Crawler-type hybrid power intelligent orchard operation robot and method thereof

Abstract

The invention discloses a crawler-type hybrid power intelligent orchard operation robot and a method thereof. The operation robot comprises a walking system, a power system, an operation system and a control system, a crawler wheel train is sequentially connected with a driving motor and a driving motor controller, an energy storage unit and an engine jointly or independently output power to the walking system according to an energy strategy, and a whole vehicle controller is responsible for coordinated control over the whole vehicle energy flow direction. And the intelligent navigation controller is responsible for receiving a navigation signal, planning a track and calculating and outputting a motion control quantity, and automatically spraying liquid medicine to the orchard through the operation system in the automatic driving process of the operation robot. The crawler-type hybrid power intelligent orchard operation robot fills the blank that no crawler-type hybrid power intelligent orchard operation robot exists in the current market, the single-time operation endurance time of the operation robot is prolonged, people and pesticide are separated in the orchard pesticide spraying process, the intelligent function is easier to achieve, and the crawler-type hybrid power intelligent orchard operation robot conforms to the development strategy of high-end agricultural machinery and intelligent manufacturing.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural machinery, and in particular to a crawler-type hybrid power intelligent orchard operation robot and a method thereof. Background Art

[0002] With the continuous improvement of social science and technology, agricultural production is gradually developing in the direction of new energy and intelligence. Under this new situation, the orchard operation machinery currently on the market faces a huge test. Orchard operation machinery usually uses diesel as the main energy source, which has high fuel consumption and serious emissions, which is contrary to the development direction of green and energy-saving agricultural machinery; and the machinery for spraying orchards is usually driven or remotely controlled, which makes it difficult to separate people and medicines, bringing potential health damage to agricultural workers. Pure electric crawler operation robots have short single-operation endurance due to low battery energy density and the lack of charging equipment in farming areas, which limits their application scenarios and scope.

[0003] In response to the above problems, we designed a tracked hybrid intelligent orchard operation robot and its method to solve the above problems. Summary of the invention

[0004] The purpose of the present invention is to solve the problem in the prior art that it is difficult to separate people and medicines when using machinery to spray orchards, which brings potential health damage to agricultural workers. Pure electric crawler operating robots, on the one hand, have short single-operation endurance, and on the other hand, are difficult to charge, which limits their application scenarios and scope. The present invention proposes a crawler hybrid intelligent orchard operating robot and a method thereof, which adopts hybrid power drive to improve the single-operation endurance mileage, optimizes the structure of the operating robot, and performs intelligent driving operations, and provides multiple navigation solutions and good scene adaptability.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions: A crawler-type hybrid intelligent orchard operation robot, comprising: The walking system includes crawler wheel systems located on the left and right sides of the frame, and a motor, a reducer and a motor driver that establish a power transmission connection with each crawler wheel system; A power system, including an energy storage unit, an engine and a generator, wherein the energy storage unit and the engine jointly or individually transmit power to the travel system and the working system; The operating system includes a liquid medicine tank, a hydraulic motor, a water pump, a switch valve and a fan. The hydraulic motor is driven and controlled by the engine through a clutch. The hydraulic motor drives the water pump to pressurize and regulate the liquid medicine in the liquid medicine tank. The fan is powered by an energy storage unit and driven and controlled by an agricultural implement controller to blow the liquid medicine sprayed through the switch valve into the orchard. The control system includes an intelligent navigation controller, a vehicle controller, a generator controller and an implement controller. The intelligent navigation controller includes a satellite positioning device, a camera and a laser radar, and controls the travel path of the travel system according to the positioning and planned driving trajectory; the generator controller is connected to the generator, and the implement controller is connected to and controls the operation of the implements of the operating system; the vehicle controller is connected to the travel system, the power system and the implement controller of the operating system.

[0006] Further preferably, the track wheel system in the walking system includes a driving wheel, a road wheel, a tensioner wheel and a towing wheel, and the driving wheel, the road wheel, the tensioner wheel and the towing wheel are arranged on the track in sequence from front to back, and the driving wheel establishes a power transmission connection with the reducer, the power output end of the motor is connected to the reducer, the motor driver establishes a control connection with the motor, and the motor driver receives control instructions from the vehicle controller of the control system.

[0007] Further preferably, the energy storage unit in the power system includes a fast-charging and discharging power battery and a supercapacitor, the supercapacitor and the power battery are connected in parallel on a DC bus, and the DC bus and the motor driver provide electrical energy to the motor driver to drive the motor to operate, the vehicle controller is connected to the power battery through a battery management system of the power battery, and is connected to the engine through an engine controller of the engine, and an electronic control unit is used for regulation and control between the engine and the generator.

