Agricultural robot chassis field scene working condition simulation device and usage method

Abstract

The present invention discloses a field scene working condition simulation device for an agricultural robot chassis and a method for using the device, belonging to the technical field of working condition simulation devices. The device comprises: a base, a control center, a wheel-leg assembly, a steering loading assembly, a walking loading assembly, a ground contact loading assembly, an oil source power assembly, and a field scene and working condition data recording and playback assembly. The field scene and working condition data recording and playback assembly is used to record and playback field scenes and working conditions. The field scene and working condition data recording and playback assembly and the control center exchange data within the same local area network via a switch. Through the simulation device, characteristic analysis and verification of multiple subsystems such as the automatic driving system, wire-controlled chassis, and operating equipment of the equipment can be completed indoors, and control system performance analysis can be completed on an indoor test bench to find out the causes of failures and weak links and take corresponding countermeasures, thereby achieving the purpose of shortening the product development cycle and saving research and development costs.

Description

Technical Field

[0001] The present invention relates to the technical field of working condition simulation equipment, and in particular to a field scene working condition simulation equipment for an agricultural robot chassis and a use method thereof. Background Art

[0002] Agricultural robot technology is gradually being applied in the field of agricultural machinery. The testing and verification of intelligent navigation algorithms still relies on field tests, which are affected by uncertain factors such as crop growing season, test terrain, soil properties, and driving proficiency. The repeatability of field tests is poor, and the test accuracy and credibility are difficult to guarantee, resulting in a long product design and development cycle. Therefore, it is urgent to design an agricultural robot chassis field scene working condition simulation device and its use method, so that the characteristics analysis and verification of multiple subsystems of the equipment, such as the automatic driving system, wire-controlled chassis, and operating equipment, can be completed indoors. This will accelerate the research and development of key technologies and improve equipment quality. With the help of the agricultural robot chassis field scene working condition simulation device, it is ensured that the agricultural robot reaches the expected reliability level, and the control system performance analysis is completed on the indoor test bench to find out the causes of failures and weak links, and take corresponding countermeasures, so as to shorten the product development cycle and save R&D costs. Summary of the Invention

[0003] The technical problem to be solved by the present invention is: through indoor testing, the performance analysis of the control system can be completed, the causes of its failures and weak links can be found, and corresponding countermeasures can be taken to shorten the product development cycle and save R&D costs.

[0004] The technical solution adopted by the present invention to solve the technical problem is: an agricultural robot chassis field scene working condition simulation device, including: a base, a control center, a wheel-leg assembly, a steering loading assembly, a walking loading assembly, a ground contact loading assembly, an oil source power assembly, and a field scene and working condition data recording and playback assembly;

[0005] The seat body is provided with a frame and a mounting plate, the frame is provided with a damping adjustable shock absorber, the frame is movably connected to a fixed arm, and the damping adjustable shock absorber is movably connected to the fixed arm;

[0006] The wheel-leg assembly includes a steering shaft, a wheel bracket and a steering cylinder. The steering shaft is arranged at the end of the fixed arm and connected to the top of the wheel bracket. A travel motor is arranged at the bottom of the wheel bracket, and the output end of the travel motor is connected to the wheel. The steering cylinder is arranged on the side of the fixed arm, and the output end of the steering cylinder is connected to one end of the steering cantilever. The other end of the steering cantilever is connected to the wheel bracket through a steering torque sensor.

[0007] The steering loading assembly includes a steering angle sensor, a steering loading cylinder, and a steering loading cantilever. The steering angle sensor is arranged on the top of the steering shaft, the steering loading cylinder is arranged on the side of the fixed arm, one end of the steering loading cantilever is arranged on the wheel bracket, and the output end of the steering loading cylinder is connected to the other end of the steering loading cantilever.

[0008] The travel loading assembly is arranged on the mounting plate, and the output end of the travel loading assembly is connected to the wheel drive to apply a travel loading force to the wheel;

[0009] The ground contact loading assembly is located below the wheel and in contact with the wheel for applying a ground contact loading force to the wheel;

[0010] The control center and the oil source power assembly cooperate with each other to provide power output to the travel motor, steering cylinder, steering loading cylinder, travel loading assembly and ground contact loading assembly and control the power output;

[0011] The field scene and working condition data recording and playback component is used to record and playback field scenes and working conditions. The field scene and working condition data recording and playback component exchanges data with the control center in the same local area network through a switch.

[0012] As a preferred technical solution of the present invention, the field scene and working condition data recording and playback component includes a robot, and the robot includes four wheel speed sensors, four steering angle sensors, a depth camera, an inertial attitude measurement unit, a multi-line laser radar sensor, a positioning and orientation terminal, and a high-performance computing processing center. The four wheel speed sensors are respectively installed on the four wheel body shafts of the robot for real-time measurement of the wheel body speed data during field walking. The four steering angle sensors are installed on the four wheel body steering shafts of the robot for real-time measurement of the wheel body steering angle data during field walking. The depth camera and the multi-line laser radar sensor are centrally installed at a position point. Located in the upper front of the robot's front and on the longitudinal symmetry plane of the vehicle body, the inertial attitude measurement unit is installed at the center of the robot's chassis frame. The multi-line lidar sensor collects three-dimensional point cloud data of the working scene, and the depth camera collects image information in the forward direction. The inertial attitude measurement unit collects the attitude angle of the center point of the agricultural robot. The real-time collection of sensor data and the alignment of image and point cloud are realized on the high-performance computing processing center. The positioning and orientation terminal has two GPS antennas and is installed on the roof of the robot in a front-to-back layout. The positioning and orientation terminal collects the robot's position information and heading angle. The software used for data collection is developed based on the ROS robot operating system, and the recorded data is saved as a bag file type.

[0013] As a preferred technical solution of the present invention, the walking loading assembly includes a walking loading motor and a walking sensor assembly. A walking loading bracket and a sensor bracket are provided on the mounting plate. The walking loading motor is provided on the walking loading bracket. The walking sensor assembly is provided on the sensor bracket. The output end of the walking loading motor is connected to a first cross universal joint, the first cross universal joint is connected to the walking sensor assembly, the walking sensor assembly is connected to a second cross universal joint, the second cross universal joint is connected to an electromagnetic telescopic coupling, the electromagnetic telescopic coupling is connected to a third cross universal joint, the third cross universal joint is fixedly connected to the center point of the wheel, and the oil source power assembly is connected to the walking loading motor.

