“New Retail”has innovated the channel and sales model of unmanned fresh food retail. Based on research conducted on the technical advantages of networked unmanned selling, with an online dynamic pricing mechanism introduced, the flaw of a poor quality control and a low customer viscosity in unmanned fresh food sales have been overcome, thus effectively reducing the overall cost of the system. Meanwhile, a hybrid trunk-and-branch distribution model for unmanned retail of fresh goods has been constructed, with breakthroughs made in multi-objective solution, problem decoupling, and PSO global optimization. A location path model with time windows for secondary transportation under dynamic price constraints (2E-dPLRPTW) has been established, followed by a case study to verify the effectiveness of the proposed method in improving the efficiency of unmanned sales of fresh goods.