Manipulator Generative Design Optimization for Orchard Environments
Marcus Rosette, Tianhai Wang, James Burridge, Wei Guo, Joseph Davidson
AI summary
Problem
Off-the-shelf industrial robot arms lack the reach, dexterity, and collision-free workspace required for precise agricultural manipulation in unstructured orchard settings.
Approach
The authors use an NSGA-II evolutionary algorithm to automatically generate and evaluate candidate manipulator kinematics in a physics-based simulation, optimizing for reachability, torque, and motion planning costs against real 3D orchard scans.
Key results
- Automated generation of task-specific manipulator kinematics via NSGA-II evolutionary search
- Multi-objective optimization balancing reachability, torque, manipulability, and path planning cost
- Demonstration on real-world dormant cherry tree pruning using expert-collected 3D LiDAR scans
- Optimized designs achieve improved workspaces and reduced operational constraints compared to commercial industrial arms
Why it matters
Provides an automated pathway for designing deployable, task-specific agricultural robots that overcome the limitations of off-the-shelf hardware in complex field environments.
Abstract
Manipulators are essential for orchard robotics tasks such as pruning and harvesting, which require precise, dexterous motion in cluttered and unstructured environments. Off-the-shelf industrial arms, while readily available, often lack the reach and dexterity required for these settings. In this paper, we present a simulation-driven, multi-objective opti- mization framework for task-specific manipulator kinematics, leveraging the NSGA-II evolutionary algorithm and physics- based evaluation. Candidate designs are encoded with high-level parameters – joint type, axis orientation, link length, and joint count – then automatically generated as URDF models and evaluated in simulation for reachability, manipulability, torque demand, and motion planning cost. Trade-offs are revealed on a Pareto front, enabling exploration across diverse designs. The framework is demonstrated on a real-world tree pruning task, using collected 3D scans of expert-pruned trees and an automated prune point identification pipeline to generate target points to guide the optimization. Results show that the proposed approach produces task-specific manipulator designs with improved workspaces and reduced operational constraints compared to a commercial industrial arm, offering a viable pathway toward deployable agricultural manipulation systems.