Seeder positioning and precise seeding control device based on visual inertial navigation

Abstract

The utility model provides a seeding machine positioning and precise seeding control device based on visual inertial navigation, which comprises a miniature industrial personal computer, a camera fixed on the miniature industrial personal computer, a seeding machine, a seeding machine control cabinet and a WIFI (Wireless Fidelity) serial server, and is characterized in that the miniature industrial personal computer is in line connection with the camera; the micro industrial personal computer is connected with the WIFI serial port server through WIFI, the WIFI serial port server is connected with a seeder control cabinet wire, the micro industrial personal computer is provided with a Ubuntu18.04 system, and the Ubuntu18.04 system is provided with an ROS system of a melodic version. The utility model provides a positioning and seeding device based on the combination of an ROS system and an embedded system, displacement information is calculated by utilizing vision and IMU information, and the positioning and seeding device has the capability of realizing real-time positioning and multi-information fusion in a complex environment.

Description

Technical Field

[0001] The utility model relates to the technical field of agricultural system positioning, and in particular to a seeder positioning and precise sowing control device based on visual inertial navigation. Background Art

[0002] my country's agricultural production is undergoing major historical changes, among which intelligent agricultural machinery is the top priority of the change and also the fuel for agricultural modernization. In 2021, the "Opinions on Comprehensively Promoting Rural Revitalization and Accelerating Agricultural and Rural Modernization" formally proposed to improve the independent research and development capabilities of agricultural machinery and equipment, support the research and development and manufacturing of high-end intelligent and hilly and mountainous agricultural machinery and equipment, and develop smart agriculture as the key to realizing agricultural and rural modernization. In February this year, the "No. 1 Document" also proposed: vigorously implement the action to make up for the shortcomings of agricultural machinery and equipment, improve the subsidy policy for the purchase and application of agricultural machinery, and open up a "green channel" for the identification of urgently needed and applicable agricultural machinery. The key to agricultural modernization is the modernization of agricultural science and technology, improving agricultural production efficiency, alleviating the problem of agricultural labor shortage, and promoting the transformation and upgrading of the agricultural industry.

[0003] At present, most agricultural positioning and navigation technologies still use mechanical methods such as measuring wheels, which have problems such as large positioning errors, resulting in sowing intervals that do not meet the accuracy requirements and low agricultural production efficiency. In addition, there are studies that apply global positioning system technology (GPS) to agricultural positioning, but agricultural working environments are mostly located in rural areas, with harsh working environments and poor GPS signals, which in turn lead to problems such as GPS being unable to work stably and signal data being lost. In addition, GPS is expensive and cannot be put into production on a large scale. Based on the continuous maturity of industrial machine vision technology, the utility model applies visual inertial navigation technology to agricultural positioning and navigation technology. Compared with traditional mechanical positioning and GPS positioning, it has the advantages of strong environmental adaptability, high accuracy and low price. Utility Model Content

[0004] Aiming at the defect of large positioning error of existing seed drill equipment, the utility model provides a seed drill positioning and precise seeding control device based on visual inertial navigation.

[0005] In order to solve the problems raised in the above-mentioned background technology, the utility model provides a seeder positioning and precise sowing control device based on visual inertial navigation, characterized in that the device includes a micro-industrial computer, a camera fixed on the micro-industrial computer, a seeder, a control cabinet fixed on the side of the seeder, and a WIFI serial port server connected to the control cabinet, the camera is fixed above the micro-industrial computer, the micro-industrial computer is connected to the camera line, the micro-industrial computer is connected to the WIFI serial port server via WIFI, the micro-industrial computer selects Zhanmei j1900, the micro-industrial computer is installed with Ubuntu 18.04 system, and the melodic version of ROS system is installed in the Ubuntu 18.04 system.

[0006] The camera is Inter RealSense D435i, which includes two infrared camera modules and one RGB camera module. The camera also includes an IMU module, and the IMU module is Bosch BMI055. The camera and IMU are connected to a data processing chip, and the data processing chip is connected to the micro industrial computer via a USB3.0 data cable.

[0007] The seeder is LOVOL-2BMXE-2A, which is a two-row seeder including a furrow opener, a seed meter, and a pressing wheel, and is an actuator.

[0008] The control cabinet chip uses the stm32f103c8t6 single-chip microcomputer. The control cabinet is connected to control various actuators required for sowing by the seeder, and completes data communication with the WIFI serial port server in Modbus TCP mode through the USART serial port.

[0009] The WIFI serial port server includes a core processor and an external antenna. The WIFI serial port server uses Hanfeng HF7221-0. The WIFI serial port server can transmit WIFI signals. The micro industrial computer is connected to the WIFI of the WIFI serial port server to achieve the effect of transmitting data.

[0010] Compared with the prior art, the beneficial effects of the utility model are:

[0011] The utility model provides a seeder positioning and precise sowing control device based on visual inertial navigation. By deeply integrating and fusing machine vision, embedded systems, digital communications and other technologies in multiple directions and fields, a precise positioning system based on visual inertial navigation is established to achieve real-time background operation and improve work efficiency. The positioning system provided by the utility model has low environmental requirements, and the required signal acquisition is only the camera-captured image and the acceleration and angular velocity of the system itself. There is no problem of poor technical adaptability such as unstable GPS signals. The utility model combines embedded software and hardware development with network communication technology to achieve data transmission stability and improve the practicality and convenience of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 It is a three-dimensional diagram of the utility model;

[0013] Figure 2 It is a structural schematic diagram of the connection of various components of the utility model;

[0014] Figure 3 The utility model is implemented as a work flow chart. DETAILED DESCRIPTION

[0015] The utility model is further described below in conjunction with the accompanying drawings and specific implementation methods.

