Task2Morph: Differentiable Task-inspired Framework for Contact-Aware Robot Design (2403.19093v1)
Abstract: 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.
- Josh Bongard. Why morphology matters. The horizons of evolutionary robotics, 6:125–152, 2014.
- Task-agnostic morphology evolution. In International Conference on Learning Representations (ICLR), 2020.
- Metamorph: Learning universal controllers with transformers. In International Conference on Learning Representations (ICLR), 2021.
- Embodied intelligence via learning and evolution. Nature communications, 12(1):1–12, 2021.
- Real-world embodied ai through a morphologically adaptive quadruped robot. Nature Machine Intelligence, 3(5):410–419, 2021.
- Embodied cognition is not what you think it is. Frontiers in psychology, 4:58, 2013.
- Springer handbook of computational intelligence. Springer, 2015.
- An interaction-modeling mechanism for context-dependent text-to-sql translation based on heterogeneous graph aggregation. Neural Networks, 142:573–582, 2021.
- Neural graph evolution: Towards efficient automatic robot design. In International Conference on Learning Representations (ICLR), 2018.
- Robogrammar: graph grammar for terrain-optimized robot design. ACM Transactions on Graphics (TOG), 39(6):1–16, 2020.
- Modular mobile robot design selection with deep reinforcement learning. In NeurIPS Workshop on ML for engineering modeling, simulation and design, 2020.
- Jointly learning to construct and control agents using deep reinforcement learning. In 2019 International Conference on Robotics and Automation (ICRA), pages 9798–9805. IEEE, 2019.
- Transform2act: Learning a transform-and-control policy for efficient agent design. In International Conference on Learning Representations (ICLR), 2021.
- An end-to-end differentiable framework for contact-aware robot design. In Robotics: Science and Systems (RSS), 2021.
- Functional co-optimization of articulated robots. In 2017 IEEE International Conference on Robotics and Automation (ICRA), pages 5035–5042. IEEE, 2017.
- One robot for many tasks: Versatile co-design through stochastic programming. IEEE Robotics and Automation Letters (RA-L), 5(2):1680–1687, 2020.
- Joint optimization of robot design and motion parameters using the implicit function theorem. In Robotics: Science and systems (RSS), volume 8, 2017.
- Hardware as policy: Mechanical and computational co-optimization using deep reinforcement learning. In Conference on Robot Learning (CoRL), pages 1158–1173. PMLR, 2021.
- C^ 2: Co-design of robots via concurrent networks coupling online and offline reinforcement learning. arXiv preprint arXiv:2209.06579, 2022.
- Karl Sims. Evolving virtual creatures. In Proceedings of the 21st annual conference on Computer graphics and interactive techniques, pages 15–22, 1994.
- Evolution gym: A large-scale benchmark for evolving soft robots. Advances in Neural Information Processing Systems (NIPS), 34:2201–2214, 2021.
- Contrasting exploration in parameter and action space: A zeroth-order optimization perspective. In The 22nd International Conference on Artificial Intelligence and Statistics, pages 2926–2935. PMLR, 2019.
- N-limb: Neural limb optimization for efficient morphological design. arXiv preprint arXiv:2207.11773, 2022.
- Modular robot design synthesis with deep reinforcement learning. In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), volume 34, pages 10418–10425, 2020.
- Modular robot design optimization with generative adversarial networks. In 2022 International Conference on Robotics and Automation (ICRA), pages 4282–4288. IEEE, 2022.
- Fit2form: 3d generative model for robot gripper form design. In Conference on Robot Learning (CoRL), pages 176–187. PMLR, 2021.
- Redmax: Efficient & flexible approach for articulated dynamics. ACM Transactions on Graphics (TOG), 38(4):1–10, 2019.
- Add: Analytically differentiable dynamics for multi-body systems with frictional contact. ACM Transactions on Graphics (TOG), 39(6):1–15, 2020.
- Joint object detection and multi-object tracking with graph neural networks. In 2021 IEEE International Conference on Robotics and Automation (ICRA), pages 13708–13715. IEEE, 2021.
- Prioritized experience replay. International Conference on Learning Representations (ICLR), 2016.