Abstract: An emergency collision avoidance strategy integrating active braking and active steering is proposed for complex conditions in this paper. Firstly, according to the vehicle speed and road adhesion coefficient and combining with the braking and steering safety distance model, a steering and braking priority that considers the effect of frontal obstacles is obtained. Based on this and in view of the other traffic factors in the side lane, the collision avoidance mode under steering priority is further subdivided into three types: steering, braking and steering plus braking. For collision avoidance by steering, a fifth-degree polynomial is used for path planning. Based on the safety and smoothness cost functions, an optimal safe and smooth desired path is comprehensively selected, and then achieving path tracking by using feedforward and LQR feedback control. A driver-in-the-loop simulation is conducted to verify the effectiveness of the active collision avoidance algorithm proposed. The test results under three typical operating conditions show that the intelligent vehicle can make reasonable decision on collision avoidance mode according to different emergency operating conditions, and can successfully complete the active collision avoidance operations of steering, braking and steering plus braking. Compared with the path generated by Sigmoid function, the planned path based on the fifth-degree polynomial is more suitable for emergency collision avoidance

Key words: intelligent vehicle, emergency collision avoidance, path planning, path tracking, driver-in-the-loop simulation