[0016] like Figure 1 As shown, a visual inertial navigation based seeder positioning and precision seeding control system is characterized in that the system includes a micro industrial computer, a camera fixed on the micro industrial computer, a seeder, a control cabinet fixed on the side of the seeder, and a WIFI serial port server connected to the control cabinet.

[0017] Figure 2 This is a structural diagram of the connection of each component of the utility model. The camera module is InterRealSense D435i camera, which includes dual infrared camera modules, RGB camera modules and Bosch BMI055 IMU modules; the micro industrial computer is Zhanmei j1900 micro industrial computer, which includes a central processor Interi5-4200Y module, dual USB3.0 interface module, and dual-band antenna module; the WIFI serial port server module is Hanfeng HF7221-0, which includes RJ45 network interface, external RS485 interface module, and signal transmission module; the chip of the seed drill control cabinet is stm32f103c8t6 single-chip microcomputer, which includes a processor core module, a serial communication module and an interrupt controller module.

[0018] Figure 3 The specific workflow for the implementation of the utility model includes the following steps:

[0019] 1. Initialize and calibrate the camera and the IMU respectively, record ROS data packets, and use Kalibr and imu_util calibration tools to obtain the intrinsic parameters of the camera, the accelerometer bias, gyroscope bias and scale factor of the IMU, and then jointly calibrate the camera and the IMU to obtain the relative rotation matrix, time offset, etc. between the camera and the IMU.

[0020] 2. The camera collects field ridge image information and transmits the data to a micro-industrial computer through a USB3.0 module. The micro-industrial computer extracts feature points using the Shi-Tomasi algorithm and sets appropriate thresholds to prevent redundant and repeated recognition of feature points. The pyramid-LK forward and reverse optical flow method is then used to track the feature points, match the feature points, set matching thresholds to prevent mismatching, and estimate the motion state based on the pixel values ​​of the same feature points between images.

[0021] 3. The IMU records the data changes of the accelerometer and gyroscope, and obtains the position movement information and attitude change information of the IMU by integrating and solving the six-dimensional data output by the IMU. The system motion status is calculated by combining the two, and the motion data is transmitted to the micro industrial computer through the USB3.0 module.

[0022] 4. Based on the calibrated parameters, the camera and IMU data are bundled and adjusted to construct an error function, which is then fused in a tightly coupled manner using the least squares nonlinear optimization method. The fused data is further optimized globally using loop closure detection and global optimization to obtain the system's x, y, and z axis motion in the world coordinate system and the changes in the quaternion representing the posture.

[0023] 5. According to the actual situation of agricultural seeding, the three-dimensional data is processed, the data is projected into a two-dimensional plane, and the cumulative travel distance of the seed drill is calculated.

[0024] 6. The micro industrial computer is connected to the WIFI signal transmitted by the WIFI serial port server. The micro industrial computer is configured with the specified length, baud rate and other information, completes the signal communication with the WIFI serial port server, and transmits the calculated cumulative travel distance to the WIFI serial port server.

[0025] 7. The internal chip stm32f103c8t6 of the seed drill control cabinet is connected to the WIFI serial port server via RS485 in Modbus TCP data format to read the cumulative travel distance.

[0026] 8. The stm32f103c8t6 single-chip microcomputer inside the seed drill control cabinet determines the read cumulative travel distance and controls the seed drill to execute the seeding instruction when the predetermined distance interval is reached.

Claims

1. A device for positioning and precise seeding of a seed drill based on visual inertial navigation, characterized in that: The device comprises a micro industrial computer (1), a camera (2) fixed on the micro industrial computer (1), a sowing machine (3), a furrow opener (4) on the sowing machine, a seed meter (5), a pressing wheel (6), a sowing machine control cabinet (8) fixed on the side of the sowing machine (3), and a WIFI serial port server (7) connected to the control cabinet (8); the camera (2) is fixed above the micro industrial computer (1); the micro industrial computer (1) is connected to the camera (2) by wire; the micro industrial computer (1) is connected to the WIFI serial port server (7) via WIFI; the micro industrial computer (1) is a J1900; the micro industrial computer (1) is installed with an Ubuntu 18.04 system; and the melodic version of the ROS system is installed in the Ubuntu 18.04 system.

2. According to claim 1, a device for positioning and precise sowing of a seeder based on visual inertial navigation is characterized in that: The camera (2) is Inter RealSense D435i, which includes two infrared camera modules and one RGB camera module. The camera also includes an IMU module, and the IMU module is Bosch BMI055. The camera and the IMU are connected to a data processing chip, and the data processing chip is connected to the micro industrial computer (1) via a USB3.0 data cable.

3. The device for positioning and accurately sowing a seed drill based on visual inertial navigation according to claim 1, characterized in that: The seeder (3) is LOVOL-2BMXE-2A, and is a two-row seeder, comprising a furrow opener (4), a seed meter (5), and a pressing wheel (6), and the seeder (3) is an actuator.

4. The device for positioning and accurately sowing a seed drill based on visual inertial navigation according to claim 1, characterized in that: The control cabinet (8) uses a stm32f103c8t6 single-chip microcomputer as the chip. The control cabinet (8) is connected to various actuators of the sowing machine (3) to control sowing, and data communication is completed in Modbus TCP mode via a USART serial port and a WIFI serial port server (7).

5. The device for positioning and accurately sowing a seed drill based on visual inertial navigation according to claim 1, characterized in that: The WIFI serial port server (7) comprises a core processor and an external antenna. The WIFI serial port server (7) uses Hanfeng HF7221-0. The WIFI serial port server (7) can transmit WIFI signals. The micro industrial computer (1) is connected to the WIFI of the WIFI serial port server (7) to achieve the effect of transmitting data.

Images