The research on picking robots is vital to the transformation and upgrading of the agroforestry industry and the revitalization and development of rural areas. This paper examines the research field of agroforestry picking robots by meticulously combing and analyzing 623 CNKI and 648 WoS core literature from 2004 to 2022 selected in China Knowledge Network (CNKI) and Web of Science (WoS) databases using Cite Space 6.1R3 software. The analysis includes the quantity of literature, issuing countries, organizations, keywords, keyword clustering, emerging terms, etc. On this basis, research hotspots in the field of agroforestry picking robots are identified, such as research based on the identification of picking targets, the control of motion planning, structural design and simulation, and the planning of walking paths. This paper analyzes and discusses these research hotspots and main lines, providing a reference for future studies in this field. This bibliometric approach can provide comprehensive literature information for research in related fields, as well as identify and summarize the major research hotspots in a shorter time, allowing new researchers to enter the field more quickly and obtain more valuable scientific information.