A Novel Control Strategy for Offset Points Tracking in the Context of Agricultural Robotics

TL;DR

The paper addresses the need to control an offset point on a rigidly attached agricultural implement rather than the vehicle center, to achieve precise field operations. It introduces two control strategies: (1) a direct extension of classical lateral-deviation control for the implement, and (2) a backstepping-based controller that directly targets the offset point with a two-stage design for exponential convergence. Modeling, analysis, and real-world experiments show that center-based control is inadequate for offset-point tasks and that the backstepping approach provides theoretical guarantees and often faster convergence, reducing lever-arm-induced errors. The work highlights the importance of tool-position effects in trajectory tracking and points to predictive control and multi-point coordination as directions for future enhancements in agricultural robotics.

Abstract

In this paper, we present a novel method to control a rigidly connected location on the vehicle, such as a point on the implement in case of agricultural tasks. Agricultural robots are transforming modern farming by enabling precise and efficient operations, replacing humans in arduous tasks while reducing the use of chemicals. Traditionnaly, path_following algorithms are designed to guide the vehicle's center along a predefined trajetory. However, since the actual agronomic task is performed by the implement, it is essential to control a specific point on the implement itself rather than vehicle's center. As such, we present in this paper two approaches for achieving the control of an offset point on the robot. The first approach adapts existing control laws, initially inteded for rear axle's midpoint, to manage the desired lateral deviation. The second approach employs backstepping control techniques to create a control law that directly targets the implement. We conduct real-world experiments, highlighting the limitations of traditional approaches for offset points control, and demonstrating the strengths and weaknesses of the proposed methods.

Figures (5)

• Figure 1: Agricultural robot equipped with a mounted implement. In this situation, the main focus is on the implement rather than the robot itself, as precision is required at the point where the implement interacts with the ground, thus making standard control laws of the robot's center inadequate.
• Figure 2: Notations used in the paper. The robot is assumed to follow a bicycle model. Contrary to standard robotics control problem, the goal is not to make the center $O$ of the robot to converge to the trajectory, but rather an offset point $T$ that represents in our case the implement (tool) of an agricultural robot.
• Figure 3: Reference trajectory used in the experiments. It consists of three straight lines (L1, L2, L3) and two curves (C1, C2) of different curvatures.
• Figure 4: Implement lateral error of a classical (control of the center point $O$), desired deviation, and backstepping approaches. Left: Lateral error as the function of the curvilinear abscissa. The lines (L1, L2, L3) and curves (C1, C2) correspond to the areas highlighted in \ref{['fig:Expe_setup']}. Right: Overall distribution of the errors for each method, with the median and quartiles at 25% and 75% represented as box plots.
• Figure 5: Distributions of the absolute deviation error as a function of the implement's longitudinal offset $T_s$. The box plots represent the medians with quartiles at 25% and 75%.