Description: we hypothesis optimising morphology for behaviour diversity with the proxies of stiffness range and number of fingerip interaction behaviours with different tendon configurations will lead to better hand designs. One significant observation we make is the human-like morphology being one of the highest performing by these metrics, this makes sense evolutionarily as a highly efficient design. We are also able to tune performance in different kinds of tasks by selecting designs over our simulated workspace.

Full article: https://doi.org/10.1109/RoboSoft63089.2025.11020976

Cite: Carlet, R., Gilday, K., & Hughes, J. (2025, April). Behaviour Range Optimization of the Dexterous Robotic Open Parametric Hand. In 2025 IEEE 8th International Conference on Soft Robotics (RoboSoft) (pp. 1-7). IEEE.