A nonlinear control allocation algorithm for DYC in distributed-motor drive electric vehicles using S-SQP

The yaw stabilization in distributed-motor drive electric vehicles is usually based on hierarchical control structure, namely an upper-level motion following controller and a lower-level control allocation scheme. The existing strongly coupling property of tire forces gives rise to extreme difficulty in solving the control allocation problem for its nonlinearities in the moment/effector relationships, especially under critical cornering. In this paper, an improved approach is presented for the solution of the control allocation where the control variable rates or moment are nonlinear functions of control position. A single-iteration sequential quadratic programming method based on simplified tire model is exploited to solve the control allocation problem. The performance of this approach will be compared with the traditional approach which utilizes linear efficiency matrix without considering the tire coupling property by simulation. The results show that the yaw moment generated by the change of lateral force cannot be neglected, which will affect both the allocation accuracy and control energy. The single-iteration sequential quadratic programming method that satisfies the real-time requirement not only efficiently achieves four wheel longitudinal forces according to the wheel states, but also significantly reduces the allocation errors. Additionally, control energy consumption is lower when compared with the approach using linear efficiency matrix.