[0014] As a preferred technical solution of the present invention, the walking sensor assembly includes a walking torque sensor and a speed sensor, the speed sensor is mounted on the walking torque sensor, the first cross universal joint is connected to the end of the walking torque sensor, a sensor rotating shaft is provided in the walking torque sensor, the sensor rotating shaft extends to the outer end of the walking torque sensor and is connected to the second cross universal joint, the electromagnetic telescopic coupling includes a telescopic sleeve and a telescopic special-shaped shaft, the telescopic special-shaped shaft is provided in the telescopic sleeve, the telescopic sleeve is connected to the second cross universal joint, and the telescopic special-shaped shaft is connected to the third cross universal joint.

[0015] As a preferred technical solution of the present invention, the ground contact loading assembly includes a ground contact force loading cylinder, a force sensor is provided on the output end of the ground contact force loading cylinder, a contact disc is provided on the force sensor, the contact disc is located under the wheel and in contact with the wheel, and the oil source power assembly is connected to the ground contact force loading cylinder.

[0016] As a preferred technical solution of the present invention, the ground contact loading assembly further includes a displacement sensor, which is located below the connection position between the telescopic special-shaped shaft and the wheel through the third cross universal joint.

[0017] A method for using the above-mentioned agricultural robot chassis field scene working condition simulation device to perform chassis motion simulation includes the following steps:

[0018] S1: With the help of the robot, the four wheel speed sensors, four steering angle sensors, a depth camera, an inertial attitude measurement unit, a multi-line laser radar sensor, a positioning and orientation terminal, and a high-performance computing processing center are used to record field scene and working condition data, and the high-performance computing processing center is used to play back the data indoors to complete the three-dimensional scene mapping of the farmland;

[0019] S2: With the help of the control center, field scenes and working conditions are simulated indoors through the wheel-leg assembly, steering loading assembly, walking loading assembly, ground contact loading assembly and oil source power assembly;

[0020] S3: The staff compares and analyzes the acquired information with the wheel steering torque information, wheel torque information, expected speed and expected steering angle, operates the control center, and adjusts the control gain coefficient for controlling the wheel steering and speed so that the wheel-leg assembly meets the requirements.

[0021] The method for using the above-mentioned agricultural robot chassis field scene working condition simulation device to perform chassis motion simulation, wherein S1 specifically includes the following steps:

[0022] S1.1: Measure the distance from each sensor to the coordinate origin using the center of the robot chassis as the coordinate origin, and publish the coordinate transformation matrix of each sensor to the chassis coordinate origin in the ROS system;

[0023] S1.2: The high-performance computing processing center includes industrial computer 1, which starts the driver and data acquisition program for all sensors on industrial computer 1, collects data from each sensor at a frequency of 10 Hz, and saves the data in real-time in the form of a bag file to industrial computer 1. The robot's driving path traverses the entire field in a full-coverage manner.

[0024] S1.3: The high-performance computing processing center also includes an industrial computer 2, which collects and transmits the recorded data to the industrial computer 2. The industrial computer 2 initiates a data playback program based on the ROS robot operating system, reads the contents of one or more bag files, and performs a first data playback in a time-synchronized manner, republishing the sensor data from the data packets.

[0025] S1.4: The high-performance computing processing center has a built-in farmland 3D mapping program for performing lidar mapping and robot positioning. The program starts the 3D farmland scene lidar mapping and robot positioning program, constructs a 3D map of the farmland scene, and outputs the robot's trajectory points on the map. The center waits for the data packet playback to complete, and saves the farmland scene point cloud map and all robot movement trajectory data.

[0026] S1.5: The second industrial computer has a pre-installed robot real-time simulation program, a sensor simulation program, a path planning program, and a sensor fusion positioning program. The robot real-time simulation program is started, a three-dimensional map of the farmland scene is loaded, a robot simulation model is loaded, the sensor simulation program is loaded, the path planning program is started, and the sensor fusion positioning program is started.

[0027] S1.6: Launching the trajectory tracking software in the robot real-time simulation program on the second industrial computer. The trajectory tracking software subscribes to the robot's target path information and current position and posture information. The trajectory tracking algorithm calculates the robot's heading angle and forward speed in real time.

[0028] S1.7: The second industrial computer also includes a robot real-time kinematics solver. The robot real-time kinematics solver is started, subscribes to the robot's heading angle and forward speed in real time, obtains the expected speed and expected steering angle of each wheel through inverse kinematic calculations, and publishes the results to the ROS robot operating system server in real time.

[0029] The method for using the above-mentioned agricultural robot chassis field scene working condition simulation device to perform chassis motion simulation, wherein S2 specifically includes the following steps:

[0030] S2.1: The control center is pre-installed with a wheel control program for controlling the travel motor and steering cylinder via the oil source power assembly, as well as a loading control program for controlling the steering loading cylinder, travel loading motor, and ground contact force loading cylinder. Wheel steering torque information and wheel torque information are pre-set in the control center. After the data playback system is activated, the control center receives the desired speed information and desired steering angle information published by the ROS server in step S1.7.

[0031] S2.2: The control center controls the travel motor and steering cylinder using the wheel control program, so that the wheels move according to the desired speed and desired steering angle calculated in real time in step S1.7;

[0032] S2.3: The control center controls the steering loading cylinder using the loading control program. The steering loading cylinder applies a steering loading force to the wheel bracket via the steering loading cantilever, so that the wheel is subjected to the steering loading force and can obtain a simulated steering load environment. The steering torque sensor measures the steering loading torque information of the wheel in real time, and the steering angle sensor measures the steering angle information of the wheel.

[0033] S2.4: The control center controls the travel loading motor using the loading control program. The travel loading motor applies a travel loading force to the wheel via the first cross universal joint, the travel torque sensor, the speed sensor, the second cross universal joint, the telescopic sleeve, the telescopic special-shaped shaft, and the third cross universal joint, so that the wheel can obtain a simulated travel load environment. The travel torque sensor measures the travel loading torque information of the wheel in real time, and the speed sensor measures the speed information of the wheel.

