Architecting Self-Driving Laboratories: AI and Robotics in Chemical Synthesis

Aim

Workshop Objectives

Workshop Structure

Day 1: Autonomous Discovery & Lab Computer Vision

• Introduction to self-driving laboratories and autonomous chemical experimentation.
• Understanding the Design–Make–Test–Analyze (DMTA) closed-loop workflow.
• Role of cameras, sensors, and digital data capture in modern laboratories.
• Basics of computer vision for monitoring chemical experiments.
• Detecting color changes, phase boundaries, turbidity, and reaction-state transitions.
• Converting visual observations into machine-readable experimental signals.
• Hands-on: Build an OpenCV-based reaction monitoring script for color/phase-change detection.
• Tools: Python, OpenCV, NumPy, Google Colab.

Day 2: Laboratory Automation & Robotic Workflow Control

• Fundamentals of laboratory robotics and automated chemical workflows.
• Introduction to Python-based instrument control and serial communication.
• Understanding automated liquid handling, pumps, stirrers, heaters, and sensors.
• Programming dispensing volume, flow rate, temperature, mixing, and timing parameters.
• Designing automated experimental sequences using workflow/state-machine logic.
• Implementing safety triggers, exception handling, and automated event logging.
• Hands-on: Build a virtual liquid-handling controller with automated experiment logs.
• Tools: Python, PySerial, Pandas, CSV/JSON.

Day 3: AI Optimization & Closed-Loop Autonomous Experimentation

• Introduction to AI-driven experimental optimization in chemical synthesis.
• Understanding Bayesian Optimization and data-efficient experiment selection.
• Optimizing parameters such as temperature, concentration, catalyst loading, and reaction time.
• Using experimental feedback to predict and recommend the next experiment.
• Integrating vision and sensor data into an intelligent decision-making loop.
• Introduction to active learning, AI agents, and emerging autonomous laboratory orchestration.
• Hands-on: Build a simulated AI-driven reaction optimization loop that recommends improved experimental conditions.
• Tools: Python, Scikit-Learn, Scikit-Optimize, NumPy, Pandas.

Who Should Enrol?

Important Dates

Workshop Outcomes