As a strategic initiative to enhance industrial competitiveness and drive high-quality economic development, intelligent manufacturing is currently undergoing a paradigm transformation from digitalization and networking to full intelligence. Robot cluster control, as a key enabling technology for complex engineering systems, offers novel theoretical perspectives and technological pathways for reconfiguring the organizational form and operational logic of intelligent manufacturing systems and for facilitating the associated paradigm transformation. Building upon an elucidation of the current status, practical drivers, and theoretical foundations underlying the integration of robot cluster control into manufacturing production systems, this study analyses the reconfiguration mechanism by which robot cluster control empowers the paradigm transformation of intelligent manufacturing, along four dimensions: task execution logic, collaborative decision-making structure, production scheduling mode, and resource allocation path. It further identifies three typical application modes (i.e., distributed assembly, unmanned collaborative factory, and dynamic production network) and examines the practical case of Zoomlion's intelligent factory for shared manufacturing, thereby illustrating the evolutionary changes at the task, system, and network levels driven by robot cluster control. The study also summarizes the future challenges confronting this paradigm transformation from technological (collaborative algorithms, real-time performance, reliability), industrial (standards, ecosystem, security), and strategic-implementation (technology supply, system integration, organizational execution) perspectives, and proposes targeted recommendations encompassing key technological breakthroughs, industrial ecosystem optimization, and strengthened strategic guidance. The findings contribute to deepening the integration of robot cluster control with intelligent manufacturing, and provide both theoretical interpretation and application-oriented support for the paradigm transformation of intelligent manufacturing systems.