This chapter examines key technologies for offline data collection in ISAC systems. It first compares manual and automated data collection methods, highlighting their strengths, weaknesses, and applicable scenarios, while addressing the challenges of balancing data quality and quantity in offline CSI collection. The chapter then details the design of automated systems, including robotic devices and power-driven sampling techniques, to streamline large-scale data collection. Additionally, it contrasts device-based and device-free methods, proposing strategies to minimize data loss. Two core algorithms are introduced: the A3C-IP algorithm, which uses asynchronous reinforcement learning to optimize data collection paths and fingerprint prediction, and the CPPU algorithm, which integrates GAN to dynamically update CSI data and improve collection efficiency through optimal path planning. Performance evaluations validate their effectiveness in intelligent positioning systems, offering robust solutions for efficient offline data collection.