[0008] Further preferably, the liquid medicine tank includes at least two detachable liquid medicine storage units, each of which has a different capacity, and the liquid medicine storage units are used in combination according to the operation requirements to adjust the total capacity of the liquid medicine tank, and the liquid medicine storage units are equipped with liquid level sensors to monitor the amount of liquid medicine in each liquid medicine storage unit in real time and feed the data back to the control system. The control system adjusts the combination of the liquid medicine storage units according to the liquid medicine amount data and the operation plan; the liquid medicine flow sensor and the pressure sensor are installed on the switch valve, the hydraulic motor drives the water pump to extract the liquid medicine in the liquid medicine tank, and sprays it through the switch valve, and the fan is arranged next to the switch valve to blow the liquid medicine sprayed by the switch valve into the orchard.

[0009] Further preferably, the intelligent navigation controller of the control system has a built-in environmental perception module to detect obstacles on the working path, automatically avoid obstacles and plan the path.

[0010] Further preferably, the power system includes a whole-machine power monitoring unit and a feedback unit to monitor and adjust the output power of the engine and the energy storage unit in real time; an energy distribution module is provided between the energy storage unit and the engine to analyze the energy requirements of the power battery and the engine in real time.

[0011] Further preferably, the power system adopts pure electric mode and hybrid mode to provide power energy to the working robot; during the pure electric mode operation, the power battery outputs power to the motor, and the motor transmits power to the track wheel system under the coordination of the motor driver and the reducer, driving the entire track wheel system to operate and push the working robot forward; during the hybrid mode operation, the power battery and the engine serve as the power source of the working robot, and jointly output power to the walking system and the working system, the engine drives the generator to generate electricity and the current generated by the power battery outputs power to the motor together, forming a hybrid power, driving the entire track wheel system to push the working robot forward.

[0012] An operating method of a crawler hybrid intelligent orchard operating robot, the operating method comprising the following contents: Trajectory planning: The operation robot is planned to travel by manual trajectory planning or intelligent trajectory planning. Manual trajectory planning includes: pre-marking in the operation area with a handheld dotting device to plan the operation robot's travel trajectory; intelligent trajectory planning includes collecting images of fruit trees between rows based on the principle of visual navigation, marking the points where the images of fruit trees between rows are located, and planning the operation robot's travel trajectory; Task execution: The operating robot moves along the planned driving trajectory. During the movement, according to the set work tasks, it performs plant protection operations on the driving trajectory, uses the fan to perform air atomization, and blows the liquid sprayed by the switch valve into the orchard.

[0013] Further preferably, in the manual trajectory planning process, a handheld dotting device pre-dots in the working area, and planning the driving trajectory of the working robot includes the following steps: a) Step, dotting and recording: The staff uses a handheld dotting device to dot in the work area and record the coordinates of each key point; b) Step, trajectory planning: record the coordinate data of key points through a handheld dotting device, and form the driving trajectory of the working robot in the intelligent navigation controller; c) Step, moving navigation: The working robot receives real-time position information through a satellite positioning device, and navigates according to the driving trajectory formed by the coordinates of key points recorded by a handheld dotting device.

[0014] Further preferably, the intelligent trajectory planning includes collecting images of fruit trees between rows based on the principle of visual navigation, marking the points of the images of fruit trees between rows, and planning the driving trajectory of the working robot, including the following steps: 1) Step, data collection and training: Relying on the camera carried by the operating robot, collect orchard image data between rows in different orchards, different weather and different lighting environments; 2) Step, data annotation and model training: Based on the principle of visual navigation, the collected orchard images between rows are annotated. During the annotation process, the annotation lines of the left and right tree rows are defined and input into the deep learning model for learning and training; 3) Step, obtain the navigation line: Run the deep learning model in the intelligent navigation controller to obtain the training results of the fitted straight line along the two tree rows, select two parallel lines at different height layers of the image, the two intersection points of the first parallel line and the fitted straight line are recorded as p1 (x1, y1), p2 (x2, y2), the two intersection points of the second parallel line and the fitted straight line are recorded as p3 (x3, y3), p4 (x4, y4), find the intersection points p_s and p_e of p1, p2 and p3, p4, and the line connecting the two intersection points is the navigation line obtained by the plane image; 4) Step, navigation line conversion: project the navigation line onto the real working robot for coordinate conversion, convert the pixel coordinates of the navigation line into the driving trajectory of the working robot in the metric coordinate system, and the working robot moves according to the driving trajectory in the metric coordinate system; Step 5: Navigation in the U-turn area: When it is detected that the length of the fitted straight line between two rows of trees gradually decreases to a certain set threshold, the working robot is considered to be close to the end of the row, and the camera on the working robot is started to identify the fruit trees on both sides. If it is judged that the fruit trees no longer appear in the camera within a certain time threshold, it is considered that the working robot has arrived at the U-turn area and starts automatic navigation in the U-turn area; Step 6), re-enter the line for straight segment navigation: After the U-turn is completed, the working robot automatically returns to the navigation line of the straight segment, and the working robot repeats the driving trajectory of steps 4) to 5).

