Preprints
https://doi.org/10.5194/ms-2026-198
https://doi.org/10.5194/ms-2026-198
19 Nov 2026
 | 19 Nov 2026
Status: this preprint is currently under review for the journal MS.

Inverse Kinematics for a Redundant Wing-Inspection Robot via Multiobjective Optimization with an Improved Second-Generation Nondominated Sorting Genetic Algorithm

Xiaolin Dai, Xingqi Zhou, Nuo Tu, Yu He, Yang Zhang, and Pengyuan Zhao

Abstract. Wing scanning requires feasible probe poses and continuous joint configurations with adequate joint-limit margins. This study develops an adaptive search-space module (ASM) coupled with the second-generation nondominated sorting genetic algorithm (NSGA-II), denoted ASM-NSGA-II, for a ten-degree-of-freedom robot comprising four external axes and a six-axis industrial arm. The forward-kinematic chain is decomposed to optimize only the four redundant parameters; analytical arm inverse kinematics recovers candidate configurations, which are screened by joint-limit and probe-pose constraints. Joint continuity, joint-limit margin, and manipulability guide multiobjective selection. The ASM scores nested candidate regions through repeated presampling, while normalized adjacent-pose similarity controls migration from the preceding population. ASM selects the sampling region, and NSGA-II retains nondominated solutions within it. Ten independent runs on each of four paths show mean-fitness reductions of 41.6–51.5 % relative to standard NSGA-II and 6.2–25.0 % relative to the published T-IK method; mean solution time differs from T-IK by no more than 8.7 %. A wing-surface task connects mesh-based pose construction to joint-sequence execution in MATLAB–CoppeliaSim, providing qualitative model-level motion verification. The results support improved aggregate trade-offs for the tested system and paths; physical tracking accuracy, real-time operation, and collision safety require further validation.

Competing interests: At least one of the (co-)authors is a member of the editorial board of Mechanical Sciences.

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.
Share
Xiaolin Dai, Xingqi Zhou, Nuo Tu, Yu He, Yang Zhang, and Pengyuan Zhao

Status: open (until 26 Dec 2026)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
Xiaolin Dai, Xingqi Zhou, Nuo Tu, Yu He, Yang Zhang, and Pengyuan Zhao
Xiaolin Dai, Xingqi Zhou, Nuo Tu, Yu He, Yang Zhang, and Pengyuan Zhao
Metrics will be available soon.
Latest update: 19 Nov 2026
Download
Short summary
We developed a way for a robot to move a probe along an aircraft wing. It uses information from one point on the scanning path to narrow the search for how the robot should move at the next point. In computer tests on four paths, it found better ways to balance smooth motion and keep joints away from their limits, with similar computing times to an existing method. We also demonstrated a virtual wing scan.
Share