[0034] S2.5: The control center controls the ground contact force loading cylinder using the loading control program. The ground contact force loading cylinder applies a contact loading force to the wheel via the contact disk, so that the wheel can obtain a simulated ground contact load environment. The force sensor measures the ground contact loading force information, and the displacement sensor measures the displacement information of the telescopic special-shaped shaft relative to the telescopic sleeve in real time.

[0035] S2.6: The control center is connected to the high-performance computing processing center, the steering torque sensor, the angle sensor, the walking torque sensor, the speed sensor, the displacement sensor and the force sensor for information exchange. The steering torque sensor transmits the steering loading torque information to the control center, the angle sensor transmits the angle information to the control center, the walking torque sensor transmits the walking loading torque information to the control center, the speed sensor transmits the speed information to the control center, the displacement sensor transmits the displacement information to the control center, and the force sensor transmits the ground contact loading force information to the control center.

[0036] The method for using the above-mentioned agricultural robot chassis field scene working condition simulation device to perform chassis motion simulation, wherein S2.1 specifically includes the following steps:

[0037] S2.1.1: After the data playback system is started, the high-performance computing processing center publishes the robot status data to the ROS robot operating system server in real time, and the control center receives the data published by the ROS robot operating system server in real time via Ethernet.

[0038] The beneficial effects of the present invention are embodied in:

[0039] 1. By setting up steering loading components, walking loading components, ground contact loading components, and coordinating with wheel-leg components and the control center, the characteristic analysis and verification of multiple subsystems of the equipment, such as the automatic driving system, wire-controlled chassis, and operating equipment, can be completed indoors. This will accelerate the research and development of key technologies and improve equipment quality. With the help of the field scene working condition simulation equipment of the agricultural robot chassis, it is ensured that the agricultural robot reaches the expected reliability level. The control system performance analysis is completed on the indoor test bench, the causes of failures and weak links are found, and corresponding countermeasures are taken to shorten the product development cycle and save R&D costs.

[0040] 2. By setting up field scene and working condition data recording and playback components to record field scenes and working conditions and play them back indoors, it is convenient to cooperate with the control center, wheel-leg components, steering loading components, walking loading components, and ground contact group loading components to simulate indoor working conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is a schematic diagram of the present invention;

[0042] Figure 2 It is a front view schematic diagram of the present invention;

[0043] Figure 3 It is an enlarged schematic diagram of part A of the present invention;

[0044] Figure 4 Schematic diagram of the electromagnetic telescopic coupling of the present invention;

[0045] Figure 5 1 is a schematic front view of the electromagnetic telescopic coupling of the present invention;

[0046] Figure 6 It is a schematic diagram of the BB portion of the present invention;

[0047] Figure 7 It is a flow chart of the method of the present invention.

[0048] In the figure: 1. Base; 2. Frame; 3. Mounting plate; 4. Travel loading bracket; 5. Sensor bracket; 6. Damping adjustable shock absorber; 7. Fixed arm; 8. Steering shaft; 9. Steering torque sensor; 10. Wheel bracket; 11. Travel motor; 12. Wheel; 13. Steering cylinder; 14. Steering cantilever; 15. Angle sensor; 16. Steering loading cylinder; 17. Steering loading cantilever; 18. Travel loading motor; 19. First cross universal joint; 20. Travel torque sensor; 21. Speed sensor; 22. Second cross universal joint; 23. Telescopic sleeve; 24. Telescopic special-shaped shaft; 25. Third cross universal joint; 26. Displacement sensor; 27. Surface contact force loading cylinder; 28. Force sensor; 29. Contact disc. DETAILED DESCRIPTION

[0049] The present invention will now be described in further detail with reference to the accompanying drawings.

[0050] Combined with attachment Figure 1-7 As shown, an agricultural robot chassis field scene working condition simulation device and a use method include a base body 1, a frame body 2, a mounting plate 3, a walking loading bracket 4, a sensor bracket 5, a damping adjustable shock absorber 6, a fixed arm 7, a steering shaft 8, a steering torque sensor 9, a wheel bracket 10, a walking motor 11, a wheel 12, a steering cylinder 13, a steering cantilever 14, an angle sensor 15, a steering loading cylinder 16, a steering loading cantilever 17, a walking loading motor 18, a first cross universal joint 19, a walking torque sensor 20, a speed sensor 21, a second cross universal joint 22, a telescopic sleeve 23, a telescopic special-shaped shaft 24, a third cross universal joint 25, a displacement sensor 26, a surface contact force loading cylinder 27, a force sensor 28 and a contact disc 29.

[0051] Combined with attachment Figure 1 、 7 As shown, an agricultural robot chassis field scene working condition simulation device includes: a base body 1, a control center, a wheel-leg assembly, a steering loading assembly, a walking loading assembly, a ground contact loading assembly, and an oil source power assembly. The control center pre-stores collected data, and the collected data is collected by the robot;

[0052] Preferably, the robot includes four wheel speed sensors, four steering angle sensors, a depth camera, an inertial attitude measurement unit (i.e., IMU sensor), a multi-line laser radar sensor, a positioning and orientation terminal, and a high-performance computing and processing center. When the robot is driving in the field, the high-performance computing and processing center collects chassis kinematic parameters and farmland operation scene data. The scene data includes soil ground, crop rows, and other obstacles in the field (telephone poles, trees, people, etc.). Preferably, the positioning and orientation terminal selects a Beidou differential positioning and orientation terminal;

[0053] Specifically, the four wheel speed sensors are respectively installed on the four wheel shafts of the robot to measure the wheel speed data in real time during field walking. The four steering angle sensors are installed on the four wheel steering shafts of the robot to measure the steering angle data of the wheel in real time during field walking. The depth camera and multi-line lidar sensor are centrally installed at a position point, which is located in the upper front of the robot's front and on the longitudinal symmetry plane of the vehicle body. The inertial attitude measurement unit (i.e., IMU sensor) is installed at the center of the robot's chassis frame. The multi-line lidar sensor collects three-dimensional point cloud data of the operation scene. The depth camera collects image information in the forward direction, and the inertial attitude measurement unit (IMU sensor) collects the attitude angles (including pitch, yaw, and roll angles) of the agricultural robot's center point. The real-time acquisition of sensor data and the registration of images and point clouds are implemented on a high-performance computing processing center. The positioning and orientation terminal has two GPS antennas, which are installed on the roof of the robot in a front-to-back layout with a spacing greater than 1.2m. They are arranged symmetrically with the center of the chassis. The positioning and orientation terminal collects the robot's position information and heading angle. The software used for data acquisition is developed based on the ROS robot operating system, and the recorded data is saved as a bag file type.

