Double junction tandem solar cells consisting of two absorbers with designed different band gaps show great advantage in breaking the Shockley-Queisser limit efficiency of single junction solar cell by differential absorption of sunlight in a wider range of wavelengths and reducing the thermal loss of photons. Owing to the advantages of adjustable band gap and low cost of perovskite cells, perovskite/crystalline silicon tandem solar cells have become a research hotspot in photovoltaics. We systematically review the latest research progress of perovskite/crystalline silicon tandem solar cells. Focusing on the structure of perovskite top cells, intermediate interconnection layers and crystalline silicon bottom cells, we summarize the design principles of high-efficiency tandem devices in optical and electrical aspects. We find that the optical and electrical engineering of each layer structure in perovskite/crystalline silicon tandem solar cells goes through the whole process of device preparation. We also summarize the challenges of limiting the further improvement of the efficiency of the perovskite/crystalline silicon tandem solar cells and the corresponding improvement measures, which covers the following respects: 1) Improving the balance between <i>V</i><sub>oc</sub> and <i>J</i><sub>sc</sub> of the broadband perovskite cell through additive engineering and interface engineering; 2) improving the bandgap matching between the electrical layers and reducing the carrier transport barrier through adjusting the work function or conductivity of layers; 3) improving the photocurrent coupling between sub-cells and the photocurrent of tandem solar cells by using light engineering and conformal deposition technology of perovskite cells. At present, there have been many technologies to improve the stability of perovskite solar cells, such as additive engineering and interface engineering, but the problem has hardly been solved. Therefore, improving the stability of broadband gap perovskite solar cells to the level of crystalline silicon solar cells will become an important challenge to limit its large-scale application. In terms of efficiency, the mass production efficiency of perovskite/crystalline silicon tandem solar cells is far lower than that of the laboratory level. One of the reasons is that it is difficult to achieve low-cost and deposition of uniform large area perovskite solar cells. Therefore improving the stability of broadband gap perovskite solar cells and developing low-cost large-area perovskite deposition technology will become extremely critical. Finally we look forward to the next generation of higher efficient low-cost tandem solar cells. We believe that with the increasing demand for higher efficiency photovoltaic devices, the triple junction solar cells based on the perovskite/crystalline silicon stack structure will become the future photovoltaics.