A Hybrid Force-Position Strategy for Shape Control of Deformable Linear Objects With Graph Attention Networks

arXiv 2025 29.7 method, application

TLDR

Proposes hybrid force-position MPC with graph attention dynamics model for deformable linear object shape control, validated in simulations and real-world experiments.

Reasoning

Strengths: addresses challenging DLO manipulation with combined force/position and GNN-based dynamics, including sim and real validation. Weaknesses: no evidence of general world model claims; scope limited to DLO shape control and novelty is incremental.

Read-first score

Read-first score 29.7, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 5.

Recency 8%

86.7

Uses a gentle age decay so recent papers surface without erasing older foundations. 2025

Methodology quality 25%

Screens visible abstract and analysis fields for experiment, dataset, baseline, metric, and limitation evidence. markers=experiment,result

Reproducibility 25%

Screens links and visible text for paper, code, dataset, artifact, and repository signals. pdf=True; code=False; dataset=False; markers=code

Topical relevance 42%

7.1

Uses existing LLM keyword relevance scores normalized to 0-100. world model,world simulator,generative world model,interactive world model,video world model,world dynamics prediction,model-based reinforcement learning world model

Field roles

Frontier

Rank sensitivity

Stability: volatile; rank range: 318.

Keyword Scores

world dynamics prediction

world model

world simulator

model-based reinforcement learning world model

generative world model

interactive world model

video world model

Resources

Paper