alpha beta滤波器
The g and h coefficients refer to the two scaling factors, where g is the scaling we used for the measurement, and h is the scaling for the change in measurement over time .
g和h系数指的是两个缩放因子,其中g是我们用于测量的缩放比例,h是测量值随时间变化的缩放比例。
Take tracking a one-dimensional uniform speed object as an example.
以跟踪一维匀速物体为例。
Measurement is typically denoted z .Subscript k indicates the time step, so z_k is the data for this time step and x denotes our state and v the speed of the object that is the derivative of distance:
测量通常用z表示。下标k表示时间步长,因此z_k是该时间步长的数据,x表示我们的状态, v表示物体的速度,该速度是距离的导数:
The sensor sends a tracking beam in the direction of the target at a constant rate. The tracking period is Δt . Assuming a constant speed, the dynamic model of the system can be described by two motion equations:
传感器以恒定速率向目标方向发送跟踪光束。 跟踪周期为Δt。 假设速度恒定,则可以通过两个运动方程来描述系统的动力学模型:
The above equations are called State Extrapolation Equation , which is one of the five Kalman filter equations. This system of equations extrapolates the current state to the next state (prediction).
上面的方程式称为状态外推方程,它是五个卡尔曼滤波方程式之一。 该方程式系统将当前状态外推到下一个状态(预测)。
Now make a dynamic system that changes state over time. In this example, we will use the g-h filter to track a constant velocity object in one dimension.
现在创建一个动态系统,该系统会随着时间的推移而改变状态。 在此示例中,我们将使用gh滤波器在一个维度上跟踪恒速对象。
We assume that an object is moving in constant speed in a fixed direction.The state update equation of the object position:
我们假设一个物体在一个固定的方向上以恒定的速度运动,物体位置的状态更新方程为:
The speed state update equation:
速度状态更新方程式:
Where,\hat{x}_{n,n-1} represents the predicted state of x at time n based on the measurement at time n-1.
其中,\帽子{X} _ {N,N-1}表示x的预测状态在时间n在时间n-1的基础上进行测量。
The g-h filter is not one filter — it is a classification for a family of filters,each filter is differentiated by how g and h are chosen.
gh过滤器不是一个过滤器,它是一系列过滤器的分类,每个过滤器的区别在于g和h的选择方式。
Different filters choose g and h in different ways depending on the mathematical properties of the problem, some filters set g and h as constants, others vary them dynamically.
不同的过滤器根据问题的数学属性以不同的方式选择g和h,一些过滤器将g和h设置为常数,另一些过滤器动态地改变它们。
When g is larger we more closely follow the measurement instead of the prediction and h affects how much we favor the measurement of $\dot{x}$ vs our prediction.
当g较大时,我们会更密切地遵循度量而不是预测,而h影响我们对$ \ dot {x} $相对于我们的预测的支持程度。
A particular choice might perform well in one situation, but very poorly in another. Even when you understand the effect of g and h it can be difficult to choose proper values. In fact, it is extremely unlikely that you will choose values for g and h that is optimal for any given problem. Filters are designed, not selected ad hoc.
特定选择可能在一种情况下表现良好,但在另一种情况下表现很差。 即使您了解g和h的影响,也可能难以选择合适的值。 实际上,极不可能为任何给定问题选择最佳的g和h值。 过滤器是经过设计的,不是临时选择的。
翻译自: https://towardsdatascience.com/alpha-beta-filter-21d3276cf35e
alpha beta滤波器