The growing global demand for critical raw materials (CRMs) essential to renewable energy, electromobility, and digital technologies has accelerated the need for advanced exploration methods capable of operating in increasingly challenging geological environments. Traditional drilling systems, designed primarily for shallow mineral and hydrocarbon exploration, face limitations in heterogeneous and consolidated formations where rock heterogeneity, variable mechanical strength, and borehole instability restrict operational efficiency. This review bridges geological science and robotic engineering by analyzing the evolution of next-generation autonomous drilling technologies integrating sensor systems, artificial intelligence (AI), and real-time geotechnical feedback. The current work explores how robotic drilling systems can autonomously adapt to variable lithologies, optimize penetration rates, and ensure borehole stability through intelligent sensing and control. The paper reviews the geological, geomechanical and ore deposit characteristics of CRMs, discusses state-of-the-art drilling optimization strategies, and highlights advances in measurement while drilling (MWD), logging while drilling (LWD), and geochemical analysis techniques. It also suggests a list of sensor techniques for possible future integration in autonomous subsurface robotic systems. It concludes by emphasizing the need for integration between subsurface geological modeling and intelligent drilling robotics as a pathway toward sustainable and efficient CRM exploration.
Avrantinis et al. (Fri,) studied this question.