A multi-armed robot for assisting with agricultural tasks

2026-03-27 · Source: ΑΙhub · Field: Agriculture & Food Systems — Precision Agriculture & Smart Farming, Robotics & Autonomous Systems · Depth: Advanced, medium

Summary

Madhav Rijal and colleagues presented a methodology at IROS 2025 for safely manipulating plant branches using a multi-armed robotic system, addressing the challenge of occluded flowers or fruits in agricultural tasks like pollination and harvesting. Their approach combines RRT*-based motion planning with real-time force feedback to prevent branch damage. The system models branches as deformable linear objects, using a geometric heuristic to guide planning and identify safer configurations. During execution, if measured force exceeds a predefined threshold (e.g., 40 N), the robot re-plans its motion online to find an alternative path that reduces stress. Experiments with five different branch poses, each tested 10 times, achieved a 78% success rate across 50 trials, demonstrating that force-aware replanning reduced interaction forces from over 100 N to below the 40 N threshold, ensuring task feasibility and branch safety.

Key takeaway

For robotics engineers developing agricultural automation, this research demonstrates a robust method for handling delicate biological structures. You should consider integrating force-aware motion planning with online replanning capabilities into your systems to minimize damage to plants while maximizing task success rates. This approach is crucial for tasks like precision pollination or harvesting where targets are often occluded and require careful branch manipulation.

Key insights

Robots can safely manipulate delicate plant branches using force-aware motion planning and real-time feedback.

Principles

• Combine geometric heuristics with force feedback.
• Online replanning enhances safety and adaptability.

Method

The method uses an RRT*-based planner with geometric heuristics for initial path generation. During execution, it monitors force feedback and triggers online replanning if forces exceed a safe threshold, adapting motion to reduce branch stress.

In practice

• Integrate force sensors for delicate manipulation.
• Model deformable objects with geometric heuristics.
• Implement online replanning for dynamic environments.

Topics

• Multi-armed Robotics
• Agricultural Automation
• Branch Manipulation
• Force-Aware Motion Planning
• RRT* Algorithm

Best for: Robotics Engineer, AI Scientist, AI Student

Related on AIssential

• AI-powered robot learns how to harvest tomatoes more efficiently — Robots can improve harvesting efficiency by assessing "harvest-ease" and adapting their picking strategy.
• Mana: Dexterous Manipulation of Articulated Tools — Mana reinterprets dexterous articulated tool manipulation as an animation problem, enabling zero-shot sim-to-real transfer via a coarse-to-fine pipeline.
• Robots that refuse to fail: AI evolves 'legged metamachines' that reassemble and withstand injury — Modular robots with athletic intelligence can reconfigure, recover from injury, and maintain movement in dynamic environments.
• Multi Graph Search for High-Dimensional Robot Motion Planning — Multi-Graph Search (MGS) offers a complete and bounded-suboptimal approach for high-dimensional robot motion planning.
• Vine-inspired robotic gripper gently lifts heavy and fragile objects — Vine-inspired robots can achieve strong, gentle grasps by transforming between open and closed loop configurations.
• Eternal.ag raises €8M to automate greenhouse harvesting with AI-powered robots — AI-powered autonomous robots are critical for addressing greenhouse labor shortages and ensuring year-round food supply.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by ΑΙhub.