Search papers, labs, and topics across Lattice.
O-ConNet, an end-to-end deep learning framework, is introduced to infer structural parameters and reconstruct motion trajectories of over-constrained spatial mechanisms from sparse reachable points. By implicitly learning kinematic constraints, O-ConNet avoids explicit constraint equation solving during inference. Experiments on a Bennett 4R dataset show O-ConNet significantly outperforms LSTM-Seq2Seq in both parameter estimation (65.1% improvement) and trajectory reconstruction (88.2% improvement).
End-to-end learning can now infer the full motion of complex mechanical linkages from just three sparse points, bypassing the need to explicitly solve kinematic constraint equations.
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.
