Abstract: The existing safety distance model is established based on longitudinal relative speed or deceleration, without considering the lateral motion characteristics of moving target. In this paper, the relative speed and position of the moving target crossing the road are used to set up lateral safe distance model, and an active obstacle avoidance algorithm based on lateral safe distance model is proposed. Firstly, a lateral safety distance model is established according to the relative speed and position of the moving target crossing the road and the braking distance of ego vehicle, and the active obstacle avoidance algorithm is designed. Then, considering the influence of road conditions on braking effect, the predicted peak adhesion coefficient of current road surface is introduced to estimate the maximum braking deceleration, with the expected obstacle avoidance deceleration constrained to adjust braking intensity for adapting to the safe obstacle avoidance driving in different road conditions. Finally, with the typical lateral moving target-cyclist as research object, a PreScan/Simulink/Carsim co-simulation is conducted to verify the effectiveness of the obstacle avoidance algorithm. The results show that the active obstacle avoidance algorithm based on lateral safe distance model can effectively avoid vehicle collision with cyclist, so enhancing the active safety of driving

Key words: lateral moving target, lateral safe distance model, active obstacle avoidance algorithm, peak adhesion coefficient, active safety