Abstract: The shift point of electric vehicle automated mechanical transmission (AMT) system has an important impact on the power and economy level of the whole vehicle. In view of the poor adaptability of electric vehicle AMT look-up table shift law under working conditions, which cannot be solved quantitatively based on working conditions, and does not take into account power and economy, a real-time optimal control method of electric vehicle AMT shift point based on greedy strategy was proposed. Firstly, the driving condition of electric vehicle was predicted by Markov chain, and then greedy algorithm was used to optimize the shift point. Under world light vehicle test cycle (WLTC) and China automotive test cycle (CATC) conditions, the simulation results show that compared with the electric vehicle with economic shift strategy, the electric vehicle with greedy algorithm shift strategy consumes less power, the battery SOC decreases more slowly, and the economy of the vehicle is improved.