the central controller of the drone, and the drone returns.

[0030] The innovation of the digital fruit picking intelligent drone equipment is that it responds quickly, does not damage the outer skin of the fruit, can autonomously identify the fruit and complete the picking autonomously, which is different from some road fruit picking equipment on the market and fruit picking drones that require manual control.

[0031] The digital fruit picking intelligent drone equipment described above will cause damage to the connection between the fruit stem and the skin when the fruit stem is separated. The end effector has a blade cutting structure at the front end of the picking finger to cut the fruit stem and protect the fruit skin from damage.

[0032] The digital fruit picking intelligent drone equipment has four picking fingers on its end effector whose curvature conforms to the shape of the fruit, can wrap the fruit well, and can adjust the picking finger angles according to the different fruit sizes and shapes.

[0033] The digital fruit picking intelligent drone equipment can make picking predictions based on different fruits and adjust the threshold value according to the type and size of the fruit to ensure that the skins of multiple fruits can be protected during picking.

[0034] The digital fruit picking intelligent drone equipment has mechanical arms on both sides that can complete the picking work independently, thereby improving the picking efficiency.

[0035] The digital fruit picking intelligent drone equipment can perform real-time signal communication to ensure that the remote operator can grasp the position and status of the drone in real time. The digital twin system can monitor the health status of the drone equipment in real time and simulate the planning of the picking path.

[0036] The digital fruit picking intelligent drone can improve the fruit picking efficiency and ensure that the fruit skin is protected from damage during the picking process.

[0037] The digital fruit picking intelligent drone can perform remote monitoring through a digital twin system to detect and control the fruit type and quality, picking quantity and picking efficiency.

[0038] The digital fruit picking intelligent drone can set the required weight of picked fruits in advance. When the weight in the fruit basket reaches the required weight, the gravity sensor will send a return command to the drone central controller, thereby realizing full-process automated picking.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] The present invention provides a fruit picking drone device that can automatically search for and pick fruits without damaging the fruits, and can realize functions such as automatic fruit identification, grabbing, and cutting fruit stems. At the same time, the drone is equipped with 5G signal equipment and a digital twin system, which can monitor and predict fruit picking from a remote end, and ensure that the signal is good when completing the picking work outside. The drone picking mechanical arm uses visual image processing and pressure sensors to enable the drone to autonomously identify the position and size of the fruit. The digital twin system predicts fruit picking and monitors the health status of the picking equipment in real time. If the equipment is damaged, it can return to the home in time for repair. The gravity sensor can measure the weight of the fruit in the fruit picking basket, and when the required weight is reached, the drone can automatically return. Digital fruit picking intelligent drones can reduce labor intensity and improve picking efficiency.

[0041] The present invention makes full use of the real-time sensing of the pressure sensor and the recognition diversity of the visual sensor. It can autonomously control the strength of the picking fingers according to the perception of the pressure sensor to ensure the protection of the fruit skin during the fruit picking process; the picking fingers can be adjusted according to the pressure data and image recognition data of different fruits to complete the picking of different fruits; the digital twin system can detect the health status of each device of the drone after completing the picking to determine whether it needs to return; the drone has a built-in 5G signal receiving device, a remote sensing device and a GPS positioning device to ensure that the drone can accurately reach the picking location, and the remote operator can obtain the real-time location of the drone; the gravity sensor can ensure the weight requirement of the fruit picking, thereby completing the fully automatic fruit picking process. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 Schematic diagram of intelligent drone for digital fruit picking;

[0043] Figure 2Schematic diagram of the end effector of the robot arm of the intelligent drone for digital fruit picking;

[0044] Figure 3 A bird’s-eye view of a smart drone for digital fruit picking;

[0045] Figure 4 Schematic diagram of the intelligent drone robotic arm for digital fruit picking;

[0046] Figure 5 Schematic diagram of the connection of the special-shaped plate of the robotic arm. DETAILED DESCRIPTION

[0047] The details of the present invention and its specific implementation methods are further described below in conjunction with the accompanying drawings.

[0048] Combination Figure 1 , 2 , 3, 4, 5. The present invention designs a fruit picking drone device which has a rapid response, almost no damage to the fruit, can search for the fruit by itself and complete the picking.

[0049] like Figure 1 The figure shows that it is mainly composed of 13 a four-wing UAV body, 14 a universal transmission device, 15 a four-axis mechanical arm, 16 a fruit picking basket, and 17 a hanging rope; the universal transmission device 14 is connected to the UAV body 13 through a top circular guide rail, and ensures that the universal transmission device and the UAV can rotate freely, and the fruit picking basket 16 is connected to the hanging rope through hooks around it.

