LATIDIA · Robótica

LEMCA: Síntesis guiada por LLM de arquitecturas eficientes de control de conmutación de modo

arXiv: 2609.21319v1Tipo de anuncio: nuevo Resumen: Las tareas de control físico en el mundo natural, como conducir o manipular objetos, con frecuencia exhiben variaciones dramáticas en la complejidad sensorial y de cómputo a lo largo del tiempo. Cor

La noticia

arXiv:2609.21319v1 Announce Type: new Abstract: Physical control tasks in the natural world, such as driving or object manipulation, frequently exhibit dramatic variations in sensory and compute complexity over time. Correspondingly, a natural resource-efficient choice for robot control is to dynamically switch between control modes with varying resource allocations. However, such "mode-switching controllers" (MSCs) have historically required laborious, expert-driven design and synthesis for each new task. Driven by these design difficulties, modern robotic control architectures often fall back to a wasteful "monolithic" one-size-fits-all structure, where resource allocation is permanently anchored to the hardest, most resource-intensive task phases. To facilitate the design of performant yet efficient MSCs, we propose LLM-Guided synthesis of