摘要
针对单神经元自适应PID控制算法在电磁导航智能车角度偏差处理中存在输出误差和控制增量大的缺点,提出了基于二次型性能指标学习算法的单神经元自适应PID控制算法。在加权系数的调整中引入二次型性能指标,使输出误差和控制增量的加权平方和为最小来调整加权系数,从而间接实现对输出误差和控制增量加权的约束控制。Matlab仿真测试表明,与单神经元自适应PID控制算法相比,二次型性能指标学习算法在智能车角度控制中具有响应快,超调量小、鲁棒性和适应性强的优点,大大提高了智能车舵机控制系统的性能。
In view of the single neuron adaptive PID control algorithm in electromagnetic navigation of intelligent vehicle output errors exist in the angular deviation processing and control incremental big shortcoming, put forward based on quadratic performance index is proposed learning algorithm of single neuron adaptive PID control algorithm. In the adjustment of the weighted coefficient of the introduction of quadratic performance index, make the output error and control incremental weighted sum of squares to a minimum to adjust the weighting coefficient, and indirectly, to realize to control the output error and incremental weighted constraint control.Matlab simulation tests show that compared with single neuron adaptive PID control algorithm, the quadratic performance index of learning algorithms in smart car Angle control has quick response, less overshoot, robustness and strong adaptability,the advantages of greatly improving the performance of the intelligent vehicle steering gear control system.
出处
《自动化与仪器仪表》
2015年第11期178-179 182,共3页
Automation & Instrumentation
基金
甘肃省高等学校研究生科研项目基金(1115-02)
关键词
单神经元自适应PID控制
二次型性能指标
电磁导航智能车
舵机控制
MATLAB仿真
Single neuron adaptive PID control
Quadratic performance index
Electromagnetic navigation intelligence car
Steering gear control
Matlab simulation