[0050] like Figure 2 , Figure 4 and Figure 5 As shown, the four-axis robot arm 15 is mainly composed of an end effector 18, a hydraulic rod 19, a robot arm connecting rod 20, a robot arm special-shaped connecting plate 21, and a rotatable base 22; the end effector is mainly composed of a picking finger 1, a push rod 7, a clamp ring 10, a connecting rod 3, a rack 4, a pin, a motor driver 9, a pressure sensor 2, a visual sensor 11, and an 8-chip microcomputer; the robot arm special-shaped connecting plate 21, the robot arm connecting rod 20 and the hydraulic rod 19 are connected on both sides of the robot arm special-shaped connecting plate 21 through an axis, and the lower hydraulic rod and the robot arm rod are fixed on the fixed special-shaped plate by screws, and the end effector 18 is connected with the hydraulic rod and the robot arm rod by the axis to ensure that the axis and the rod and the axis hole are in clearance fit and can rotate freely; the robot arm and the universal transmission device are directly connected by the guide rail to ensure that the robot arm and the universal transmission device can rotate freely; there is a circuit assembly in the universal transmission device, and the universal transmission device and the robot arm are powered by the internal battery of the drone.

[0051] The specific working process of the digital fruit picking intelligent drone is as follows:

[0052] The drone is operated by a remote operator and reaches the required picking location according to the built-in GPS system and remote sensing system settings of the drone. The 11 visual sensors identify the fruit and transmit the signal to the 8 single-chip microcomputers. The 8 single-chip microcomputers transmit the instructions and requirements to the end effector controller and the drone central controller respectively according to the signal. The drone central controller judges the distance, position and direction according to the signal and controls the drone to adjust its posture. The digital twin system simulates the picking work according to the signal transmitted by the visual sensor, transmits the simulation result to the drone central controller, and adjusts the drone posture again according to the signal. After the adjustment, the simulated path signal is transmitted to the 8 single-chip microcomputer in the end effector. The single-chip microcomputer issues a command to the end effector according to the signal to control the 6 motors to start rotating. After transmission, the 1 picking finger moves inward to clamp the fruit. The 2 pressure sensor detects whether the contact force between the picking finger and the fruit reaches the pressure threshold. If the threshold is reached, a signal is sent to the 8 single-chip microcomputer to control the 6 motors to stop running, thereby achieving 1 The picking fingers clamp the fruit, and the cutting device 12 starts to run. After the cutting is completed, the 8 single-chip computer transmits the signal to the drone central controller to control the 15 mechanical arm to rotate and move the 15 mechanical arm to the fruit picking basket 16 below. When the 15 mechanical arm reaches the specified position, the 6 motor rotates, the 1 picking finger is released, and the fruit is sent to the 16 fruit picking basket to complete a picking. After the picking is completed, the digital twin system detects this behavior and the health status of the entire equipment, and detects the blade wear, drone wing wear and other components. If return is required, the demand will be submitted to the remote operator, and the operator will determine whether return is required based on the actual situation. After a certain amount of fruit is picked, the gravity sensor above the fruit picking basket determines whether the weight in the fruit basket reaches the threshold. When the threshold is reached, the gravity sensor sends a command to the drone central controller and returns.

[0053] Since the drone is picking fruits in a dense orchard, it has an automatic obstacle avoidance function, which allows it to avoid obstacles such as leaves and branches, thus ensuring the safety of the drone itself. After completing the picking work, the drone records the fruit picking, and the digital twin system monitors the health of the drone in real time to ensure that it can continue to complete the task. The visual sensor can identify different fruit types, sizes and positions, and transmit the signal. The pressure sensor on the inside of the picking finger can select and set the threshold according to the different fruit types and sizes identified by the visual sensor, ensuring that different fruit types and sizes can wrap the fruit while leaving the outer skin undamaged. The digital twin system combines visual sensors and pressure sensors to evaluate the quality of the fruit, ensure the freshness of the fruit when it is picked, and return the real-time data of fruit picking to the remote operator, thus ensuring the quality of the picked fruit. The fruit in the fruit picking basket can be observed in real time, and the fruit picking basket rope is connected to the universal A camera can be installed at the connection of the transmission device to monitor the fruits in the fruit picking basket and transmit the real-time image to the operation end; since the connection between the fruit stalk and the epidermis will be damaged when the fruit stalk is detached, a blade cutting device is installed at the top of the end effector of the drone's mechanical arm, and the fruit stalk is cut by the blade to protect the fruit epidermis from damage; the curvature of its four picking fingers conforms to the shape of the fruit, and the fruit can be well wrapped, and the angle of the picking fingers can be adjusted according to different fruit sizes and shapes; the built-in 5G signal receiving device, remote sensing device and GPS positioning device can ensure that the drone can accurately reach the picking location, and the remote operator can obtain the real-time location of the drone; it is suitable for areas such as hilly and mountainous areas where conventional picking equipment cannot be used, with a wide range of use, and at the same time solves the problem that fruits of taller fruit trees are difficult to pick; the present invention does not require the operator to perform on-site operations. Different from the market where operators need to control the drone at close range, the present invention can be operated from a remote end, which reduces labor costs and improves picking efficiency.

