Redesigning Chemistry Laboratories: Accelerating Material Discovery through Simulation and Cutting-Edge Robotics

Voica, Andrei ; Moreno Paris, Daniel

4. semester

Robotics, M.Sc.

2023

2023-06-01

88

Denne afhandling undersøger, hvordan kemilaboratorier kan redesignes til selvdrevne miljøer, der accelererer opdagelsen af nye materialer ved hjælp af robotteknologi og simulation. Med udgangspunkt i Material Acceleration Platforms (MAPs) udvikles en fleksibel robotplatform, der kan designe, udføre, teste og analysere eksperimenter i en lukket sløjfe. Den fysiske AAU Matrix Production-opsætning med fem Kuka robotmanipulatorer, B&R Automation Acopos 6D magnetisk levitationsplatform og specialfremstillede dele blev replikeret i Nvidia Isaac Sim for at udvikle og evaluere arbejdsgange og eksperimenter. ROS1 blev brugt til at styre både simulerede og virkelige Kuka-robotter. Det centrale forskningsspørgsmål er: Hvordan kan en robotplatform hjælpe med at opdage nye materialer på en fleksibel og effektiv måde? Simulationseksperimenter viser, at systemet kan fuldautomatisere en kemisk proces, mens overførsel til den fysiske opsætning var udfordrende. Projektet bidrager med design-, styrings- og integrationsindsigter, der kan understøtte fremtidig autonom materialeforskning.

This thesis explores how chemistry laboratories can be redesigned into self-driven environments that accelerate the discovery of new materials through robotics and simulation. Building on the concept of Material Acceleration Platforms (MAPs), it develops a flexible robotic platform capable of designing, executing, testing, and analyzing experiments in a closed loop. The AAU Matrix Production setup—comprising five Kuka robotic manipulators, the B&R Automation Acopos 6D magnetic levitation platform, and purpose-made parts—was replicated in Nvidia Isaac Sim to develop and evaluate workflows and experiments. ROS1 was used to control both simulated and real Kuka robots. The core research question is: How can a robotic platform help in the discovery of new materials in a flexible and efficient way? Simulation experiments demonstrate that the system can automatically complete a chemical process, while transferring the solution to the physical setup proved challenging. The project contributes design, control, and integration insights to support future autonomous materials research.

[This summary has been generated with the help of AI directly from the project (PDF)]

MAPs ; Robotics ; Self-driving laboratories ; Kuka ; Acopos 6D ; ROS ; Nvidia Isaac Sim

Download
View record in AAU Student Projects