改进扩展卡尔曼滤波算法在智能足球机器人比赛中的应用

Application of extended Kalman filter algorithm in intelligent robot soccer competition

为了减少系统延时对高实时性机器人足球比赛的影响,在深入研究卡尔曼滤波算法原理的基础上,提出了一种改进的扩展卡尔曼滤波(EKF)算法,建立了足球机器人竞赛中小球的运动模型,通过仿真实验给出了小球运动状态的预测轨迹。实验结果表明,改进后的EKF算法可以有效地解决以往算法在高度非线性化区域的不稳定性等问题,同时改进后的算法提高了系统实时性,算法易于实现且预测效果较好。

In order to reduce the system latency affection on the high real-time robot soccer competition,a kind of improved extended Kaman filter（EKF） algorithm was proposed based on a deep insight into the basic Kalman filter algorithm. Also the corresponding ball motion model was established and the simulation results were shown. The simulation results show that the improved EKF algorithm can solve the instability problem effectively in the nonlinear area and the real-time characteristic is improved. The improved EKF algorithm can achieve a better performance on the location forecast of the ball and it is esay to implement.