This article introduces an accurate, robust, and efficient direct yaw-moment control (DYC) architecture for four-wheel independently actuated (FWIA) autonomous vehicles via the nonsingular terminal sliding mode (NTSM) control method. To improve the DYC performance of FWIA autonomous vehicles for path tracking under critical maneuvers, a more accurate vehicle lateral dynamics model that considers the tire nonlinear mechanics under combined conditions is established first. On this basis, a method based on phase plane theory is introduced to analyze the vehicle lateral stability, and the two optimal external yaw moments are calculated by using the NTSM control algorithm that takes the reference yaw rate and the reference sideslip angle as the control objectives, respectively. Then, the external yaw moment finally applied to the vehicle is the weighting of the two calculated values, and the weight coefficient is optimized in real time by the particle swarm optimization method according to the vehicle lateral instability analysis results. A real-time and fast torque distribution method is further proposed to achieve the target yaw moment, in which the tire force saturations, the adaptive weight coefficient of each wheel, and the partial derivative of the target function are involved. Simulation results are, finally, provided to verify the effectiveness of the proposed controller in improving the lateral stability of FWIA autonomous vehicles for path tracking under four extreme conditions and eliminating the steady errors of the vehicle target yaw rate and sideslip angle.