Force Aware Branch Manipulation To Assist Agricultural Tasks

TL;DR

This work addresses the challenge of manipulating natural, deformable branches in agricultural settings under occlusion by proposing a geometry-guided, force-aware motion-planning framework. A 2D differential-geometry-based branch model informs a constrained RRT* planner, while online force feedback triggers re-planning to keep contact forces below a defined threshold. The approach does not require individual branch parameter tuning, instead relying on safety zones and force-based replanning to generalize across branch geometries. Experimental results with a UR5 and force sensor demonstrate substantial force reduction and a 39/50 success rate under force constraints, indicating practical potential for improving tasks such as pollination, pruning, and harvesting in canopy environments.

Abstract

This study presents a methodology to safely manipulate branches to aid various agricultural tasks. Humans in a real agricultural environment often manipulate branches to perform agricultural tasks effectively, but current agricultural robots lack this capability. This proposed strategy to manipulate branches can aid in different precision agriculture tasks, such as fruit picking in dense foliage, pollinating flowers under occlusion, and moving overhanging vines and branches for navigation. The proposed method modifies RRT* to plan a path that satisfies the branch geometric constraints and obeys branch deformable characteristics. Re-planning is done to obtain a path that helps the robot exert force within a desired range so that branches are not damaged during manipulation. Experimentally, this method achieved a success rate of 78% across 50 trials, successfully moving a branch from different starting points to a target region.

Figures (5)

• Figure 1: Humans often use one hand to grasp the branch for better accessibility, while the other hand is used to perform primary tasks like (a) branch pruning and (b) hand pollination of the flower. (c) An overview of our approach, where one robot manipulates the branch to move the flower to the field of view of another robot by planning a force-aware path.
• Figure 2: (a) Geometric modeling of deformable linear objects (DLO) provides the general idea which branch configuration are safe to manipulate and which ones are prone to break the branches. (b) Guided by the deformable model, RRT* can sample configurations efficiently to generate the path to the goal region so that the path is safe to manipulate the branch. (c) With the help of force feedback, the path is re-planned to another goal point that is within the field of view of a camera attached to a black robot arm such that the force remains below a certain threshold to prevent branches from getting damaged
• Figure 3: Geometric modeling provides knowledge of any random endpoint constraints' manipulation feasibility and safety level.
• Figure 4: (a) When moving a branch without branch model information or force feedback, significant manipulation force (100 N) occurs. (b) Introducing branch model information reduces the manipulation force to below 60 N (c). Utilizing both branch information and force feedback maintains the applied force below the 40 N Force Threshold.
• Figure 5: (a) Path with orientation to manipulate branch from start configuration to goal region. (b) Among the different planned path to reach the goal, the planner followed the dotted line which minimized the force exerted on branch.