Self-Driving Labs: A New Paradigm in Cell Culture

Cell culture research remains manual and error prone, and automation has been adopted more slowly here than in chemistry or materials science. This survey reviews the components required to build a Self-Driving Laboratory (SDL) for cell culture, defined as an autonomous platform that executes experiments, quantifies and analyses results, and uses that analysis to select future experiments. The review comprises 27 control platforms, 14 perception instruments and 33 knowledge components, supported by a bibliometric analysis of SCOPUS records from 2010 to 2025. Systems are decomposed through a control-systems lens into control, perception, knowledge and communication elements, compared within each layer. Open-source liquid handlers and microscopes approach commercial performance, while centrifuges and integration software lag. Five recurring barriers keep the reviewed systems automated rather than autonomous: firmware and protocol lock-in, software fragmentation, mechanical modification, sparse and asynchronous sensing, and data-pipeline brittleness. Addressing them together is the integration challenge this review sets out.