[0054] Furthermore, during data collection, 1) first, the center of the robot chassis is used as the coordinate origin, and the distance from each sensor to the coordinate origin is measured. The coordinate transformation matrix of each sensor to the chassis coordinate origin is published in the ROS system. 2) The high-performance computing processing center includes industrial computer 1 and industrial computer 2. Industrial computer 1 is installed on the D disk of the robot. The driver and data acquisition program of all sensors on industrial computer 1 are started, and the data of each sensor is collected at a frequency of 10Hz. The data is saved in real time to industrial computer 1 in the bag file format. The field driving control program of the robot is started, and manual remote control driving and automatic driving are optional. The driving path of the robot traverses the entire field in a full coverage manner. 3) The collected and recorded data is transmitted to industrial computer 2. 4) Start the data playback program based on the ROS robot operating system, read the contents of one or more bag files, and perform the first data playback in a time-synchronized manner, and republish the sensor data from the data packet. 5) Start the farmland 3D scene lidar mapping and robot positioning program. Preferably, existing 3D mapping algorithms such as LIO-SAM and NDT can be used. The program subscribes to data from sensors such as depth cameras, inertial attitude measurement units (i.e., IMU sensors), multi-line lidar sensors, and positioning and orientation terminals to build a 3D map of the farmland scene and output the robot's trajectory points on the map. Wait for the data packet playback to end, save the farmland scene point cloud map and all the robot's walking trajectory data, and 5) On the industrial computer 2 Start the real-time simulation program of the robot: ① Load the three-dimensional map of the farmland scene (point cloud map); ② Load the robot simulation model; ③ Load the sensor simulation program, including the simulation program of the depth camera, inertial attitude measurement unit (i.e. IMU sensor), and multi-line lidar sensor, and output the data perceived by various sensors in real time under the simulation environment; ④ Start the path planning program, plan the robot's driving path on the map, and publish it to the node server in real time in the form of a standard path topic in ROS. Preferably, directly read the walking trajectory data saved in the previous step and publish it; ⑤ Start the sensor fusion positioning program on the industrial computer 2, call the coordinate transformation matrix parameters of the sensor to the chassis origin coordinates; use the subscribed lidar and depth camera to calculate ... The simulation data of multiple sensors such as the robot and IMU sensors are integrated into the positioning program to predict the robot's position and posture information on the point cloud map of the farmland scene in real time. 6) The trajectory tracking software in the robot's real-time simulation program is started on the second industrial computer. This program subscribes (calls) the robot's target path information and current position and posture information. The trajectory tracking algorithm calculates the robot's heading angle and forward speed in real time. The accuracy of trajectory tracking can be tuned by adjusting the control gain coefficient. 7) The robot's real-time kinematic solution program is started, subscribing to the robot's heading angle and forward speed in real time. Through the inverse kinematic operation, the expected speed and expected steering angle of each wheel are obtained and published to the ROS server in real time, thus completing data collection and playback.

[0055] The control center is connected to the second industrial computer in the same local area network via a switch.

[0056] Combined with attachment Figure 1-7 As shown, the seat body 1 is provided with a frame body 2 and a mounting plate 3, the frame body 2 is provided with a damping adjustable shock absorber 6, the frame body 2 is movably connected with a fixed arm 7, the damping adjustable shock absorber 6 is movably connected to the fixed arm 7, the wheel leg assembly includes a steering shaft 8, a wheel bracket 10 and a steering cylinder 13, the steering shaft 8 is provided at the end of the fixed arm 7, the steering shaft 8 is connected to the top of the wheel bracket 10, a travel motor 11 is provided at the bottom of the wheel bracket 10, the output end of the travel motor 11 is connected to the wheel 12, the steering cylinder 13 is provided on the side of the fixed arm 7, and the output end of the steering cylinder 13 is connected to the steering cantilever 14 One end of the steering cantilever 14 is connected to the wheel bracket 10 through the steering torque sensor 9. The steering loading assembly includes a steering angle sensor 15, a steering loading cylinder 16 and a steering loading cantilever 17. The steering angle sensor 15 is arranged on the top of the steering shaft 8, and the steering loading cylinder 16 is arranged on the side of the fixed arm 7. One end of the steering loading cantilever 17 is arranged on the wheel bracket 10, and the output end of the steering loading cylinder 16 is connected to the other end of the steering loading cantilever 17. The walking loading assembly is arranged on the mounting plate 3, and the output end of the walking loading assembly is connected to the wheel 12 for applying a walking loading force to the wheel 12. The ground contact loading component is located below the wheel 12 and contacts the wheel to apply a ground contact loading force to the wheel 12. The control center cooperates with the oil source power component to provide power output to the travel motor 11, the steering cylinder 13, the steering loading cylinder 16, the travel loading component and the ground contact loading component and control the power output. Preferably, the oil source power component is connected to the travel motor 11, the steering cylinder 13, the steering loading cylinder 16, the travel loading component and the ground contact loading component to provide power output to the travel motor 11, the steering cylinder 13, the steering loading cylinder 16, the travel loading component and the ground contact loading component. The load component provides power, and the control center is connected to the oil source power component to control the power output of the oil source power component. The steering cylinder 13 is provided to provide power to drive the wheel bracket 10 to steer with the help of the steering cantilever 14, so as to achieve the purpose of steering the wheel 12. The steering loading cylinder 16 provides power and provides a steering loading force for the steering of the wheel 12 through the steering loading cantilever 17. When in use, the user can set the torque loading curve according to the field scene and working conditions, and then adjust the steering loading cylinder 16 through the program until the actual torque value measured by the steering torque sensor 9 is within the allowable error range with the set torque value;

