Search

• Developed a wheat cut-edge detection method based on stereo vision to provide support for the automatic navigation of combine harvesters. • Designed a classification method for crop areas based on the density-based spatial clustering of applications with noise method. • Designed an extraction method for wheat edge feature points based on low-pass filtering to eliminate the interference from the ridge in the yield. A cut-edge detection method for wheat based on stereo vision was proposed in this work to obtain the navigation path of a combine harvester. First, the point cloud was acquired by the stereo camera. The crop area was extracted with the threshold obtained by the Otsu method. Then, the point cloud of the crop area was gridded. The grids were clustered by the density-based spatial clustering of applications with noise method to classify the different crop areas. After filtering the noise caused by the ridge in the yield, the grids of interest were extracted. The edge point was extracted in each grid of interest. The polynomial fitting method was then used to acquire the straight or curved cut-edge. A total of 300 images were selected for the test of crop area extraction and crop areas classification. The results showed that the success rate of crop area extraction was 93.7% and the success rate of crop areas classification was 91.1%. 100 images were selected to extract the edge points and compare with the true value of manual measurement. Experiment results showed that the average deviation of the edge points was 8.47 cm, the maximum deviation was 23.1 cm, and the standard deviation was 5.97 cm. The proposed method is thus capable of providing support for the automatic navigation of combine harvesters. [ABSTRACT FROM AUTHOR]

• Global path conflict detection. • Conflict resolution strategy. • Time window generation. Path planning is one of the crucial problems of multi-machine cooperative navigation. In the context of regional farmland operation, multiple agricultural machines often need to complete multiple tasks together. Studies on global path conflict detection algorithms based on topographic maps and time windows have been conducted to solve the global path conflict problem of multiple agricultural machinery cooperation in the farmland operation environment. First, the global path conflict problem of multiple agricultural machinery cooperation was analyzed. The constraints of time window conflict detection and the time parameters to be considered in calculating the time window were proposed. Then, the global path preplanning was performed based on a topographic map and Dijkstra algorithm. According to the path preplanning results, the global path conflict was detected based on the time window. Finally, the global path conflict detection algorithm was simulated on MATLAB with the topographic map of Zhuozhou Experimental Farm as an example. The global path preplanning results show that the four paths overlapped in different degrees at nodes 1, 2, 7, 11, 15, 20, and 27. Path 3 and Path 1 were inclusive interval-type conflicts, and Path 4 and Path 1 were node-type conflicts. Path 3 adopted the change path strategy after the conflicting paths were replanned. The path was changed from [15-11-7-2-1-5-10-14-18-31] to [15-21-24-18-31]. Path 4 adopted the waiting strategy, and its paths did not change. Although the four paths still had different degrees of path overlap, there were no time conflict in the overlapping part of the replanned path. The simulation results show that the global path optimization and path conflict detection of multiple agricultural machinery cooperation could be realized based on the topographic map and time window. Moreover, the conflict resolution strategy with the least time could be obtained to achieve a safe, efficient, and conflict-free global path of multiple agricultural machinery cooperation. Thus, a foundation for further solving the path planning of multiple agricultural machinery cooperation in the complex operation environment of regional farmland is laid. [ABSTRACT FROM AUTHOR]

• This paper reviewed applications of machine vision in agricultural robot navigation from 2017 to 2021. • This paper summarized the challenges that machine vision needs to overcome in agricultural robot navigation. • This paper introduced development directions of machine vision in agricultural robot navigation. Many tasks in smart agriculture have further requirements for the autonomous navigation of agricultural robots. Due to irreplaceable visual information and low-cost hardware costs, machine vision (MV) is widely used in agricultural robot navigation. This paper reviewed applications of MV in agricultural robot navigation. First, the advantages, disadvantages and roles of different vision sensors and MV algorithms in agricultural robot navigation were introduced. And then, the development status of MV in agricultural robot navigation was reviewed. Finally, the challenges of MV in agricultural robot navigation were discussed. To address these challenges, this paper looks forward to the development directions of future research. [ABSTRACT FROM AUTHOR]

