Shanghai University has successfully developed China's first dual-arm humanoid robot designed for tomato harvesting, marking a significant breakthrough in embodied intelligence and agricultural automation. Led by Professor Miao Zhonghua from the School of Mechanical Engineering and Automation, the team has created a system that integrates autonomous navigation, large AI models, and advanced mechanical control to address the complex challenges of smart farming. The robot features a self-developed control system that combines a dual-purpose tracked and ground mobile base with 7-degree-of-freedom anthropomorphic arms. This hardware design allows for efficient track switching and crab-style steering, enabling the machine to navigate narrow greenhouse spaces and operate on both rails and the ground. At the core of the robot is a sophisticated intelligence layer that merges deep learning, active visual perception, and AI large models. This enables the system to perform real-time posture estimation, precise target recognition from multiple angles, maturity grading, and accurate stem position calculation. The robot can also coordinate both arms to harvest continuously and switch sides independently without human intervention. Through advanced task decision algorithms and intelligent trajectory planning, the system achieves full-process autonomous operation with real-time status monitoring and adaptive parameter adjustment. The technology was recently tested at large-scale tomato planting bases in Chongming, Shanghai. Field trials demonstrated a harvesting success rate and full-area coverage rate exceeding 90 percent. This achievement represents a culmination of over a decade of research by the team in robotics and intelligent control. It highlights Shanghai University's strategic alignment with the national "Artificial Intelligence Plus" initiative and its commitment to integrating scientific research with industrial applications. The development leverages the school's strengths in mechanical engineering and automation to provide a green, efficient, and secure solution for the modern agricultural sector. Looking ahead, the research team plans to focus on three key areas for further optimization: continuous operation precision, stability of full-process autonomous execution, and the ability for intelligent self-evolution. These improvements aim to accelerate the industrialization and commercialization of agricultural robots. The project is expected to provide high-end intelligent equipment support that enhances the quality and sustainability of China's smart agriculture. The success of this innovation has attracted significant attention from major media outlets, including Xinhua News Agency, the Wenhui Daily, and Shangguan News. The development serves as a landmark achievement for the school, demonstrating how deep technological integration can drive practical solutions for real-world industrial needs. By combining cutting-edge AI with mechanical engineering, Shanghai University continues to contribute to the advancement of high-end intelligent equipment and the digital transformation of traditional industries.