[0057] Preferably, the control center is pre-installed with a real-scale three-dimensional simulation model program of the robot. The program uses the ROS multi-machine communication service software in the local area network to receive the expected motion instructions of the robot's four wheel legs leg1, leg2, leg3, and leg4 in real time, including the expected rotation speeds (vd1, vd2, vd3, and vd4) and expected steering angles (wd1, wd2, wd3, and wd4) of the four wheel legs (left front wheel, right front wheel, left rear wheel, and right rear wheel). According to the expected motion instructions of the robot's four wheel legs, the four wheel rotation joints and the four wheel steering joints of the simulation model program are driven to move according to the expected instructions to obtain the expected rotation speed information and the expected steering angle information used by the control center when performing simulation tests;

[0058] Preferably, the oil source power assembly includes an oil tank, a cooling fan, an oil pump, a three-phase asynchronous motor, a filter, a travel loading control valve block, a steering loading control valve block, a travel control valve block and a steering control valve block.

[0059] Combined with attachment Figure 1-3As shown, the walking loading assembly includes a walking loading motor 18 and a walking sensor assembly. The mounting plate 3 is provided with a walking loading bracket 4 and a sensor bracket 5. The walking loading motor 18 is provided on the walking loading bracket 4. The walking sensor assembly is provided on the sensor bracket 5. The output end of the walking loading motor 18 is connected to a first cross universal joint 19. The first cross universal joint 19 is connected to the walking sensor assembly. The walking sensor assembly is connected to a second cross universal joint 22. The walking sensor assembly includes a walking torque sensor 20 and a speed sensor 21. The speed sensor 21 is mounted on the walking torque sensor 20, the first cross universal joint 19 is connected to the end of the walking torque sensor 20, a sensor rotating shaft is provided in the walking torque sensor 20, the sensor rotating shaft extends to the outer end of the walking torque sensor 20 and is connected to the second cross universal joint 22, the second cross universal joint 22 is connected to an electromagnetic telescopic coupling, the electromagnetic telescopic coupling is connected to a third cross universal joint 25, wherein the electromagnetic telescopic coupling includes a telescopic sleeve 23 and a telescopic special-shaped shaft 24, preferably, an electromagnet is provided in the telescopic sleeve 23, the telescopic special-shaped shaft 24 is located at the telescopic sleeve An electromagnet 2 is provided at the inner end of the sleeve 23, and the control center monitors the driving status of the robot in real time. When the robot is moving straight, an electromagnet 1 is provided in the telescopic sleeve 23, and an electromagnet 2 is provided at the inner end of the telescopic sleeve 23. The electromagnet 1 and the electromagnet 2 are attracted after being energized; when the robot turns, the steering cylinder 13 is actuated, and the electromagnet 1 and the electromagnet 2 are de-energized and released. A displacement grating scale reading head is provided on the inner sleeve wall of the telescopic sleeve 23 near the telescopic special-shaped shaft 24, and a displacement grating scale that matches the displacement grating scale reading head is provided on the telescopic special-shaped shaft 24. The telescopic special-shaped shaft 24 is provided with a displacement grating scale that matches the displacement grating scale reading head. Inside the telescopic sleeve 23, the telescopic sleeve 23 is connected to the second cross universal joint 22, the telescopic special-shaped shaft 24 is connected to the third cross universal joint 25, the third cross universal joint 25 is fixedly connected to the center point of the wheel 12, and the oil source power assembly is connected to the travel loading motor 18. The travel loading motor 18 is set to provide power so as to provide a travel loading force to the wheel 12. During use, the user can set the torque loading curve according to the field scene and working conditions, and then adjust the travel loading motor 18 through the program until the actual torque value measured by the travel torque sensor 20 is within the allowable error range with the set torque value.

[0060] Combined with attachment Figure 1-3As shown, the ground contact loading assembly includes a ground contact force loading cylinder 27, and a force sensor 28 is provided on the output end of the ground contact force loading cylinder 27. A contact disc 29 is provided on the force sensor 28. The contact disc 29 is located below the wheel 12 and contacts the wheel 12. The oil source power assembly is connected to the ground contact force loading cylinder 27. The ground contact loading assembly also includes a displacement sensor 26. The displacement sensor 26 is located below the connection position between the telescopic special-shaped shaft 24 and the wheel 12 through the third cross universal joint 25. With the help of the ground contact force loading cylinder 27, power is provided to drive the contact disc 29 to apply a ground contact loading force to the wheel 12. During use, the user can control the ground contact force loading cylinder 27 in real time through the control program until the actual values measured by the force sensor 28 and the displacement sensor 26 are within the allowable error range of the set loading curve (force curve or displacement curve).

[0061] A method for using the above-mentioned agricultural robot chassis field scene working condition simulation device to perform chassis motion simulation includes the following steps:

[0062] S1: With the help of the robot, the four wheel speed sensors, four steering angle sensors, a depth camera, an inertial attitude measurement unit, a multi-line laser radar sensor, a positioning and orientation terminal, and a high-performance computing processing center are used to record field scene and working condition data, and the high-performance computing processing center is used to play back the data indoors to complete the three-dimensional scene mapping of the farmland;

[0063] S2: With the help of the control center, field scenes and working conditions are simulated indoors through the wheel-leg assembly, steering loading assembly, walking loading assembly, ground contact loading assembly and oil source power assembly;

[0064] S3: The staff compares and analyzes the acquired information with the wheel steering torque information, the wheel torque information, the expected speed and the expected steering angle, operates the control center, and adjusts the control gain coefficient for controlling the steering and speed of the wheel (12) so that the wheel-leg assembly meets the requirements;

[0065] Among them, the control center receives information from the steering torque sensor 9, the angle sensor 15, the walking torque sensor 20, the speed sensor 21, the displacement sensor 26 and the force sensor 28, obtains the measured speed value of the wheel 12 and the steering angle value, and sends it to the industrial computer 2 in the form of a ROS topic, replacing the action state of the corresponding wheel leg in the simulation model, so as to perform data analysis of the joint work of the four wheel legs. The remaining wheel legs are tested in turn, and the navigation trajectory tracking error is recorded with the help of the industrial computer 2, and the steering angle tracking error and the wheel speed tracking error are recorded with the help of the control center, and the control gain coefficient is adjusted until the control accuracy meets the field operation requirements of the robot chassis.

