Search papers, labs, and topics across Lattice.
This paper introduces RopeDreamer, a latent dynamics framework for modeling deformable linear objects (DLOs) that combines a Recurrent State Space Model with a Quaternionic Kinematic Chain representation. By encoding DLO state as a sequence of relative rotations, the model enforces physical validity and link-length constancy, avoiding issues like self-intersection and stretching. Experiments on simulated pick-and-place tasks show RopeDreamer reduces long-horizon prediction error by 40.52% and inference time by 31.17% compared to state-of-the-art baselines, while also improving topological consistency.
By representing deformable linear objects as a chain of relative rotations, RopeDreamer achieves state-of-the-art prediction accuracy and topological consistency in long-horizon manipulation tasks.
The robotic manipulation of Deformable Linear Objects (DLOs) is a fundamental challenge due to the high-dimensional, non-linear dynamics of flexible structures and the complexity of maintaining topological integrity during contact-rich tasks. While recent data-driven methods have utilized Recurrent and Graph Neural Networks for dynamics modeling, they often struggle with self-intersections and non-physical deformations, such as tangling and link stretching. In this paper, we propose a latent dynamics framework that combines a Recurrent State Space Model with a Quaternionic Kinematic Chain representation to enable robust, long-term forecasting of DLO states. By encoding the DLO as a sequence of relative rotations (quaternions) rather than independent Cartesian positions, we inherently constrain the model to a physically valid manifold that preserves link-length constancy. Furthermore, we introduce a dual-decoder architecture that decouples state reconstruction from future-state prediction, forcing the latent space to capture the underlying physics of deformation. We evaluate our approach on a large-scale simulated dataset of complex pick-and-place trajectories involving self-intersections. Our results demonstrate that the proposed model achieves a 40.52% reduction in open-loop prediction error over 50-step horizons compared to the state-of-the-art baseline, while reducing inference time by 31.17%. Our model further maintains superior topological consistency in scenarios with multiple crossings, proving its efficacy as a compositional primitive for long-horizon manipulation planning.
