Saved in:
Bibliographic Details
Main Authors: Sun, Haoyu, Zhao, Meng, Wang, Tianhao, Wu, Jianxu
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.02038
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908933544738816
author Sun, Haoyu
Zhao, Meng
Wang, Tianhao
Wu, Jianxu
author_facet Sun, Haoyu
Zhao, Meng
Wang, Tianhao
Wu, Jianxu
contents Deep learning has shown strong potential for scientific discovery, but its ability to model macroscopic rigid-body kinematic constraints remains underexplored. We study this problem on spatial over-constrained mechanisms and propose O-ConNet, an end-to-end framework that infers mechanism structural parameters from only three sparse reachable points while reconstructing the full motion trajectory, without explicitly solving constraint equations during inference. On a self-constructed Bennett 4R dataset of 42,860 valid samples, O-ConNet achieves Param-MAE 0.276 +/- 0.077 and Traj-MAE 0.145 +/- 0.018 (mean +/- std over 10 runs), outperforming the strongest sequence baseline (LSTM-Seq2Seq) by 65.1 percent and 88.2 percent, respectively. These results suggest that end-to-end learning can capture closed-loop geometric structure and provide a practical route for inverse design of spatial over-constrained mechanisms under extremely sparse observations.
format Preprint
id arxiv_https___arxiv_org_abs_2604_02038
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle O-ConNet: Geometry-Aware End-to-End Inference of Over-Constrained Spatial Mechanisms
Sun, Haoyu
Zhao, Meng
Wang, Tianhao
Wu, Jianxu
Robotics
Deep learning has shown strong potential for scientific discovery, but its ability to model macroscopic rigid-body kinematic constraints remains underexplored. We study this problem on spatial over-constrained mechanisms and propose O-ConNet, an end-to-end framework that infers mechanism structural parameters from only three sparse reachable points while reconstructing the full motion trajectory, without explicitly solving constraint equations during inference. On a self-constructed Bennett 4R dataset of 42,860 valid samples, O-ConNet achieves Param-MAE 0.276 +/- 0.077 and Traj-MAE 0.145 +/- 0.018 (mean +/- std over 10 runs), outperforming the strongest sequence baseline (LSTM-Seq2Seq) by 65.1 percent and 88.2 percent, respectively. These results suggest that end-to-end learning can capture closed-loop geometric structure and provide a practical route for inverse design of spatial over-constrained mechanisms under extremely sparse observations.
title O-ConNet: Geometry-Aware End-to-End Inference of Over-Constrained Spatial Mechanisms
topic Robotics
url https://arxiv.org/abs/2604.02038