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention provides energy for the walking system and the operating system separately or jointly through the engine and the energy storage unit, optimizes the energy supply mode of the operating robot, flexibly adjusts the energy supply mode according to the usage scenario, improves the single operation endurance of the operating robot, solves the problem of charging difficulty of the operating robot, and expands its application scenarios and application scope; the operating robot is planned to drive its driving trajectory by adopting manual trajectory planning or intelligent trajectory planning, so that the operating robot can automatically drive according to the driving trajectory, and the separation of people and drugs during the orchard spraying process is achieved, avoiding potential health damage to agricultural workers; the present invention fills the gap of the current lack of crawler hybrid intelligent orchard operating robots on the market. Compared with the existing remote-controlled diesel crawler sprayer, it greatly reduces the use cost and fuel consumption, makes the intelligent function easier to realize, and conforms to the development strategy of high-end agricultural machinery and intelligent manufacturing. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1A system composition block diagram of a crawler-type hybrid intelligent orchard operation robot provided by an embodiment of the present invention is shown; Figure 2 A schematic diagram showing the structural arrangement of a crawler-type hybrid intelligent orchard operation robot provided by an embodiment of the present invention is shown; Figure 3 A schematic diagram of energy flow in a pure electric mode of a crawler-type hybrid intelligent orchard operation robot provided by an embodiment of the present invention is shown; Figure 4 A schematic diagram of energy flow in a hybrid mode of a crawler-type hybrid intelligent orchard operation robot provided by an embodiment of the present invention is shown; Figure 5 A schematic diagram of intelligent navigation of a crawler-type hybrid intelligent orchard operation robot based on a satellite positioning device (handheld dotting device) provided by an embodiment of the present invention is shown; Figure 6 A schematic diagram showing the determination of an intelligent navigation marking line based on a visual sensor of a crawler-type hybrid intelligent orchard operation robot provided by an embodiment of the present invention is shown; Figure 7 A schematic diagram showing the determination of a navigation line in intelligent navigation based on a visual sensor of a crawler-type hybrid intelligent orchard working robot provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0018] Figure 1 The system composition block diagram of a crawler hybrid intelligent orchard operation robot according to an embodiment of the present application is schematically shown. The crawler hybrid intelligent orchard operation robot comprises a walking system, a power system, an operation system and a control system.

[0019] The walking system includes two track wheel systems located on opposite sides of the frame, a reducer, a motor and a motor driver that establish power transmission connections with each track wheel system in turn. The track wheel system is made of high-strength wear-resistant materials to meet the needs of orchard operations on different terrains. The motor uses a DC brushless motor to reduce maintenance requirements and improve power transmission efficiency. The track wheel system includes a driving wheel, a road wheel, a tensioner and a traction wheel. The driving wheel is responsible for transmitting power, the road wheel is only responsible for bearing weight and does not provide any power, the tensioner is responsible for tightening the track and maintaining a certain shape, and the traction wheel is responsible for dragging the track to prevent it from sagging. The driving wheel, the road wheel, the tensioner and the traction wheel are arranged on the track from front to back in sequence, and the driving wheel is connected to the reducer for power transmission, the power output end of the motor is connected to the reducer, the motor driver is connected to the motor, and the motor driver receives the control instructions of the vehicle controller of the control system.

[0020] The power system includes an energy storage unit (including power batteries and supercapacitors) and directly connected engines and generators. The energy storage unit and the engine output power to the travel system and the operating system together or individually according to the energy strategy. The energy storage unit in the power system includes fast-charging and discharging power batteries and supercapacitors to provide efficient energy conversion and storage. The supercapacitor and the power battery are connected in parallel on the DC bus, and the DC bus and the motor driver provide electrical energy to the motor driver to drive the motor to run. The vehicle controller is connected to the power battery through the battery management system of the power battery, and is connected to the engine through the engine controller of the engine. The electronic control unit is used to adjust and control the engine and the generator to optimize energy efficiency.

