Optimizing the Performance of a Wheeled Mobile Robot for Use in Agriculture

Amertet, Sairoel; Gebresenbet, Girma; Mohammed Alwan, Hassan · 2024 · Crossref

DOI: 10.5772/intechopen.1008161

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the challenge of optimizing path tracking and stability for wheeled mobile robots (WMRs) in agricultural environments. The authors identify that while WMRs are essential for precision agriculture, their inherent instability and susceptibility to disturbances make standard proportional-integral-derivative (PID) controllers inadequate. To overcome these limitations, the study investigates the application of Linear-Quadratic Regulator (LQR) control, a strategy that minimizes a quadratic cost function to balance state deviations and control effort. The research aims to determine the suitability of LQR for a four-wheel skid-steering mobile robot, specifically analyzing how different weighting matrices affect system performance. The methodology involves developing a kinematic model for a four-wheel skid-steering mobile robot designed for agricultural use. The authors justify the four-wheel configuration over single, double, or three-wheel designs based on superior stability, route efficiency, and control capabilities. The kinematic model assumes planar motion, minimal wheel distortion, and smooth rolling, deriving relationships between the robot’s center of mass, instantaneous center of rotation, and individual wheel velocities. The study validates the model’s controllability and observability using state-space representations. The control system design utilizes LQR theory, where the performance index is defined by state penalty matrix Q and control penalty matrix R. The authors analyze three specific scenarios based on the trade-offs between state and control costs: only non-zero state expensive; control expensive with non-zero state cheap; and control cheap with non-zero state expensive. The results demonstrate that the LQR controller effectively stabilizes the system by adjusting the weighting matrices. Specifically, under the "cheap control" scenario, the system achieved a peak time of 0.1 seconds, a settling time of 7.82 seconds, and a rising time of 4.39 seconds. The analysis confirms that while the system is nonholonomic, it is stabilizable and detectable, allowing the LQR algorithm to manage unstable modes. The study highlights that selecting appropriate values for Q and R is critical, as small R values allow for stabilization with minimal control effort, whereas large R values penalize control effort more harshly. The significance of this work lies in providing a robust control framework for agricultural robotics, which is crucial for addressing global challenges such as food security and sustainability. By optimizing the control algorithm, the research supports the development of autonomous systems capable of navigating complex terrains with high precision. This contributes to the broader field of agricultural robotics by offering a method to enhance stability and reduce the need for manual intervention, thereby increasing productivity and enabling precision farming practices.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• steering pattern