More physics

shell.nix

@@ -0,0 +1,12 @@
+let
+	pkgs = import <nixpkgs> {};
+in
+pkgs.mkShell {
+	packages = [
+		(pkgs.python3.withPackages (ps : [
+			ps.numpy
+			ps.pyaudio
+			ps.matplotlib
+			ps.scipy]))
+	];
+}

simulator.py

@@ -3,12 +3,12 @@ import os
 import time
 import numpy as np
 import matplotlib.pyplot as plt
-from colour import Color
+#from color import Color
 
 
-c1 = Color("#7d1dd3")
-c2 = Color("#ffe500")
-colors = list(c1.range_to(c2, 42))
+#c1 = Color("#7d1dd3")
+#c2 = Color("#ffe500")
+#colors = list(c1.range_to(c2, 42))
 
 
 def infinite_colors():
@@ -116,7 +116,6 @@ class Douche:
         self.line, = ax.plot(self.xs, self.ys)
 
     def add_point(self, x, y):
-        # print(x, y)
         self.xs.append(x)
         self.ys.append(y)
         self.line.set_data(self.xs, self.ys)
@@ -128,11 +127,16 @@ class NaiveSimulator(Simulator, Douche):
     steps = [0, 0, 0]
 
     time_step = 0
-    last_update = 0.
+    last_update = np.nan
 
-    J = np.array([1., 1., 1.]) # Moment of inertia vector times 1/R
+    # We have on each axis
+    # Jw' = Gamma(motor) - f0*w
+    # which can be expressed as
+    # Jv' = (R*Gamma)(motor) - f0*v
+
+    J = np.array([1., 1., 1.]) # Moment of inertia vector
     f = np.array([0., 0., 0.]) # viscous friction coefficient
-    G = np.array([1., 1., 1.]) # Motor gain : Gamma = G * delta_w
+    G = np.array([1., 1., 1.]) # Motor gain : Gamma*R = G * delta_v
     # FIXME: this is not very realistic, this looks like a model of
     # an asynchronous motor, not a stepper.
 
@@ -148,11 +152,17 @@ class NaiveSimulator(Simulator, Douche):
     def simulate(self):
         lt = time.time()
         self.time_step = lt - self.last_update
+        if(self.time_step == np.nan):
+            return
         self.last_update = lt
         gamma = self.G*(self.command_spd - self.speed)
-
-        self.speed += 1./self.J * (gamma - self.f*self.speed ) * self.time_step
+        
         self.pos += self.speed * self.time_step
+        self.speed += 1./self.J * (gamma - self.f*self.speed) * self.time_step
+        print("timestep ", self.time_step)
+        print("gamma ", gamma)
+        print("speed ", self.speed)
+        print("position ", self.pos)
 
         self.steps = np.ceil(self.pos * np.array([self.X_STEPS_PER_MM,
                                                   self.Y_STEPS_PER_MM,