• The command intelligent scheduling model of harvester and grain truck is established. • The global optimal scheduling problem is solved based on NSGA-III genetic algorithm. • Local obstacle avoidance shortest path planning is carried out based on the improved ant colony algorithm. • The optimal scheduling problem of multi machine cooperative operation between harvester and grain truck is solved. Aiming at the problems of over-reliance on traditional manual harvest operation experience and lack of dispatching planning scheme under the environment of insufficient grain trucks in agricultural harvest scenario. A multi-objective combined optimization intelligent scheduling model of harvesters and grain transport vehicles was established to save agricultural machinery resources, and improve the overall efficiency of harvesters and transport vehicles, to meet the constraints of plot location, task number, operation time window, and path planning. The model was solved by an intelligent search algorithm, and an intelligent scheduling method of command multi-machine cooperative operation based on NSGA-III and an improved ant colony algorithm was proposed. When the harvester is full loaded, there is non harvest area on one side of the grain unloading cylinder, there the grain truck can not reach and unload cooperatively. To solve this problem, the prediction method of early unloading point was designed and further improved by determining the workload constraints and time constraints. When solving the model, the reference point setting in NSGA-III can ensure the diversity of the population. And then the ant colony algorithm that sets the detour vertex instead of the grid map is used to solve the optimal command call time and the optimal unloading position, and then the dynamic time window and the local dynamic obstacle avoidance path of the transport machine are divided. Finally, the designed model can output an accurate scheduling planning scheme with deployment information. It can be seen from the simulation test that the scheduling model established, and the solution algorithm designed in this research can obtain many groups of good optimization results. Moreover, it can make the unproductive waiting time of harvester 0, and minimize the transfer distance and number of grain transporters. This method can be combined closely with the practical application of unmanned farms, which lays a theoretical foundation for improving the efficiency of multi-machine cooperative operations. [ABSTRACT FROM AUTHOR]

• A feed rate monitoring system for combine harvesters is designed. • A method based on BPNN and optimized PSO is proposed to achieve feed rate monitoring. • An estimation method for rice yield distribution through calculating the feed rate and stem-to-grain ratio is proposed. • The relationship of header height and stem-to-grain radio and the relationship of grain moisture content and stem-to-grain radio are verified. • A binary model between the header height, the grain moisture content and the stem-to-grain ratio of the rice fed to the harvester was established. Real-time monitoring of harvester feed rate is of great significance for guiding harvesting work and improving harvesting efficiency. In this study, a feed rate monitoring system is designed and developed. The system consists of a torque sensor for measuring the header power shaft torque, an angel sensor for measuring header height, a humidity sensor for measuring grains moisture content, a Global Navigation Satellite System receiver, and a vehicle industrial computer. Data, such as the moisture content of grains and the working position of the harvester, are collected by the system during harvest. The height of the header, the moisture content of grains, and the torque of the header power shaft are selected as the input quantity, and feed rate is selected as the output quantity. A calculation model for feed rate is established using a particle swarm optimization–back propagation neural network, and then real-time monitoring of feed rate is realized. On the basis of the monitoring of feed rate, an estimation method for yield distribution information is proposed. Through analyzing the rice samples in the experimental area, the method establishes a binary linear model between the header height, the grain moisture content, and the stem-to-grain ratio of the rice fed to the harvester. The model is combined with the monitoring results of feed rate to obtain the rice yield data fed to the harvester. Then, the rice yield data are matched with the positional information at the time of harvesting to obtain a yield distribution map of the harvested area. Field experiments are conducted in a rice-growing area in Northeast China, and the average relative error of the feed rate monitoring system is 7.69%. The field sample verification accuracy of the stem-to-grain ratio model is higher than 90%, and the correlation coefficient R2 is 0.93. The estimation accuracy of the yield distribution is higher than 90%, which provides a scientific basis for agricultural decision making. [ABSTRACT FROM AUTHOR]

• Yield monitoring system design. • Yield reconstruction model. • Analysis of spatiotemporal variability. An impact-type grain yield monitoring system was developed and tested in laboratory and field to meet the requirement of precision agriculture in China. The system consisted of a mass flow sensor, a clean elevator speed sensor, a ground speed sensor, a moisture content sensor, a controller area network-bus processor, a global navigation satellite system receiver, and an industrial personal computer. The system was used to collect yield data and create the grid map in a field of winter wheat from 2018 to 2019. Laboratory and field calibrations were implemented to verify the accuracy of the mass flow sensor and the moisture content sensor. The errors of time delay and parking were removed, and exponential smoothing was performed in the data preprocessing. A yield reconstruction method was proposed to calculate the yield data of the sampling points based on the empirical model between flow voltage and grain weight. The predicted errors of total yield in both years were less than 5%, which indicated that the accuracy of yield monitoring system was high. An averaging method was used to generate the yield grid map. The spatiotemporal variation of the two-year yield was analyzed. Hierarchical management for precision farming was studied according to spatial variability ( CV S = 31%) in 2018. Results showed that the temporal variability was CV T = 34%, which confirmed that the yield increased by about 0.3 t compared with the previous year. Moreover, the variability of the grid map showed the distribution of the yield reduction area, which was related to the characteristics of the seeds and the regional properties of the soil. The results proved that the developed yield monitoring system can be used as a tool to provide a reference for precision agriculture and high-yield management of winter wheat. [ABSTRACT FROM AUTHOR]

