Searcharxiv⌕ Search

arXiv subjects

Aldair Gongora

Publications and source records attributed to Aldair Gongora.

2 recordsLinked to original sources

A Physical Testbed for Self-Driving Laboratories: Comparing Bayesian and Agentic Optimization in a Minimal Closed-Loop Color Formulation Task

Self-driving laboratories (SDLs) integrate laboratory robotics and automation with AI-driven decision-making to close the design-build-test-learn (DBTL) cycle. However, physical testbeds that combine off-the-shelf hardware and AI into a functioning closed-loop workflow remain in early stages of development, and their integration with commercial laboratory automation is comparatively less developed. In this work, we develop a physical testbed that assembles and integrates commercially available laboratory automation instruments and demonstrate continuous, closed-loop, autonomous operation over 24 cumulative hours of unattended experimentation. We validate this platform on a minimal closed-loop discovery task in which the system autonomously formulates aqueous dye mixtures to match a target absorbance spectrum. The system completed six independent campaigns, each consisting of 48 closed-loop iterations over approximately four hours. In every campaign, the system autonomously selected, dispensed, and measured each mixture, with an AI-driven decision-making algorithm determining subsequent formulations without human intervention. Using this testbed, we compared the performance of two AI-driven decision-making algorithms, traditional Bayesian optimization (BO) and an agentic approach, in selecting mixtures under a fixed total-volume constraint. The agentic optimizer matched target absorbance spectra using fewer iterations than BO, with reductions up to 52%, and converged on a solution for one target that BO failed to match within 48 iterations. These results demonstrate a reference physical platform for SDLs that can evaluate both established and emerging AI strategies, including large language model (LLM)-directed optimization, before their deployment in real-world applications where experiments may be costly and time-consuming to prepare.

cs.RO↗

Autonomous Discovery of Tough Structures

A key feature of mechanical structures ranging from crumple zones in cars to padding in packaging is their ability to provide protection by absorbing mechanical energy. Designing structures to efficiently meet these needs has profound implications on safety, weight, efficiency, and cost. Despite the wide varieties of systems that must be protected, a unifying design principle is that protective structures should exhibit a high energy-absorbing efficiency, or that they should absorb as much energy as possible without mechanical stresses rising to levels that damage the system. However, progress in increasing the efficiency of such structures has been slow due to the need to test using tedious and manual physical experiments. Here, we overcome this bottleneck through the use of a self-driving lab to perform >25,000 machine learning-guided experiments in a parameter space with at minimum trillions of possible designs. Through these experiments, we realized the highest mechanical energy absorbing efficiency recorded to date. Furthermore, these experiments uncover principles that can guide design for both elastic and plastic classes of materials by incorporating both geometry and material into a single model. This work shows the potential for sustained operation of self-driving labs with a strong human-machine collaboration.

physics.app-ph↗