[0021] The power system includes a whole-machine power monitoring unit and a feedback unit, which monitor and adjust the output power of the engine and energy storage unit in real time to cope with dynamic workloads and optimize overall energy efficiency. An energy distribution module is installed between the energy storage unit and the engine to analyze the energy requirements of the power battery and engine in real time and optimize energy distribution to improve the overall efficiency of the system.

[0022] The operating system includes a liquid medicine tank, a hydraulic motor, a water pump, a switch valve, a fan and sensors related to spraying control. The hydraulic motor is driven by the engine through a clutch. The hydraulic motor drives the water pump to pressurize and regulate the liquid medicine in the liquid medicine tank. The fan is powered by an energy storage unit and driven and controlled by the farm implement controller, so as to blow the liquid medicine sprayed by the switch valve into the orchard, achieving the effect of uniform atomization and dispersion.

[0023] It is worth mentioning that the liquid medicine tank has an adjustable capacity design to meet the operational needs of orchards of different sizes. The liquid medicine tank includes at least two detachable liquid medicine storage units, each of which has a different capacity, and the liquid medicine storage units are used in combination according to the operational needs to adjust the total capacity of the liquid medicine tank. The liquid medicine storage unit is equipped with a liquid level sensor to monitor the amount of liquid medicine in each liquid medicine storage unit in real time and feed the data back to the control system. The control system adjusts the combination of liquid medicine storage units according to the liquid medicine volume data and the operation plan; the liquid medicine flow sensor and pressure sensor are installed on the switch valve to achieve precise control of the liquid medicine spraying efficiency. The hydraulic motor drives the water pump to extract the liquid medicine in the liquid medicine tank and spray it through the switch valve. The fan is set next to the switch valve to blow the liquid medicine sprayed from the switch valve into the orchard. In addition to spraying the liquid medicine through the switch valve, an adjustable nozzle can also be set to adjust the spraying mode according to the height and shape of different fruit trees. At this time, the fan is still set within the spraying range of the adjustable nozzle.

[0024] The control system includes an intelligent navigation controller, a vehicle controller, a generator controller and an agricultural implement controller.

[0025] Among them, the intelligent navigation controller includes a satellite positioning device (GPS / Beidou satellite positioning), a camera and a lidar. The intelligent navigation controller can receive one or more source data from a satellite positioning device, a camera, a lidar, etc., and perform single or fusion positioning, so as to calculate the control quantity according to the positioning and planning for output, form a driving trajectory, and control the walking path of the walking system.

[0026] The control system has remote monitoring and operation functions, allowing operators to conduct real-time control and monitoring through mobile devices.

[0027] The control system's intelligent navigation controller has a built-in environmental perception module that detects obstacles on the operating path, automatically avoids obstacles and plans the path; it can also detect the location and density of fruit trees, and automatically adjust operating parameters according to environmental changes in the orchard to improve operating efficiency and safety.

[0028] The generator controller is connected to the generator, specifically to the generator and the power battery and the supercapacitor.

[0029] The implement controller connects and controls the operation of the implements of the operating system, specifically controls the operation of the water pump, switch valve and fan in the operating system.

[0030] The vehicle controller is connected to the agricultural implement controllers of the travel system, power system and operation system, specifically the engine, motor driver and generator controller, and is configured to coordinate the charging and discharging of the power battery (or supercapacitor) and the power output of the engine, and is configured to actively output the liquid spraying requirements to the agricultural implement controller, thereby controlling the liquid spraying amount. The vehicle controller has a fault self-diagnosis function, which can automatically sound and light alarms when abnormalities are detected and take protective measures such as parking and shutting down.

[0031] It can be understood that the vehicle controller is connected to the power battery through the BMS (Battery Management System) of the power battery, and the vehicle controller is connected to the engine through the ECU (Engine control unit) of the engine. That is, the vehicle controller controls the power battery and the engine respectively through the BMS and the ECU, which is easy to know for those skilled in the art, so for the sake of simplicity, the BMS and the ECU are not specifically mentioned in the drawings and below.

