Because of the sensitive factors, such as the larger loads, higher mass center and relatively narrow tread in comparison with the height of heavy vehicles that have the poor dynamic rollover stability. This paper set of anti-rollover LQR control algorithm based on early warning. The model-based early rollover warning algorithm utilize the TruckSim® models and early warning reference model to predict the impeding vehicle rollover time in advance and told the driver the warning signals so that drivers had enough time to take appropriate measures to prevent vehicle rollover that called time to rollover (TTR), thereby greatly improving the vehicles active safety performance of the heavy vehicles. As to the anti-rollover LQR control algorithm, the principle was to use the optimal additional yaw moment obtained from the control algorithm and then made it reasonable impose the corresponding wheels by taking the moment distribution methods based on the differential braking for the purpose of reducing the risk of rollover. The simulation results show that the algorithm was presented in this paper can effectively reduce the lateral load transfer ratio and actively void the occurrence of rollover accidents.