[0066] The method for using the above-mentioned agricultural robot chassis field scene working condition simulation device to perform chassis motion simulation, wherein S1 specifically includes the following steps:

[0067] S1.1: Measure the distance from each sensor to the coordinate origin using the center of the robot chassis as the coordinate origin, and publish the coordinate transformation matrix of each sensor to the chassis coordinate origin in the ROS system;

[0068] S1.2: The high-performance computing processing center includes industrial computer 1, which starts the driver and data acquisition program for all sensors on industrial computer 1, collects data from each sensor at a frequency of 10 Hz, and saves the data in real-time in the form of a bag file to industrial computer 1. The robot's driving path traverses the entire field in a full-coverage manner.

[0069] S1.3: The high-performance computing processing center also includes an industrial computer 2, which collects and transmits the recorded data to the industrial computer 2. The industrial computer 2 initiates a data playback program based on the ROS robot operating system, reads the contents of one or more bag files, and performs a first data playback in a time-synchronized manner, republishing the sensor data from the data packets.

[0070] S1.4: The high-performance computing processing center has a built-in farmland 3D mapping program for performing lidar mapping and robot positioning. The program starts the 3D farmland scene lidar mapping and robot positioning program, constructs a 3D map of the farmland scene, and outputs the robot's trajectory points on the map. The center waits for the data packet playback to complete, and saves the farmland scene point cloud map and all robot movement trajectory data.

[0071] S1.5: The second industrial computer has a pre-installed robot real-time simulation program, a sensor simulation program, a path planning program, and a sensor fusion positioning program. The robot real-time simulation program is started, a three-dimensional map of the farmland scene is loaded, a robot simulation model is loaded, the sensor simulation program is loaded, the path planning program is started, and the sensor fusion positioning program is started.

[0072] S1.6: Launching the trajectory tracking software in the robot real-time simulation program on the second industrial computer. The trajectory tracking software subscribes to the robot's target path information and current position and posture information. The trajectory tracking algorithm calculates the robot's heading angle and forward speed in real time.

[0073] S1.7: The second industrial computer also includes a robot real-time kinematics solver. The robot real-time kinematics solver is started, subscribes to the robot's heading angle and forward speed in real time, obtains the expected speed and expected steering angle of each wheel through inverse kinematic calculations, and publishes the results to the ROS robot operating system server in real time.

[0074] The method for using the above-mentioned agricultural robot chassis field scene working condition simulation device to perform chassis motion simulation, wherein S2 specifically includes the following steps:

[0075] S2.1: The control center is pre-installed with a wheel control program for controlling the travel motor 11 and steering cylinder 13 via the oil source power assembly, as well as a loading control program for controlling the steering loading cylinder 16, travel loading motor 18, and ground contact force loading cylinder 27. Wheel steering torque information and wheel torque information are pre-set in the control center. After the data playback system is activated, the control center receives the desired speed information and desired steering angle information published on the ROS server in step S1.7.

[0076] After the data playback system is started, the high-performance computing processing center publishes the robot status data to the ROS robot operating system server in real time, and the control center receives the data published on the ROS robot operating system server in real time via Ethernet;

[0077] S2.2: The control center uses the wheel control program to control the travel motor 11 and the steering cylinder 13 through the oil source power assembly, so that the wheel 12 moves according to the desired speed and desired steering angle calculated in real time in step S1.7;

[0078] According to the specific position of the selected wheel-leg assembly corresponding to the four wheel legs of the robot, the control center selects the desired rotation speed information and the desired steering angle information of the corresponding wheel legs, and calculates the control signals of the wheel-leg assembly walking motor 11 and the steering cylinder 13 through the control algorithm;

[0079] S2.3: The control center controls the steering loading cylinder 16 via the oil source power assembly using the loading control program. The steering loading cylinder 16 applies a steering loading force to the wheel bracket 10 via the steering loading cantilever 17, so that the wheel 12 is subjected to the steering loading force, thereby obtaining a simulated steering load environment. The steering torque sensor 9 measures the steering loading torque information of the wheel 12 in real time, and the steering angle sensor 15 measures the steering angle information of the wheel 12.

[0080] S2.4: The control center controls the travel loading motor 18 via the oil source power assembly using the loading control program. The travel loading motor 18 applies a travel loading force to the wheel 12 via the first cross universal joint 19, the travel torque sensor 20, the speed sensor 21, the second cross universal joint 22, the telescopic sleeve 23, the telescopic special-shaped shaft 24, and the third cross universal joint 25, so that the wheel 12 can obtain a simulated travel load environment. The travel torque sensor 20 measures the travel loading torque information of the wheel in real time, and the speed sensor 21 measures the speed information of the wheel 12.

[0081] S2.5: The control center controls the ground contact force loading cylinder 27 via the oil source power assembly using the loading control program. The ground contact force loading cylinder 27 applies a contact loading force to the wheel 12 via the contact disk 29, so that the wheel 12 can obtain a simulated ground contact load environment. The force sensor 28 measures the ground contact loading force, and the displacement sensor 26 measures the displacement of the telescopic special-shaped shaft 24 relative to the telescopic sleeve 23 in real time.

[0082] S2.6: The control center is connected to the high-performance computing processing center, the steering torque sensor 9, the angle sensor 15, the walking torque sensor 20, the speed sensor 21, the displacement sensor 26 and the force sensor 28 for information exchange. The steering torque sensor 9 transmits the steering loading torque information to the control center, the angle sensor 15 transmits the angle information to the control center, the walking torque sensor 20 transmits the walking loading torque information to the control center, the speed sensor 21 transmits the speed information to the control center, the displacement sensor 26 transmits the displacement information to the control center, and the force sensor 28 transmits the ground contact loading force information to the control center.

