This paper mainly studies the tracking control algorithm of self-designed desert planting robot, aiming to solve the tracking problem in the process of desert planting robot. In this paper, a trajectory tracking control algorithm based on model prediction is designed according to the kinematics model of the robot designed by ourselves. The MPC controller is used to combine the expected trajectory output turning Angle and the linear velocity of the robot. Then, an electronic differential distribution algorithm is used to calculate the linear velocities of the two rear wheels respectively. In order to prevent the error caused by sliding of the robot in the complex terrain, an ADRC controller is introduced to accurately control the driving wheels. As for the point tracking problem in the working process of the robot, firstly, transition points are introduced to solve the inaccessible problem caused by the turning radius of the robot. Then, from the perspective of motion planning, the point tracking problem of the mobile robot is transformed into the trajectory tracking problem between the initial position and the expected position by using the parametric trajectory generation method based on time polynomial. In order to verify the effectiveness of the proposed control algorithm, a full field experiment was carried out in wasteland. The experimental results show that the tracking accuracy of the proposed control algorithm in desert areas is up to 15cm, and the average tracking accuracy of simulated desert planting points ranges from 3.16cm to 12.65cm, which can meet the planting accuracy required by desert planting and has high application value.
The most existing instance prediction methods often lack practical understanding and enhancement of bird's‐eye view (BEV) spatial‐temporal features, particularly distant features. To alleviate the above problems, we propose a robust BEV feature enhancement network (RO‐BEV). The proposed RO‐BEV includes two major components: the novel adaptive linearly increasing dividing (ALID) strategy and the hierarchical cross‐scale fusion (HCSF) module. Specifically, the ALID strategy can mitigate instance loss caused by insufficient features by providing a linearly increasing 3D representation to build a robust BEV feature. Then, the proposed HCSF module enhances spatial‐temporal BEV features in the latent space to predict vehicle instances by sampling from feature distributions at different scales and timestamps. Experimental results on the nuScenes dataset show that our RO‐BEV outperforms existing state‐of‐the‐art methods.