Abstract: Faced with the pressing challenges of an aging population, a shrinking agricultural workforce, large-scale rural migration, and rising agricultural labor costs in China, the development of autonomous agricultural machinery has emerged as a core strategy to address these issues. This technological innovation is crucial for enhancing agricultural productivity and sustainability and plays a vital role in establishing fully functional unmanned farm systems. By reducing reliance on human labor, autonomous agricultural machinery offers solutions to labor shortages while improving operational efficiency and precision. This paper provides a comprehensive review of the technical requirements for autonomous agricultural machinery, with a focus on key research advancements in China, including environmental perception, high-precision agricultural mapping, autonomous positioning and navigation, path planning, and tracking control, which collectively form the foundation of this field. One of the hallmark features of unmanned farms is the seamless and fully automated transfer of machinery between storage facilities and fields. This involves autonomous machinery automatically departing from its designated storage location, navigating farm roads to perform field operations, and returning to the shed to park precisely at its original position upon task completion. This automated transfer process demonstrates the adaptability and operational capability of autonomous machinery in diverse environments such as fields, farm roads, and agricultural sheds, highlighting its transformative potential for traditional farming workflows. However, a critical link in the unmanned farm system remains underdeveloped: operations within the storage shed. Key processes, including machinery parking, implement attachment, and departure, still heavily depend on manual intervention. Achieving full automation across all scenarios requires advanced indoor positioning and navigation systems, as well as precise environmental perception technologies, to ensure safety and efficiency during shed operations. Unlike the open and relatively predictable environments of fields and farm roads, storage sheds are often densely populated with machinery, offering limited maneuvering space and presenting heightened demands for safety and accuracy in perception and navigation. This complexity necessitates technological sophistication exceeding that required for open-field operations. While significant progress has been made in autonomous machinery for field and road operations, research on perception and autonomous parking technologies for shed operations remains relatively limited. Addressing these challenges is essential for bridging the automation gap and unlocking the full potential of unmanned farms. This paper focuses on the task of automatic parking for autonomous agricultural machinery upon returning to the shed after field operations. It explores various indoor positioning and navigation technologies and the perception systems required to enable these operations. The core challenge of automatic parking lies in achieving precise navigation from the shed entrance to a designated parking spot without Global Navigation Satellite System（GNSS） signals. Automatic parking relies on indoor positioning and navigation technologies to perceive the environment, locate the machinery, and plan safe and efficient routes for final docking. Positioning and navigation approaches are categorized into fixed-route and non-fixed-route methods. Fixed-route navigation includes visual navigation (lane tracking and visual marker localization), rail guidance, and magnetic navigation, offering simplicity, precision, and stability. These methods also support shed management by delineating safe zones, enhancing operational safety. In contrast, non-fixed-route navigation methods, such as external source-based positioning (e.g., Radio Frequency Identification, WiFi, Bluetooth, and Ultra-Wideband) and simultaneous localization and mapping (SLAM), provide superior flexibility, scalability, and adaptability to complex and dynamic indoor environments. Finally, this paper summarizes the technical and practical challenges facing autonomous agricultural machinery, including the integration of diverse technologies, adaptation to dynamic environments, and the need for robust safety mechanisms. It also outlines future development directions, emphasizing the importance of multi-technology integration and all-weather adaptability for complex environments. Achieving these goals will not only advance the technological capabilities of autonomous agricultural machinery but also serve as a critical reference for promoting the widespread adoption of unmanned farming systems in China. This, in turn, will enhance operational quality and efficiency, driving progress in the modernization of agriculture.