基于两轮机器人运动模型的Kalman滤波视觉导航

在视觉算法对目标物识别的基础上，对两轮移动机器人运动模型进行分析，提出将其作为扩展卡尔曼滤波的系统状态方程，并应用到视觉导航系统中。仿真实验及实际测试表明，方法能较好地对两轮移动机器人的运动实现预测和估计，减轻环境干扰影响、降低处理计算量并提高系统实时性，满足了机器人在复杂环境下视觉导航的要求。

Kalman Filter for Vision Navigation Using Two-wheel Mobile Robot's Model

Based on the target recognition using vision algorithm, this paper analyzes the model of a two-wheel mobile robot and presents it as the system state equation of the extended Kalman filter which is applied to the vision navigation system. Simulation and experimental results show this method has a useful effect on prediction and estimation of a two-wheel robot movement, and satisfy the needs of robot vision navigation in a complex environment by eliminating the inference, reducing the computational complexity and improving the efficiency of the system.