Event
Proposes a Vision-LLM-Based Task Planning Method for Autism Education Robots
1 reports · 1 sources
Lada Kudláčková proposes a method in which a camera-equipped Nao robot uses a vision large language model to convert photographs of Shoebox Tasks into a PDDL representation and generate action plans, thereby autonomously executing tasks, prompting the next step, or reporting detected failures through action demonstrations, speech, and gestures. The solver takes only captured images as input, and teachers need no programming expertise; the authors state that with an adapted robot interface it can be integrated into a variety of educational robots. This 6-page paper was published at the AICCONF 2026 conference in Prague.
Report timeline
- Visual Task Modeling and Automated Planning for Nao Robots in Autism Spectrum Disorder Education
A method for autism spectrum disorder (ASD) education enables a camera-equipped Nao robot, using a vision-based large language model, to convert photos of a Shoebox Task into a Planning Domain Definition Language (PDDL) representation and generate action plans, autonomously executing the task, prompting the next step, or reporting failures through action demonstrations, voice, and gestures.
arXiv cs.RO ·