Claims

1. A digital fruit picking intelligent drone, characterized by: It includes a four-wing UAV body, a universal transmission device and a power transmission assembly under the UAV body, a retractable four-axis robotic arm mounted on both sides of the universal transmission device, the end of the four-axis robotic arm is composed of an end effector that can perform automatic identification and grasping, and a detachable fruit picking basket is equipped under the universal transmission device. The four-axis robotic arm is composed of an end effector, a hydraulic rod, a robotic arm connecting rod, a robotic arm special-shaped connecting plate and a rotatable base; the end effector is composed of a picking finger, a push rod, a clamping ring, a connecting rod, a frame, a pin, a motor driver, a pressure sensor, a visual sensor and a single-chip microcomputer, and the four-axis robotic arm can perform multi-degree-of-freedom free rotation, including XYZ axis translation and X axis rotation; a robotic arm special-shaped connecting plate, a robotic arm connecting rod and a hydraulic rod are connected to both sides of the robotic arm special-shaped connecting plate through an axis, and the lower hydraulic rod and the robotic arm rod are fixed to the robotic arm special-shaped connecting plate by screws. On the connecting plate, the end effector is connected to the hydraulic rod and the mechanical arm rod by means of an axis to ensure that the axis, the rod and the axis hole are clearance-fitted and can rotate freely; a visual sensor and a pressure sensor are arranged inside the end effector, and the visual sensor senses the type, size and position of the fruit, judges the fruit through data analysis, and simulates and plans the fruit picking path by using the digital twin system. An optimization plan for the fruit picking path is proposed based on the simulation results, and the drone picks the fruit according to the optimization plan. The pressure sensor is used to sense the pressure of the fruit to ensure that the fruit skin is protected to the greatest extent while wrapping the fruit, and the fruit stalk is cut by a cutting blade, and the fruit is placed in a fruit picking basket to complete the picking. The fruit picking basket senses the weight of the picked fruit through a gravity sensor. When the required weight is reached, the drone automatically returns.

2. The digital fruit picking intelligent drone according to claim 1, characterized in that: The four-axis robot arm and the universal transmission device are connected by a guide rail to ensure that the robot arm and the universal transmission device can rotate freely; the universal transmission device is equipped with a circuit assembly, and the universal transmission device and the robot arm are powered by the internal battery of the drone. A hanging rope is installed under the universal transmission device, and the fruit picking basket is connected to the hanging rope by a hook, and a gravity sensor is installed at the connection between the hanging rope and the universal transmission device.

3. The digital fruit picking intelligent drone according to claim 2 is characterized in that: The four picking fingers of the end effector are fixed to the frame and connected to the motor driver by a connecting rod. Pressure sensors are evenly distributed on the inner sides of the four picking fingers. The visual sensor is fixed to the frame. The type, size and position of the fruit are identified by the visual sensor. The single-chip microcomputer sends instructions to the central controller by receiving the sensor signal. The digital twin system simulates the virtual picking path according to the data sent by the visual sensor, proposes the optimal picking path, and sends instructions to the robotic arm and the end effector according to the optimal picking path. At the same time, the drone adjusts its own posture. The robotic arm and the end effector pick the fruit according to the instructions. Pressure sensors are evenly distributed on the inner sides of the picking fingers in the end effector. According to the type and size of the fruit, when the pressure reaches the threshold, the picking fingers stop moving, and the cutting device at the top of the end effector cuts the fruit stalk to complete the fruit picking process. After cutting, the single-chip microcomputer transmits the signal to the central controller, and according to the instructions of the central controller, the four-axis robotic arm places the fruit in the fruit picking basket.

4. The digital fruit picking intelligent drone according to claim 3 is characterized in that: The gravity sensor is arranged at the connection between the universal transmission device and the hanging rope.

5. The digital fruit picking intelligent drone according to claim 4 is characterized in that: The power transmission assembly includes a universal joint power transmission device and a mechanical arm power transmission device.

6. The digital fruit picking intelligent drone according to claim 5, characterized in that: The four-wing UAV body is equipped with a remote sensing device, a GPS positioning device, a central controller, an equipment detection chip and a small 5G signal receiving device.

7. The digital fruit picking intelligent drone according to claim 6, characterized in that: The remote sensing device and GPS positioning device can enable the drone to quickly and accurately reach the fruit picking location without the need for manual transportation and remote control under the fruit trees; The equipment detection chip adopts a digital twin system to predict and simulate behavior. During the fruit picking process, it can provide the optimal fruit picking angle and orientation by calculating and simulating the grabbing action, and reasonably plan the picking path, so as to protect the fruit to the greatest extent. At the same time, the health status of the drone equipment and the robotic arm equipment is checked to avoid damage to the equipment during the picking process. According to the types, sizes and states of different fruits, the digital twin system will evaluate the picking of the fruits and compare them with the actual picked fruits to ensure that there are no rotten fruits. The small 5G signal receiving device is built into the main body of the drone. Since a large number of fruit trees will shield the signal during the fruit picking process, the 5G signal receiving device ensures that the drone receives stable signals and provides feedback on the real-time position and picking situation.

Images