[0083] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions in the specification are only preferred examples of the present invention and are not intended to limit the present invention. Various changes and improvements may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and improvements fall within the scope of the present invention as claimed. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A field scene working condition simulation device for an agricultural robot chassis, comprising: The seat (1), control center, wheel leg assembly, steering loading assembly, walking loading assembly, ground contact loading assembly, oil source power assembly and field scene and working condition data recording and playback assembly are characterized by: The seat body (1) is provided with a frame body (2) and a mounting plate (3); the frame body (2) is provided with a damping adjustable shock absorber (6); the frame body (2) is movably connected to a fixed arm (7); the damping adjustable shock absorber (6) is movably connected to the fixed arm (7); The wheel-leg assembly comprises a steering shaft (8), a wheel bracket (10) and a steering oil cylinder (13); the steering shaft (8) is arranged at the end of the fixed arm (7); the steering shaft (8) is connected to the top of the wheel bracket (10); a travel motor (11) is arranged at the bottom of the wheel bracket (10); the output end of the travel motor (11) is connected to the wheel (12); the steering oil cylinder (13) is arranged on the side of the fixed arm (7); the output end of the steering oil cylinder (13) is connected to one end of the steering cantilever (14); the other end of the steering cantilever (14) is connected to the wheel bracket (10) via a steering torque sensor (9); The steering loading assembly comprises a steering angle sensor (15), a steering loading oil cylinder (16) and a steering loading cantilever (17); the steering angle sensor (15) is arranged on the top of the steering shaft (8); the steering loading oil cylinder (16) is arranged on the side of the fixed arm (7); one end of the steering loading cantilever (17) is arranged on the wheel bracket (10); and the output end of the steering loading oil cylinder (16) is connected to the other end of the steering loading cantilever (17); The walking loading assembly is arranged on a mounting plate (3), and an output end of the walking loading assembly is drivingly connected to a wheel (12) for applying a walking loading force to the wheel (12); The ground contact loading assembly is located below the wheel (12) and contacts the wheel for applying a ground contact loading force to the wheel (12); The control center and the oil source power assembly cooperate with each other to provide power output to the travel motor (11), the steering cylinder (13), the steering loading cylinder (16), the travel loading assembly and the ground contact loading assembly and control the power output; The field scene and working condition data recording and playback component is used to record and playback field scenes and working conditions. The field scene and working condition data recording and playback component exchanges data with the control center in the same local area network through a switch.

2. The agricultural robot chassis field scene working condition simulation device according to claim 1, characterized in that: The field scene and working condition data recording and playback component includes a robot, which includes four wheel speed sensors, four steering angle sensors, a depth camera, an inertial attitude measurement unit, a multi-line laser radar sensor, a positioning and orientation terminal, and a high-performance computing processing center. The four wheel speed sensors are respectively installed on the four wheel shafts of the robot to measure the speed data of the wheel bodies in real time during field walking. The four steering angle sensors are installed on the four wheel steering shafts of the robot to measure the steering angle data of the wheel bodies in real time during field walking. The depth camera and the multi-line laser radar sensor are centrally installed at a position point, which is located at the front of the robot. The inertial attitude measurement unit is installed in the center of the robot chassis frame. The multi-line laser radar sensor collects three-dimensional point cloud data of the working scene. The depth camera collects image information of the forward direction. The inertial attitude measurement unit collects the attitude angle of the center point of the agricultural robot. The real-time collection of sensor data and the alignment of image and point cloud are realized on the high-performance computing processing center. The positioning and orientation terminal has two GPS antennas and is installed on the roof of the robot in a front-to-back layout. The positioning and orientation terminal collects the position information and heading angle of the robot. The software used for data collection is developed based on the ROS robot operating system, and the recorded data is saved as a bag file type.

3. The agricultural robot chassis field scene working condition simulation device according to claim 1, characterized in that: The walking loading assembly comprises a walking loading motor (18) and a walking sensor assembly. A walking loading bracket (4) and a sensor bracket (5) are provided on the mounting plate (3). The walking loading motor (18) is provided on the walking loading bracket (4). The walking sensor assembly is provided on the sensor bracket (5). The output end of the walking loading motor (18) is connected to a first cross universal joint (19). The first cross universal joint (19) is connected to the walking sensor assembly. The walking sensor assembly is connected to a second cross universal joint (22). The second cross universal joint (22) is connected to an electromagnetic telescopic coupling. The electromagnetic telescopic coupling is connected to a third cross universal joint (25). The third cross universal joint (25) is fixedly connected to the center point of the wheel (12). The oil source power assembly is connected to the walking loading motor (18).

4. The agricultural robot chassis field scene working condition simulation device according to claim 3, characterized in that: The walking sensor assembly comprises a walking torque sensor (20) and a speed sensor (21), wherein the speed sensor (21) is mounted on the walking torque sensor (20), the first cross universal joint (19) is connected to the end of the walking torque sensor (20), a sensor rotating shaft is arranged in the walking torque sensor (20), the sensor rotating shaft extends to the outer end of the walking torque sensor (20) and is connected to the second cross universal joint (22), the electromagnetic telescopic coupling comprises a telescopic sleeve (23) and a telescopic special-shaped shaft (24), the telescopic special-shaped shaft (24) is arranged in the telescopic sleeve (23), the telescopic sleeve (23) is connected to the second cross universal joint (22), and the telescopic special-shaped shaft (24) is connected to the third cross universal joint (25).

5. The agricultural robot chassis field scene working condition simulation device according to claim 1, characterized in that: The ground contact loading assembly comprises a ground contact force loading cylinder (27), a force sensor (28) is provided on the output end of the ground contact force loading cylinder (27), a contact disc (29) is provided on the force sensor (28), and the contact disc (29) is located below the wheel (12) and in contact with the wheel (12). The oil source power assembly is connected to the ground contact force loading cylinder (27).

6. The agricultural robot chassis field scene working condition simulation device according to claim 4, characterized in that: The ground contact loading assembly further comprises a displacement sensor (26), and the displacement sensor (26) is located below the connection position between the telescopic special-shaped shaft (24) and the wheel (12) through the third cross universal joint (25).

