中文摘要：

基于模型预测控制（MPC）理论的智能车纵向速度控制问题可以转换为二次规划问题（QP）。针对该QP问题,利用一种改进的粒子群算法（IPSO）减少MPC计算成本。通过引入收缩因子保证粒子群算法收敛,引入惯性因子避免粒子在全局最优解附近振荡,引入“备胎机制”来处理QP约束。数值试验验证了改进的IPSO算法可减少迭代次数、降低计算成本。将IPSO算法与MPC算法结合形成IPSO-MPC算法,智能车纵向速度控制仿真结果证明IPSO-MPC算法有效。

英文摘要：

The problem of longitudinal velocity control of intelligent vehicle based on model predictive control （MPC） theory can be converted to quadratic programming（QP）. Considering this QP problem, an improved particle swarm optimization （IPSO） algorithm is applied to reduce MPC computational cost. In this paper, shrinkage factor is introduced to guarantee particle swarm algorithm convergence, the inertia factor is introduced to avoid the particle oscillating around the global opti- mal solution, and ＂spare tire mechanism＂ is introduced to deal with QP constraint. Numerical experiments show that the improved IPSO algorithm can reduce iterations and computational cost. IPSO - MPC algorithm is formed of IPSO algorithm and MPC algorithm, and the simulation result of intelligent vehicle longitudinal velocity control verified its effctiveness.