Simulation-Based Adaptive Multi-Stage Toolpath Optimization for Robotic Machining of Freeform Shapes

Published: 1 June 2026| Version 1 | DOI: 10.17632/t4663htkcy.1
Contributors:
Kele Zhu,
,
,
, Zichun Zhou

Description

This dataset supports the manuscript "Simulation-Based Adaptive Multi-Stage Toolpath Optimization for Robotic Machining of Freeform Shapes." It contains simulation outputs, parsed toolpath data, KUKA controller source outputs, region masks, case-level result tables, and cross-geometry summary files used to evaluate a region-aware adaptive multi-stage toolpath planning workflow for robotic freeform machining. The dataset includes four benchmark geometries: Case A, a primary irregular mixed freeform case with heterogeneous smooth, transition, and detail-sensitive regions; Case B, a smooth-dominant freeform case with localized details; Case C, a continuous freeform surface with gradual curvature transitions; and Case D, a feature-dense high-curvature freeform case. For each case, the dataset provides geometry files, SR/TBR/DR masks, parsed path CSV files, controller source files, evaluation result CSV files, and paper-ready table CSV files. The compared strategies include a single-stage single-treatment baseline, a non-adaptive two-stage baseline, and the proposed adaptive multi-stage multi-treatment workflow. The cross-geometry summary files support the manuscript tables and figures reporting total simulated cycle time, stage-wise workload redistribution, pre-finish residual proxies, finishing burden, final-quality proxies, transition metrics, and run-status records. The README.md and data_dictionary.csv files describe the folder structure, terminology, variables, and manuscript mapping.

Files

Steps to reproduce

The dataset is intended to support the manuscript results rather than to provide a full executable software workflow. To reproduce the reported tables and figures, first unzip the dataset package and review README.md and data_dictionary.csv. The cross-geometry summary files in summary_outputs/ provide the source data for the main benchmark comparisons. For each benchmark case, the cases/ folder contains geometry files, SR/TBR/DR masks, parsed toolpath CSV files, raw KUKA controller source outputs, evaluation result CSV files, and paper-ready table CSV files. The values reported in the manuscript can be checked by comparing summary_outputs/ with the case-level results/ and tables/ folders. The original simulation and evaluation were conducted in an offline Rhino/Grasshopper/KUKA|prc environment using the shared assumptions described in the manuscript.

Institutions

Categories

Manufacturing Engineering, Mechanical Engineering, Robotics

Licence