7. A method for using the agricultural robot chassis field scene working condition simulation device according to any one of claims 1 to 6 to perform chassis motion simulation, characterized in that: The steps include: S1: With the help of the robot, the four wheel speed sensors, four steering angle sensors, a depth camera, an inertial attitude measurement unit, a multi-line laser radar sensor, a positioning and orientation terminal, and a high-performance computing processing center are used to record field scene and working condition data, and the high-performance computing processing center is used to play back the data indoors to complete the three-dimensional scene mapping of the farmland; S2: With the help of the control center, field scenes and working conditions are simulated indoors through the wheel-leg assembly, steering loading assembly, walking loading assembly, ground contact loading assembly and oil source power assembly; S3: The staff compares and analyzes the acquired information with the wheel steering torque information, wheel torque information, expected speed and expected steering angle, operates the control center, and adjusts the control gain coefficient for controlling the steering and speed of the wheel (12) so that the wheel-leg assembly meets the requirements.

8. The method for using the agricultural robot chassis field scene working condition simulation device for chassis motion simulation according to claim 7 is characterized in that: The S1 specifically includes the following steps: S1.1: Measure the distance from each sensor to the coordinate origin using the center of the robot chassis as the coordinate origin, and publish the coordinate transformation matrix of each sensor to the chassis coordinate origin in the ROS system; S1.2: The high-performance computing processing center includes industrial computer 1, which starts the driver and data acquisition program for all sensors on industrial computer 1, collects data from each sensor at a frequency of 10 Hz, and saves the data in real-time in the form of a bag file to industrial computer 1. The robot's driving path traverses the entire field in a full-coverage manner. S1.3: The high-performance computing processing center also includes an industrial computer 2, which collects and transmits the recorded data to the industrial computer 2. The industrial computer 2 initiates a data playback program based on the ROS robot operating system, reads the contents of one or more bag files, and performs a first data playback in a time-synchronized manner, republishing the sensor data from the data packets. S1.4: The high-performance computing processing center has a built-in farmland 3D mapping program for performing lidar mapping and robot positioning. The program starts the 3D farmland scene lidar mapping and robot positioning program, constructs a 3D map of the farmland scene, and outputs the robot's trajectory points on the map. The center waits for the data packet playback to complete, and saves the farmland scene point cloud map and all robot movement trajectory data. S1.5: The second industrial computer has a pre-installed robot real-time simulation program, a sensor simulation program, a path planning program, and a sensor fusion positioning program. The robot real-time simulation program is started, a three-dimensional map of the farmland scene is loaded, a robot simulation model is loaded, the sensor simulation program is loaded, the path planning program is started, and the sensor fusion positioning program is started. S1.6: Launching the trajectory tracking software in the robot real-time simulation program on the second industrial computer. The trajectory tracking software subscribes to the robot's target path information and current position and posture information. The trajectory tracking algorithm calculates the robot's heading angle and forward speed in real time. S1.7: The second industrial computer also includes a robot real-time kinematics solver. The robot real-time kinematics solver is started, subscribes to the robot's heading angle and forward speed in real time, obtains the expected speed and expected steering angle of each wheel through inverse kinematic calculations, and publishes the results to the ROS robot operating system server in real time.

9. The method for using the agricultural robot chassis field scene working condition simulation device according to claim 8 to perform chassis motion simulation, characterized in that: The S2 specifically includes the following steps: S2.1: The control center is pre-installed with a wheel control program for controlling the travel motor (11) and the steering cylinder (13) through the oil source power assembly, and a loading control program for controlling the steering loading cylinder (16), the travel loading motor (18), and the ground contact force loading cylinder (27). The control center pre-sets wheel steering torque information and wheel torque information; after the data playback system is started, the control center receives the expected speed information and expected steering angle information published on the server of the ROS robot operating system in step S1.7; S2.2: The control center controls the travel motor (11) and the steering cylinder (13) by means of the wheel control program, so that the wheel (12) moves according to the desired speed and desired steering angle calculated in real time in step S1.7; S2.3: The control center controls the steering loading cylinder (16) by means of a loading control program. The steering loading cylinder (16) applies a steering loading force to the wheel bracket (10) via a steering loading cantilever (17), so that the wheel (12) is subjected to the steering loading force, so that the wheel (12) can obtain a simulated steering load environment. The steering torque sensor (9) measures steering loading torque information of the wheel (12) in real time, and the rotation angle sensor (15) measures rotation angle information of the wheel (12). S2.4: The control center controls the travel loading motor (18) by means of a loading control program. The travel loading motor (18) applies a travel loading force to the wheel (12) through a first cross universal joint (19), a travel torque sensor (20), a speed sensor (21), a second cross universal joint (22), a telescopic sliding sleeve (23), a telescopic special-shaped shaft (24), and a third cross universal joint (25), so that the wheel (12) can obtain a simulated travel load environment. The travel torque sensor (20) measures the travel loading torque information of the wheel in real time, and the speed sensor (21) measures the speed information of the wheel (12). S2.5; The control center controls the ground contact force loading cylinder (27) by means of a loading control program, and the ground contact force loading cylinder (27) applies a contact loading force to the wheel (12) through a contact disk (29), so that the wheel (12) can obtain a simulated ground contact load environment, the force sensor (28) measures ground contact loading force information, and the displacement sensor (26) measures displacement information of the telescopic special-shaped shaft (24) relative to the telescopic sliding sleeve (23) in real time; S2.6: The control center is connected to the high-performance computing processing center, the steering torque sensor (9), the angle sensor (15), the walking torque sensor (20), the speed sensor (21), the displacement sensor (26) and the force sensor (28) for information exchange. The steering torque sensor (9) transmits steering loading torque information to the control center, the angle sensor (15) transmits angle information to the control center, the walking torque sensor (20) transmits walking loading torque information to the control center, the speed sensor (21) transmits speed information to the control center, the displacement sensor (26) transmits displacement information to the control center, and the force sensor (28) transmits ground contact loading force information to the control center.

10. The method for using the agricultural robot chassis field scene working condition simulation device according to claim 9 to perform chassis motion simulation, characterized in that: The S2.1 specifically includes the following steps: S2.1.1: After the data playback system is started, the high-performance computing processing center publishes the robot status data to the ROS robot operating system server in real time, and the control center receives the data published by the ROS robot operating system server in real time via Ethernet.

Images