[0032] Figure 2 A schematic diagram of the structural layout of a crawler hybrid intelligent orchard working robot according to an embodiment of the present application is schematically given. In the schematic diagram of the robot structural layout, the engine drives the generator to generate electricity through the speed increaser, which is rectified by the rectifier and becomes high-voltage direct current as the main power source of the working robot. The supercapacitor / power battery (high-voltage battery) is connected in parallel to the DC bus as an auxiliary power source, and the two together provide electric energy to drive the motor to operate. Specifically, the DC bus provides electric energy to the motor drivers arranged on the left and right sides, and the motor drives the reducers on the left and right sides to rotate. At this time, the left and right wheels connected to the reducers on the left and right sides rotate to drive the driving wheels of the track wheel systems on the left and right sides to operate, thereby realizing the movement of the working robot.

[0033] In remote control mode, the intelligent navigation controller completes the target speed control of the motors on both sides according to the driver's control signal and feedback signal; in automatic driving mode, through absolute or relative positioning information, the intelligent navigation controller outputs the deviation between the current positioning and the planned trajectory and sends it to the control module to output the control amount. The intelligent navigation controller receives the value and outputs the control target. Finally, the target speed signal is sent to the motor driver in real time through the bus to control the motor operation.

[0034] Figure 3A schematic diagram of energy flow in pure electric mode of a crawler hybrid intelligent orchard working robot according to an embodiment of the present application is schematically given, in which the robot is working in pure electric mode. In order to more intuitively illustrate the energy flow, some transmission devices are simplified in the figure, which is easy to understand for those skilled in the art and will not cause conceptual confusion. During the pure electric mode operation, only the power battery (or supercapacitor) serves as the energy source for the operation of the working robot. In this mode, the power battery outputs power to the motor, and the motor transmits power to the crawler wheel system under the coordination of the motor driver and the reducer, driving the entire crawler wheel system to operate and push the working robot forward.

[0035] Figure 4 A schematic diagram of energy flow in a hybrid mode of a crawler hybrid intelligent orchard operation robot according to an embodiment of the present application is schematically given, in which the operation robot is operating in a hybrid mode. During the operation in the hybrid mode, the power battery (or supercapacitor) and the engine serve as the energy source of the entire operation robot, and both of them jointly output power to the walking system and the operation system. The engine drives the generator to generate electricity and the current output by the power battery jointly outputs power to the motor, which is equivalent to forming a hybrid output system. The motor drives the driving wheels of the two crawler wheel systems, thereby driving the entire crawler wheel system to push the operation robot forward. In this process, the engine can also adjust the output power distribution through the clutch to drive the operation tools in the operation system, and the hydraulic motor is driven and controlled by the engine through the clutch. It is also possible to realize functions such as rotary tillage, trenching, and transportation through the transmission of the hydraulic system, and drive the hydraulic motor (hydraulic pump) through the PTO device on the engine, and the hydraulic motor in the operation system of the operation robot drives the water pump to pressurize and adjust the liquid in the liquid tank to achieve spraying of pesticides, meeting the requirements of the hydraulic mechanism function on the operation robot.

[0036] As another implementation scenario of this embodiment, in combination with the even number content, an operation method of a crawler hybrid intelligent orchard operation robot is proposed, and the operation method mainly includes: Trajectory planning: The operating robot's driving trajectory is planned by either manual trajectory planning or intelligent trajectory planning. Manual trajectory planning includes: using a handheld dotting device to pre-dot in the operating area to plan the operating robot's driving trajectory. Intelligent trajectory planning includes collecting images of fruit trees between rows based on the principle of visual navigation, marking the points where the images of fruit trees between rows are located, and planning the operating robot's driving trajectory.

[0037] Task execution: The operating robot moves along the planned driving trajectory. During the movement, according to the set work tasks, it performs plant protection operations on the driving trajectory, uses the fan to perform air atomization, and blows the liquid sprayed by the switch valve into the orchard.

[0038] When using manual trajectory planning to perform the robot's task, such as Figure 5 The following schematic diagram shows the intelligent navigation of the crawler-type hybrid intelligent orchard operation robot based on a satellite positioning device (handheld dotting device) according to an embodiment of the present application. The specific process is as follows: a) Dot recording: First, the staff uses a handheld dot recorder (RTK dot recorder) to record the GPS-RTK coordinates of each key point in the work area. The coordinates of these key points form the basis of the robot's driving trajectory. Figure 5 As shown in the figure, the black dotted lines represent the trajectories that the robot will travel, and these trajectories are recorded by a handheld dotting device.

[0039] b) Trajectory planning: The intelligent navigation controller generates the robot's driving trajectory based on the GPS-RTK coordinate data recorded by the handheld dot-marker. Figure 5 The working robot moves in the working area according to these pre-recorded driving trajectories.

