EMNLP'26 | Auto-Robotist: An AI Agent for Collaborative Robot Design Where Every Trial Leaves Experience
EMNLP'26 | AI机器人协同设计智能体：每一次试错都留下经验
Researchers at the University of Michigan propose Auto-Robotist, an agent that combines structural search with controller training to design soft robots. Using a large language model, the system generates and maintains a continuously updated skills library from simulation results in EvoGym. Experiments show it outperforms standard genetic algorithms on seven tasks across 5x5 and 10x10 grid spaces, achieving higher scores and faster convergence while learning from both successes and failures.
Source: Chinese robotics — Research news discovery · Read original article ↗
English translation unavailable. Choose the original article to read it.
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- Environment
- SimulationOpen source S3
Reported numbers
training steps
512,000
design space grid size
5
design space grid size
10
- paper: not_reported
- code: Reported pending Open source S8
- weights: not_reported
- dataset: not_reported
- code: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
机器人的身体结构与运动控制紧密相连：驱动器放在哪里、身体怎样保持稳定，都需要通过控制训练与仿真反复验证。
Open source S3
每个候选机器人都先接受51.2万步的PPO控制训练，再进行仿真评分。
Open source S23
实验表明，这些经验能改善EvoGym中七项软体机器人仿真任务的设计搜索，并从5×5网格迁移到10×10的更大设计空间。
Open source S10
Auto-Robotist的主要结果报告三个随机种子（0、9、42）下的均值与标准差，用于观察平均表现及运行间的波动。
Open source S24
文中的「奖励」和「适应度」都用于衡量机器人完成任务的表现得分。
Open source S15
论文链接：https://arxiv.org/abs/2605.25832 代码链接：https://github.com/wangyf9/Auto-Robotist Auto-Robotist的核心是一个持续更新的「技能库」。
Open source S8
Source：Chinese robotics — Research news discovery · msn.cn