This chapter provides a comprehensive review of the integration of artificial intelligence into robotics, with a focus on Machine Learning (ML), Deep Learning (DL), and Transformer-based models. Explore how these paradigms have enabled autonomous robots to perceive, plan, and act with increasing efficiency and adaptability across diverse application domains, including healthcare, agriculture, manufacturing, and human-robot collaboration. Using the PRISMA methodology, the chapter synthesizes empirical findings from 2015 to 2025, highlighting recent advances in ML-based fault prediction, DL-driven navigation and control, and Transformer-enhanced planning and perception. The tables presented in this chapter summarize algorithmic approaches and performance benchmarks from selected studies, illustrating measurable improvements in accuracy, decision-making, and real-time performance. The chapter also addresses key challenges, such as interpretability, data dependency, and computational complexity, offering insights into emerging trends, including neuro-symbolic learning and multi-agent systems. By analyzing strengths, limitations, and future directions, this work contributes to a deeper understanding of how artificial intelligence methods shape the next generation of intelligent and adaptive robotic systems.