[0040] c) Navigation of the working robot: The intelligent navigation controller of the working robot receives the real-time position information of the working robot through a satellite positioning device (GPS-RTK may also be used in this embodiment), and navigates according to the travel trajectory formed by the coordinates of key points recorded by a handheld dotting device. The high-precision positioning information provided by the GPS-RTK used in this embodiment ensures that the robot can accurately travel along the planned trajectory for absolute positioning and navigation. It can also use visual sensors to identify tree rows for relative positioning to navigate between ridges and avoid deviations from the track. When navigating with GPS-RTK, known plot information is required for global path planning. The plot information includes plot boundaries, orchard fruit tree row end coordinates, or tree row trajectory coordinates.

[0041] In addition to the above steps, this implementation adds the steps of the operation task in the manual trajectory planning process, as shown below: d) Execution of tasks: During navigation, the robot performs specific tasks at each location of the driving track according to the set work tasks. In this embodiment, the robot performs plant protection operations in the straight line segment, that is, spraying pesticides through the wind-blown atomization device; when the robot reaches the turning area, it stops spraying pesticides and performs a turning action. Figure 5 The "robot" in the lower left corner indicates that the working robot is moving along a recorded driving trajectory and performing tasks.

[0042] e) Monitoring and adjustment: When the working robot is performing tasks, the control system can monitor the position and movement trajectory of the working robot in real time through the camera, and make dynamic adjustments when necessary to ensure the accuracy of the operation.

[0043] As another implementation scenario of this embodiment, when the working robot adopts intelligent trajectory planning, Figure 6 and Figure 7 The following schematic diagram shows the intelligent navigation of the crawler-type hybrid intelligent orchard operation robot based on the visual sensor according to the embodiment of the present application. The specific process is as follows: 1) Step, data collection and training: First, it is necessary to rely on the camera carried by the operating robot to collect inter-row orchard image data in different orchards, different weather and different lighting environments.

[0044] 2) Step, data annotation and model training: In this embodiment, based on the principle of visual navigation, a deep learning model (UFLD model can be selected) is used to output the navigation line of the inter-row orchard, so the collected inter-row orchard images need to be annotated. Figure 6 As shown, during the annotation process, it is necessary to define the annotation lines of the left and right tree rows. The principle of selecting the annotation lines is to use the tree trunk and the ground grounding point as the key points, and the line formed by connecting the key points as the annotation line. The annotation line is input into the deep learning model for learning and training. If in the annotation process, there is a situation where the grounding point cannot be determined in the data, the edge pixels where the outermost side of the tree shape intersects with the ground are used as key points to form the annotation line. In this embodiment, the selected training model is the UFLD model of the neural network framework considering the global structure, which is a lane line detection model that is more familiar to relevant practitioners. The network structure is divided into two layers, the upper layer is an auxiliary layer, which only provides segmentation loss in the training stage and is not included in the actual operation in the reasoning stage; the lower layer is the actual operation layer for network prediction, and convolution operations are performed by superimposing multiple Res modules to obtain highly extracted features and send them to the fully connected layer, thereby obtaining structural loss and classification loss.

[0045] 3) Step, obtain the navigation line: obtain the image returned by the camera on the working robot online, run the UFLD model in the intelligent navigation controller to obtain the training results of the fitting straight line along the two tree rows, such as Figure 7 As shown, the detection result is the fitting straight line of two tree rows. Draw two parallel lines at different height layers of the image, of which the first parallel line is the lower part of the image, and the second parallel line is the upper part of the image. The two parallel lines are taken at 9 / 10 and 1 / 3 height from top to bottom respectively. Get the two intersection points p1 (x1, y1) and p2 (x2, y2) of the first parallel line and the tree row fitting straight line, and the two intersection points p3 (x3, y3) and p4 (x4, y4) of the second parallel line and the tree row fitting straight line, and find the intersection points p_s and p_e of p1, p2 and p3, p4. The line connecting the two intersection points is the navigation line obtained by the plane image.

[0046] 4) Step, navigation line conversion: Project the navigation line to the real working robot for coordinate conversion. The above navigation line is a pixel coordinate, which needs to be converted to the metric coordinate system of the vehicle body coordinate to have navigation significance. Usually, the conversion process is carried out through the inverse perspective projection transformation process. The pixel coordinates are converted through other precise distance measurement sensors (such as lidar) and converted into the driving trajectory of the working robot in the metric coordinate system. The working robot moves along the driving trajectory in the metric coordinate system. This process is a technology well known to technicians in this professional field and will not be described again. According to the navigation line, the current relative deviation of the working robot can be obtained, so as to perform lateral control to keep the working robot moving along the straight line navigation.

