O-ConNet: Geometry-Aware End-to-End Inference of Over-Constrained Spatial Mechanisms

TL;DR

O-ConNet is a novel deep learning framework that infers the structure and motion of over-constrained spatial mechanisms from minimal data, outperforming traditional sequence models.

Contribution

It introduces an end-to-end approach for modeling spatial mechanisms directly from sparse points, bypassing explicit constraint solving during inference.

Findings

O-ConNet achieves significantly lower MAE scores than LSTM-Seq2Seq.

It effectively captures geometric structure from minimal observations.

The method enables practical inverse design of mechanisms with sparse data.

Abstract

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.