Agricultural environment mapping is the premise of the autonomous navigation of agricultural robots. Due to the undulating terrain and chaotic environment, it is challenging to accurately map the environmental maize field using existing LOAM (LiDAR odometry and mapping) methods. This paper proposes a LOAM method based on maize stalk semantic features for 6-DOF (degrees of freedom) pose estimation and field mapping with agricultural robots operating in a dynamic environment. The piecewise plane fitting method filters the ground points for the complex farmland terrain. To eliminate the unstable factors in the environment, we introduce the semantic information of maize plants into the feature extraction. The regional growth method segments the maize stalk instances, the instances are parameterized to a line model, and the optimization method calculates the pose transformation. Finally, the mapping method corrects the drift error of the odometry and outputs the maize field map. This paper compares our method with the GICP and LOAM methods. The trajectory relative errors of our method are 0.88%, 0.96%, and 2.12%, respectively, better than other methods. At the same time, the map drawn by our method has less ghosting and clearer plant edges. The results show that our method is more robust and accurate than other methods due to the introduction of semantic information in the environment. The mapping of corn fields can be further used in precision agriculture.

The auto-follow row technology is the basic technology to realize the uncrewed operation of corn harvesters, which helps to improve the quality and efficiency of corn harvest. A deviation detection sensor and an auto-follow row system are designed to address the difficulty for corn harvesters to harvest the corn sowing with different row spacing. The auto-follow row system is comprised of the auto-follow row controller, two deviation detection sensors, sensors for acquiring the state of the corn harvester, and the hydraulic steering system. The deviation detection sensor uses the touch rod to detect the lateral deviation and sends it to the auto-follow row controller through the Controller Area Network (CAN) bus. The auto-follow row controller obtained the lateral deviation, calculated the target steering angle according to the pure pursuit method based on the yaw angle estimation model and fuzzy control model, and adjusted the posture of the harvester by controlling the steering actuator. Two types of corn harvesters were reformed and mounted the auto-follow row system for simulation and field tests. The average errors of the auto-follow row system in the simulation tests and the field tests were 0.040 m and 0.046 m, respectively. The relative deviations within ±0.075 m were 98.33% and 97.22%, respectively, which met the operating requirements of corn harvesters. The test results show that the auto-follow row system can be applied to many kinds of corn harvesters and provide a reference for designing a completely unmanned system for corn harvesters.

The driving tasks of a maize harvester are exhausting because of the varying soil, maize conditions, and the long working time. Operators must adjust and optimize the internal settings of the harvester to modify the working parameters and reduce the harvest loss. In this paper, an intelligent control system for maize picking harvest was investigated for automatic adjustment and minimization of maize picking losses. A prediction model based on experimental data was used to predict the maize picking loss rate, and the rotational speed of the pulling rollers, operating speed, and header height were optimized to minimize the maize picking loss. The intelligent control system allows use of manual or automatic controls; the controller adjusts the rotational speed of pulling rollers, the operating speed, and header height based on the measured picking losses in automatic mode. The designed automatic control system comprises faster and slower loops. The fuzzy proportional–integral–derivative (PID) method is used to optimize the rotational speed of pulling rollers and the operating speed in the faster loop, and the PID method is used to regulate the header height in the slower loop. Field experiments were conducted to evaluate the effectiveness and stability of the system. The system test results showed that all working parts respond quickly, and the overshoot and steady-state errors of each working part were relatively small. Regardless of the load condition, the established control strategy could optimize each working parameter of the maize harvester. In experiments, maize picking loss rates of 1.676% and 1.386% were obtained, which meet the requirements of maize harvesting.