[0047] 5) Step, U-turn area navigation: In the U-turn area, the end of the field needs to be identified. Commonly used methods include end of field identification through markers, end of field identification through neural network training, and end of field identification based on physical priors. In this embodiment, the third method is adopted, that is, the cameras on the left and right sides of the working robot are used to make a judgment in combination with the length of the fitted straight line of the tree row in the above steps. Specifically, when the length of the fitted straight line of the two detected tree rows gradually decreases to below a certain set threshold, it is considered that the working robot is approaching the end of the row, and the left and right cameras carried by the working robot are started to identify the fruit trees on both sides. If it is determined that the fruit trees no longer appear in the camera within a certain time threshold (usually about 5s), it is considered that the working robot has arrived at the U-turn area and starts automatic navigation in the U-turn area. The navigation action followed by the U-turn area navigation is pre-set.

[0048] Step 6), re-enter the line for straight segment navigation: After the U-turn is completed, the working robot automatically returns to the straight segment navigation, and the navigation repeats the driving trajectory of steps 4) to 5). It should be noted that the direction of navigation in the U-turn area is alternated, that is, turn left-turn right-turn left-turn right, to ensure that the robot operates in a fixed direction.

[0049] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A crawler-type hybrid intelligent orchard operation robot, characterized in that: include: The walking system includes crawler wheel systems located on the left and right sides of the frame, and a motor, a reducer and a motor driver that establish a power transmission connection with each crawler wheel system; A power system, including an energy storage unit, an engine and a generator, wherein the energy storage unit and the engine jointly or individually transmit power to the travel system and the working system; The operating system includes a liquid medicine tank, a hydraulic motor, a water pump, a switch valve and a fan. The hydraulic motor is driven and controlled by the engine through a clutch. The hydraulic motor drives the water pump to pressurize and regulate the liquid medicine in the liquid medicine tank. The fan is powered by an energy storage unit and driven and controlled by an agricultural implement controller to blow the liquid medicine sprayed through the switch valve into the orchard. The liquid medicine tank comprises at least two detachable liquid medicine storage units, each of which has a different capacity, and the liquid medicine storage units are used in combination according to the operation requirements to adjust the total capacity of the liquid medicine tank. The liquid medicine storage units are equipped with liquid level sensors to monitor the amount of liquid medicine in each liquid medicine storage unit in real time and feed the data back to the control system; the control system adjusts the combination of the liquid medicine storage units according to the liquid medicine amount data and the operation plan; a liquid medicine flow sensor and a pressure sensor are installed on the switch valve, a hydraulic motor drives a water pump to extract liquid medicine in the liquid medicine tank and spray it out through the switch valve, and a fan is arranged next to the switch valve; The control system includes an intelligent navigation controller, a vehicle controller, a generator controller and an implement controller. The intelligent navigation controller includes a satellite positioning device, a camera and a laser radar, and controls the travel path of the travel system according to the positioning and planned driving trajectory; the generator controller is connected to the generator, and the implement controller is connected to and controls the operation of the implements of the operating system; the vehicle controller is connected to the travel system, the power system and the implement controller of the operating system.

2. A crawler-type hybrid intelligent orchard operation robot according to claim 1, characterized in that: The track wheel system in the walking system includes a driving wheel, a road wheel, a tensioner wheel and a towing wheel, which are arranged on the track in sequence from front to back, and the driving wheel establishes a power transmission connection with the reducer, the power output end of the motor is connected to the reducer, the motor driver establishes a control connection with the motor, and the motor driver receives control instructions from the vehicle controller of the control system.

3. The crawler-type hybrid intelligent orchard operation robot according to claim 1, characterized in that: The energy storage unit in the power system includes a fast-charging and discharging power battery and a supercapacitor. The supercapacitor and the power battery are connected in parallel on a DC bus. The DC bus and the motor driver provide electrical energy to the motor driver to drive the motor to operate. The vehicle controller is connected to the power battery through the battery management system of the power battery and to the engine through the engine controller of the engine. An electronic control unit is used for regulation and control between the engine and the generator.

4. The crawler-type hybrid intelligent orchard operation robot according to claim 1, characterized in that: The intelligent navigation controller of the control system is equipped with an environment perception module to detect obstacles on the operation path, automatically avoid obstacles and plan the path.

