AutoBio plus enables robot testing in realistic biology labs
AutoBio+: Benchmarking Dexterous Manipulation with Dynamic Experimental Phenomena in Realistic Laboratories
Robotics
Summary
Robots need to handle delicate tools and changing environments in biology labs, which is very hard. The authors created AutoBio+, a detailed simulation system where robots practice 14 tasks in realistic lab conditions, including flames and bubbles. This helps test and compare robot skills reliably, showing which robots do better or worse on different tasks. They also show their simulation matches real lab conditions well, making it useful for developing real robots in biology.
What this means in practice
- •For robotic automation teams: Evaluate and improve robots performing complex lab tasks with realistic dynamic effects like flames and bubbles using the AutoBio+ simulation platform.
- •For computer vision developers: Develop and test vision systems for robots handling dynamic and changing visual conditions in scientific lab environments using the hybrid rendering provided in AutoBio+.
Authors
Xuanye Wu, Yue Shi, Yongcan Yang, Yiqun Wang, Zhaoqi Wan, Dayu Xia, Jiezhang Cao, Xiangfei Li, Yao Mu, Guangtao Zhai
Abstract
Automating biological laboratories requires robots to perform dexterous tool interactions under realistic visual conditions while handling dynamic experimental phenomena, presenting significant challenges for robotic manipulation. To enable systematic evaluation under these conditions, we introduce AutoBio+, a simulation benchmark and framework for dexterous manipulation in realistic biological laboratory environments. AutoBio+ comprises 14 tasks across 4 robot embodiments, spanning functional tool use, precision manipulation, language-conditioned visual grounding, and instrument operation. We introduce controllable dynamic experimental phenomena, including flames, vapor, and bubbles, to capture the evolving visual conditions of laboratory procedures. To further enhance environmental realism, we develop a hybrid rendering pipeline that integrates interactive simulated foreground objects with Gaussian-splatting backgrounds reconstructed from 6 real laboratory scenes. For efficient large-scale data generation and evaluation, we further build a parallel simulation framework on top of mjlab, proposing Parallel Finite-State Machine (PFSM) and integrating batched rendering with decoupled trajectory replay for demonstration generation and model-independent closed-loop policy evaluation. Our evaluation reveals substantial variation in policy performance across tasks, exposing distinct strengths and weaknesses of current manipulation policies. Real-world experiments further demonstrate a reduced sim-to-real gap, supporting the validity of AutoBio+ for realistic manipulation evaluation. Supplementary materials are available at https://autobio-plus.github.io/AutoBioPlus.