How to update matplotlib pyplot on every iteration

问题:

I am using rospy to get data of my robot's position and plot this in real time.

This is what I have:

 self.plot_pose()

 def plot_pose(self):
     plt.plot(self.pose[0], self.pose[1], 'o', color='green')
     plt.plot([self.pose[0], self.pose[0] - 0.5*np.cos(self.pose[2])],
              [self.pose[1], self.pose[1] + 0.5*np.sin(self.pose[2])],
              'k-', color='red', lw=2)

     plt.show(block=False)
     plt.pause(0.0001)

Unfortunately this doesn't erase the plot but overlays everything. So I tried using

plt.clf()
plt.cla()

the first one gives me deprecation error and the second one gives me a blank plot for some reason. I am using python2.7 and rospy library.

Any suggestions on how to update the plot would be greatly appreciated.

回答1:

This is the same as usual with animations. Try to avoid creating new plots every interation and instead update the old one.

self.point, = plt.plot([],[], 'o', color='green')
self.line, =  plt.plot([],[], ls="-", color='red', lw=2)
plt.show(block=False)
self.plot_pose()

def plot_pose(self):
    self.point.set_data(self.pose[0], self.pose[1])
    self.line.set_data([self.pose[0], self.pose[0] - 0.5*np.cos(self.pose[2])],
                       [self.pose[1], self.pose[1] + 0.5*np.sin(self.pose[2])])

    plt.pause(0.0001)

You may need to adjust the limits of the plot (plt.xlim(), plt.ylim()) if the points are outside of it.