Imagen de Portada

Design and Experiment of a Greenhouse Autonomous Following Robot Based on LQR–Pure Pursuit

Guardado en:
Publicado en:Agriculture vol. 15, no. 15 (2025), p. 1615-1646
Autor principal: Hu, Yibin
Otros Autores: Jieyu, Xian, Xiao Maohua, Cheng Qianzhe, Chen, Tai, Zhu Yejun, Geng Guosheng
Publicado:
MDPI AG
Materias:
Shenzhen China
China
Humidity
Agricultural production
Algorithms
Deformation
Structural analysis
Robots
Path tracking
Motion control
Crops
Deformation analysis
Pursuit tracking
Localization
Tracking errors
Simulation analysis
Energy consumption
Radiation
Greenhouses
Efficiency
Accuracy
Agriculture
Linear quadratic regulator
Control algorithms
Planning
Root-mean-square errors
Sensors
Environmental
Acceso en línea:Citation/Abstract
Full Text + Graphics
Full Text - PDF
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
Resumen:Accurate path tracking is crucial for greenhouse robots operating in complex environments. However, traditional curve tracking algorithms suffer from low tracking accuracy and large tracking errors. This study aim to develop a high precision greenhouse autonomous following robot, use ANSYS Workbench 19.2 to perform stress and deformation analysis on the robot, then propose a path tracking method based on Linear Quadratic Regulator (LQR) to optimize the pure tracking to ensure high precision curved path tracking for curved tracking, finally perform a comparative simulation analysis in MATLAB R2024a. The structural analysis shows that the maximum equivalent stress is 196 MPa and the maximum deformation is 1.73 mm under a load of 600 kg, which are within the yield limit of 45 steel. Simulation results demonstrate that at a speed of 2 m/s, the conventional Pure Pursuit algorithm incurs a maximum lateral error of 0.3418 m and a heading error of 0.2669 rad under high curvature conditions. By contrast, the LQR–Pure Pursuit algorithm reduces the peak lateral error to 0.0904 m and confines the heading error to approximately 0.0217 rad. Experimental validation yielded an RMSE of 0.018 m for lateral error and 0.016 m for heading error. These findings confirm that the designed robot can sustain its payload under most operating scenarios and that the proposed tracking strategy effectively suppresses deviations and improves path-following accuracy.
ISSN:2077-0472
DOI:10.3390/agriculture15151615
Fuente:Agriculture Science Database