• The FIR filter is designed to adapt to the multiorder motion state. • The fuzzy controller is designed to enhance the ability to adjust abnormal noise. • The forgetting factor is designed to reduce the dependence on the initial value. • The quaternion is designed to solve the nonlinear problem of position coordinates. This study uses global navigation satellite system (GNSS) positioning equipment and inertial measurement unit integrated with accelerometer and gyroscope to improve the accuracy and stability of the current agricultural machinery automatic navigation technology. Considering the actual motion state of agricultural machinery in operation, a fuzzy adaptive finite impulse response Kalman filter (FA-FIR-KF) algorithm was proposed to integrate position information and attitude information, and some necessary auxiliary optimization algorithms were introduced to make innovative improvements. The introduction of quaternion method can suppress the actual nonlinear problem of the agricultural machinery coordinate caused by the attitude angle. A fuzzy inference system was adopted to improve the adaptive adjustment ability to abnormal noise. A forgetting factor was adopted to reduce the system's excessive dependence on prior statistical information, so that the system can quickly track the abrupt signal. The algorithm simulation program was written on MATLAB, and the performance and effect of the proposed algorithm were verified through simulation and farm experiments. Simulation results of artificially added noise simulation data show that the localization precision in the X n , Y n , and Z n directions increases by 38.95%, 38.88%, and 32.99%, respectively. This finding indicates that the FA-FIR-KF algorithm can effectively suppress the Gaussian white noise of the GNSS received signal and improve the positioning accuracy of agricultural machinery. The reliability of this algorithm applied to the automatic navigation system was verified through a tractor straight-line navigation experiment. The tractor conducts an automatic navigation test at a speed of 0.8 m/s. Under the GNSS differential state, the average error and root mean square error (RMSE) are 1.074 and 1.396 cm in filtering case and 1.17 and 1.551 cm in nonfiltering case, respectively. Under the GNSS nondifferential state, the average error and RMSE are 2.097 and 2.72 cm in filtering case and 3.663 and 4.633 cm in nonfiltering case, respectively. Compared with the nonfiltering case, the average error and RMSE reduce by 8.21% and 9.99% in the differential state and 42.75% and 41.32% in the nondifferential state, respectively. Test results show that the proposed algorithm can make the agricultural machinery track the desired path more smoothly, stably, and accurately than in the nonfiltered case, and the tracking accuracy is at the centimeter level. [ABSTRACT FROM AUTHOR]

• Task assignment model. • Dynamic and static simulation. • Crop harvesting simulation. Task assignment is a key problem in multi-machine cooperative navigation. In the context of regional farmland operation, multiple agricultural machines often need to complete multiple tasks together. In order to realize the management of multiple agricultural machinery cooperation, studies on task assignment based on the improved ant colony algorithm have been conducted under the farmland operation environment. First, a task assignment model of multiple agricultural machinery cooperation was established by combining dynamic and static task assignments. Then, according to the task assignment model, the task assignment process based on the improved ant colony algorithm was established while considering the match between supply and demand, the operation capacity of the agricultural machinery, and the operation cycle and path cost. Finally, the dynamic and static task assignments of multiple agricultural machinery cooperation based on the improved ant colony algorithm were simulated on MATLAB. Taking the crop harvesting experiment as an example, according to the actual farmland location information of the Zhuozhou Experimental Farm, the different (agricultural machinery, task) combinations were set, and the task assignment results were compared and analyzed. Results showed that the path costs of harvester and grain transporters were reduced by 51.27% and 22.00% respectively, When the quantities of tasks were set to 11, indicating that the improved ant colony algorithm can effectively reduce the path cost. When the quantities of tasks were set to 5, 11, 16 and 22, the average operation cycles were shortened by 67.32%, 37.50%, 55.95%, and 56.37% respectively. The problem of "nearby" in the task assignment was solved to a certain extent, the overload of some agricultural machinery and the idle of other agricultural machinery were avoided, and the operation cycle was shortened. At the same time, based on the static task assignment, the dynamic task assignment was realized in the two scenarios of new tasks and malfunctioning harvesters, thus laying a foundation for further solving the scheduling management problem of multiple agricultural machinery cooperation under a complex farmland operation environment. [ABSTRACT FROM AUTHOR]