The LuGre Friction Model

---

一 摩擦力的分类

1. stiction:静态摩擦力：非线性变化，符号是跟方向有关
2. Columb:库伦摩擦力：常值，符号是跟方向有关
3. Viscous:粘性摩擦力：线性变化

---

二 LuGre摩擦力模型

2.1 LuGre的原型：

Lugre的物理模型是将摩擦力看成两个都有毛的平面接触时的相互力

2.2 LuGre的曲线轨迹

LuGre模型将静态摩擦力，库伦摩擦力，粘性摩擦力糅合到一起，依据速度值来计算出一个综合的摩擦力。

---

三 LuGre的matlab实现

3.1 计算的公式

因为公式里面出现了z以及z的导数dz/dt，因此计算有积分的操作。

---

四 不同速度下的LuGre表现

4.1 几乎静止下的LuGre表现

此状态下，一旦速度方向发生改变，摩擦力会发生越阶效果（静态摩擦力），如果在此区间内速度f方向不变，则几乎保持在一定值（库伦摩擦力）

此种状态下的计算，没有使用积分操作

sigma_0 = 1e5;
sigma_1  = sqrt(1e5);
sigma_2  = 0.4;
Fc = 1;
Fs = 1.5;
vs = 0.001;

%% Draw the overall friction force at steady state condition.
%  This is not shown in the paper

v = -0.005:0.0001:0.005;

Fss = lugref_ss(v, Fc, Fs, vs, sigma_2);

figure
plot(v, Fss)
grid
xlabel('Velocity (m/s)')
ylabel('Friction force (N)')
title('Friction force at steady state condition')

%% Zoom into certain velocity to see its transient behaviour 
%  This is not shown in the paper. We here want to demonstrate that the 
%  LuGre friction model does have a transient behaviour.

function Fss = lugref_ss(v, Fc, Fs, vs, sigma_2)
    r = -(v/vs).^2;
    
    % This equation is not lebaled on the paper, it is in the bottom left 
    % of the 2nd page 
    Fss = Fc*sign(v)+ (Fs - Fc) * exp(r) .* sign(v) + sigma_2 * v;      
end

4.2 一定速度下的LuGre表现

一开始的时候，摩擦力跟速度成线性比例，最后保持在一定的常值上

ts = 1e-3;
time_span = 20;
t_sol = 0 : ts : time_span;

v = 0.002;
z = 0;


% Note, this is forward euler integration of the ODE for z. - Jason Nicholson
F = nan(size(t_sol));
for j = 1 : length(t_sol)
    [F(j), z] = lugref(z, v, Fc, Fs, vs, sigma_0, sigma_1, sigma_2, ts);
end

figure
plot(t_sol, F);
grid
xlabel('Time (s)')
ylabel('Friction force (N)')
title('Friction force for v = 0.002 m/s')
xlim([0 0.1]);

%% Apply sinusoidal velocity and measure the friction force (Fig. 3)
% The input to the friction model was the velocity which was changed
% sinusoidally around an equilibrium. The resulting friction force is given 
% as a function of velocity .

function [F, z] = lugref(z, v, Fc, Fs, vs, sigma_0, sigma_1, sigma_2, ts)
    r = -(v/vs).^2;
    g_v = (Fc + (Fs - Fc) * exp(r)) ./ sigma_0;          % eq. 4
    z_dot = v - abs(v) .* z ./ g_v;                      % eq. 1 
    
    % note, this is forward euler integration. Your results depend heavily on the size of ts. - Jason Nicholson
    z = z + z_dot.* ts;

    F = sigma_0 * z + sigma_1 * z_dot + sigma_2 * v;    % eq. 3
end

注意，上面的实现，是使用简单的欧拉法来实现积分，很依赖积分区间的大小，比如我将积分时长改为ts=0.01,则是完全不同的效果

---

reference

[1] https://github.com/auralius/LuGre
[2] A new Model for control of systems with friction