基于卡尔曼预测估计器的输入加权预测控制算法

Weighted Input Model Predictive Control Algorithm Based on Kalman Predictor and Filter

针对噪声环境下输入带有约束的系统,传统的方法要处理一个二次规划问题.本文提出用最小化无约束二次性能指标得到的输入控制量的加权和代替该时刻的输入作用于系统.用这种方法将不存在不可行问题,使得控制输入的振荡范围减小,能大大减小违反约束的机率.仿真结果证明了算法的有效性.

Aiming at the system of the input with constraints under noisy condition, the conventional method is necessary to solve a Quadratic Programming （QP） problem to which a new method is proposed. Available at each time of control inputs are calculated and made available at each time instant, the actual input applied being a weighted summation of the inputs within the set instead of the first element of the input vector. This can reduce the variation range of input greatly and the possibility of violating constraints. Computer simulation proves the effectiveness of the method presented.