ASPIRE: Agentic Skill Programming through Iterative Robot Exploration

Authors: Runyu Lu, Yubo Wu, Ethan Kou, Letian Fu, Wenli Xiao, Ajay Mandlekar, Yinzhen Xu, Guanya Shi, Ken Goldberg, Ang Chen, Mosharaf Chowdhury, Yuke Zhu, Linxi "Jim" Fan, Guanzhi Wang

arXiv ID: 2607.00272

Problem: Traditional robot programming is difficult due to the need to orchestrate multimodal perception, manage contact dynamics, and handle diverse configurations and failures, with no existing system that autonomously writes, refines, and transfers reusable control programs across tasks and embodiments.

Key Methodology:

A closed-loop robot execution engine that exposes fine-grained multimodal traces for autonomous failure diagnosis, repair synthesis, and validation
A continually expanding skill library that distills validated fixes into reusable, transferable knowledge
Evolutionary search that generates diverse task sequences and control programs to explore beyond single-trajectory refinement

Key Results:

Surpasses prior methods by up to 77% on LIBERO-Pro manipulation under perturbation, 72% on Robosuite bimanual handover, and 32% on BEHAVIOR-1K long-horizon household tasks
On LIBERO-Pro Long zero-shot generalization, ASPIRE achieves 31% success vs 4% for prior methods (despite their use of test-time reasoning and retries)
Simulation-discovered skills transfer to real robots, substantially reducing real-world programming effort across different embodiments and robot APIs

Applied Context: Builders can use ASPIRE's approach to create robotic systems that self-improve over time, autonomously building a library of reusable skills that transfers across tasks and even across different robot hardware - dramatically reducing the manual programming burden for long-horizon manipulation.