Search for: All records

Robotic systems often struggle to adapt to dynamic, unstructured environments due to top-down design constraints based on human assumptions. Inspired by biological morphogenesis, this study introduces a cellular plasticity model based on Turing patterns, enabling multi-cellular robots to self-organize their cell phenotypes in response to environmental stimuli. The model leverages reaction-diffusion dynamics to capture key cellular plasticity phenomena observed in muscle cells, neurons, and stem cells. Analytical analysis explores equilibrium points, stability, and conditions for emergent Turing patterns, while simulations examine parametric influences on system behavior. Physical experiments with the Loopy platform demonstrate that its cells dynamically self-organize mechanical properties in response to behavioral and environmental demands. This response enables Loopy to achieve similar performance to empirically optimized static parameters in obstacle-free environments and outperform the static configuration in an environment with limited space. This work advances morphogenetic robotics, presenting a scalable framework for decentralized, dynamic adaptation in unmodeled environments. 

Survival analysis is a crucial statistical technique used to estimate the anticipated duration until a specific event occurs. However, current methods often involve discretizing the time scale and struggle with managing absent features within the data. This becomes especially pertinent since events can transpire at any given point, rendering event analysis a continuous concern. Additionally, the presence of missing attributes within tabular data is widespread. By leveraging recent developments of Transformer and Self-Supervised Learning (SSL), we introduce SSL-SurvFormer. This entails a continuously monotonic Transformer network, empowered by SSL pre-training, that is designed to address the challenges presented by continuous events and absent features in survival prediction. Our proposed continuously monotonic Transformer model facilitates accurate estimation of survival probabilities, thereby bypassing the need for temporal discretization. Additionally, our SSL pre-training strategy incorporates data transformation to adeptly manage missing information. The SSL pre-training encompasses two tasks: mask prediction, which identifies positions of absent features, and reconstruction, which endeavors to recover absent elements based on observed ones. Our empirical evaluations conducted across a variety of datasets, including FLCHAIN, METABRIC, and SUPPORT, consistently highlight the superior performance of SSL-SurvFormer in comparison to existing methods. Additionally, SSL-SurvFormer demonstrates effectiveness in handling missing values, a critical aspect often encountered in real-world datasets.