The maturity of fresh fruits and vegetables is inconsistent, requiring selective harvesting based on indicators such as color and size, which consumes the most labor and becomes a bottleneck affecting the development of the fruit and vegetable industry. Selective harvesting technology (SHT) is an important research field of agricultural robots, which can reduce labor costs and increase fruit and vegetable profits, and has become an important direction for the development of fruit and vegetable harvesting technology in the world. The SHT of fresh fruits and vegetables, including the representative underground parts such as white asparagus and the representative aerial parts such as apples, strawberries, tomatoes, has accelerated iteratively and has become a research hotspot of agricultural robots in recent years. It focused on the development of SHT with market-oriented prospects in the past three years, and sorted out the implementation path, application objects and development context of technology research and development. It focused on the common key issues of end effector and harvesting mechanism, harvesting target recognition and positioning technology, and selective harvesting collaborative control technology, and summarized the open problem of technical research in this field. Finally, it summarized the challenges and opportunities faced by SHT. Aiming at the application scenarios of dehumanized or unmanned fruit and vegetable production, it pointed out that the future development of the industry and the implementation of technology needed to be balanced.