Task2Morph: Differentiable Task-inspired Framework for Contact-Aware Robot Design

2024-03-28 · Yishuai Cai, Shaowu Yang, Minglong Li, Xinglin Chen, Yunxin Mao, Xiaodong Yi, Wenjing Yang

Optimizing the morphologies and the controllers that adapt to various tasks is a critical issue in the field of robot design, aka. embodied intelligence. Previous works typically model it as a joint optimization problem and use search-based methods to find the optimal solution in the morphology space. However, they ignore the implicit knowledge of task-to-morphology mapping which can directly inspire robot design. For example, flipping heavier boxes tends to require more muscular robot arms. This paper proposes a novel and general differentiable task-inspired framework for contact-aware robot design called Task2Morph. We abstract task features highly related to task performance and use them to build a task-to-morphology mapping. Further, we embed the mapping into a differentiable robot design process, where the gradient information is leveraged for both the mapping learning and the whole optimization. The experiments are conducted on three scenarios, and the results validate that Task2Morph outperforms DiffHand, which lacks a task-inspired morphology module, in terms of efficiency and effectiveness.

📄 PDF Abstract BibTeX arXiv:2403.19093

Code (0)

등록된 구현이 없습니다.

Brain-inspired global-local learning incorporated with neuromorphic computing

2020-06-05 · Yujie Wu, Rong Zhao, Jun Zhu, Feng Chen 외

Two main routes of learning methods exist at present including error-driven global learning and neuroscience-oriented local learning. Integrating them into one network may provide complementary learning capabilities for …

Continual Learning Few-Shot Learning Meta-Learning

InkDiffuser: High-Fidelity One-shot Chinese Calligraphy via Differentiable Morphological Optimization

2026-05-07 · Kunchong Shi, Jing Zhang arxiv

Current Chinese calligraphy generation methods suffer from poor stroke rendering and unrealistic ink morphology, resulting in outputs with limited visual fidelity and artistic fluidity. To address this problem, we propos…

SoftZoo: A Soft Robot Co-design Benchmark For Locomotion In Diverse Environments

2023-03-16 · Tsun-Hsuan Wang, Pingchuan Ma, Andrew Everett Spielberg, Zhou Xian 외

While significant research progress has been made in robot learning for control, unique challenges arise when simultaneously co-optimizing morphology. Existing work has typically been tailored for particular environments…

MorphNAS: Differentiable Architecture Search for Morphologically-Aware Multilingual NER

2025-08-19 · Prathamesh Devadiga, Omkaar Jayadev Shetty, Hiya Nachnani, Prema R arxiv

Morphologically complex languages, particularly multiscript Indian languages, present significant challenges for Natural Language Processing (NLP). This work introduces MorphNAS, a novel differentiable neural architectur…

Neural Architecture Search

Externally Validated Multi-Task Learning via Consistency Regularization Using Differentiable BI-RADS Features for Breast Ultrasound Tumor Segmentation

2025-11-20 · Jingru Zhang, Saed Moradi, Ashirbani Saha arxiv

Multi-task learning can suffer from destructive task interference, where jointly trained models underperform single-task baselines and limit generalization. To improve generalization performance in breast ultrasound-base…

Multi-Task Learning Tumor Segmentation