How diffusion-based optimization discovers diverse octopus-arm coordination patterns and edits them to recover crawling performance under new actuation constraints.
Read more
How filtering-control duality produces transformer-like layers, and what a simple prediction experiment reveals about attention and memory.
Read more
A HMM-based learning framework that understand sequential data while ensuring full interpretability and fast convergence.
Read more
How a neural network extends mechanics-based soft-arm reconstruction, turning sparse marker poses into smooth posture and strain estimates fast enough for real-time use.
Read more
How reinforcement learning, fast energy shaping, and local contact reflexes work together to coordinate a simulated octopus searching for food.
Read more
How muscle geometry and energy shaping turn reaching and grasping in three dimensions into a search for a useful equilibrium.
Read more
How camera measurements and Cosserat rod mechanics recover bending, twisting, stretching, and shearing along a soft robotic arm.
Read more