5. The crawler-type hybrid intelligent orchard operation robot according to claim 1, characterized in that: The power system includes a whole-machine power monitoring unit and a feedback unit, which monitor and adjust the output power of the engine and energy storage unit in real time. An energy distribution module is provided between the energy storage unit and the engine to analyze the energy requirements of the power battery and the engine in real time.

6. The crawler-type hybrid intelligent orchard operation robot according to claim 1, characterized in that: The power system adopts pure electric mode and hybrid mode to provide power energy to the working robot; during the pure electric mode operation, the power battery outputs power to the motor, and the motor transmits power to the track wheel system under the coordination of the motor driver and the reducer, driving the entire track wheel system to operate and push the working robot forward; during the hybrid mode operation, the power battery and the engine serve as the power source of the working robot, and jointly output power to the walking system and the working system. The engine drives the generator to generate electricity and the current generated by the power battery outputs power to the motor together, forming a hybrid power, driving the entire track wheel system to push the working robot forward.

7. The operating method of a crawler hybrid intelligent orchard operating robot according to any one of claims 1 to 6, characterized in that: The work method includes the following: Trajectory planning: The operation robot is planned to travel by manual trajectory planning or intelligent trajectory planning. Manual trajectory planning includes: pre-marking in the operation area with a handheld dotting device to plan the operation robot's travel trajectory; intelligent trajectory planning includes collecting images of fruit trees between rows based on the principle of visual navigation, marking the points where the images of fruit trees between rows are located, and planning the operation robot's travel trajectory; Task execution: The operating robot moves along the planned driving trajectory. During the movement, according to the set work tasks, it performs plant protection operations on the driving trajectory, uses the fan to perform air atomization, and blows the liquid sprayed by the switch valve into the orchard.

8. The operating method of a crawler hybrid intelligent orchard operating robot according to claim 7, characterized in that: In the manual trajectory planning process, a handheld dotting device is used to pre-dot the work area. Planning the robot's travel trajectory includes the following steps: a) Step, dotting and recording: The staff uses a handheld dotting device to dot in the work area and record the coordinates of each key point; b) Step, trajectory planning: record the coordinate data of key points through a handheld dotting device, and form the driving trajectory of the working robot in the intelligent navigation controller; c) Step, moving navigation: The working robot receives real-time position information through a satellite positioning device, and navigates according to the driving trajectory formed by the coordinates of key points recorded by a handheld dotting device.

9. The operating method of a crawler hybrid intelligent orchard operating robot according to claim 7, characterized in that: Intelligent trajectory planning includes collecting images of fruit trees between rows based on the principle of visual navigation, marking the points where the images of fruit trees between rows are located, and planning the driving trajectory of the operating robot in the following steps: 1) Step, data collection and training: Relying on the camera carried by the operating robot, collect orchard image data between rows in different orchards, different weather and different lighting environments; 2) Step, data annotation and model training: Based on the principle of visual navigation, the collected orchard images between rows are annotated. During the annotation process, the annotation lines of the left and right tree rows are defined and input into the deep learning model for learning and training; 3) Step, obtain the navigation line: Run the deep learning model in the intelligent navigation controller to obtain the training results of the fitted straight line along the two tree rows, select two parallel lines at different height layers of the image, the two intersection points of the first parallel line and the fitted straight line are recorded as p1 (x1, y1), p2 (x2, y2), the two intersection points of the second parallel line and the fitted straight line are recorded as p3 (x3, y3), p4 (x4, y4), find the intersection points p_s and p_e of p1, p2 and p3, p4, and the line connecting the two intersection points is the navigation line obtained by the plane image; 4) Step, navigation line conversion: project the navigation line onto the real working robot for coordinate conversion, convert the pixel coordinates of the navigation line into the driving trajectory of the working robot in the metric coordinate system, and the working robot moves according to the driving trajectory in the metric coordinate system; Step 5: Navigation in the U-turn area: When it is detected that the length of the fitted straight line between two rows of trees gradually decreases to a certain set threshold, the working robot is considered to be close to the end of the row, and the camera on the working robot is started to identify the fruit trees on both sides. If it is judged that the fruit trees no longer appear in the camera within a certain time threshold, it is considered that the working robot has arrived at the U-turn area and starts automatic navigation in the U-turn area; Step 6), re-enter the line for straight segment navigation: After the U-turn is completed, the working robot automatically returns to the navigation line of the straight segment, and the working robot repeats the driving trajectory of steps 4) to 5).

Images