I have a trajectory planner, which calculates the trajectory assuming that the robot is following the planned trajectory. From sensors I get the position of the robot, which has an offset from the planned trajectory. I would like to define some limits, and when the "error" is over these limits I would like to send a reset signal to the trajectory planner in order to replan from the actual position of the robot.

Measurements that I can use:

  • Position (x, y)
  • Orientation ( $\theta$)
  • Velocity longitudinal (v)
  • Acceleration longitudinal (a)

The robot is a car-like robot, so my thought goes like this:

  1. Vertical distance $\Delta y$, has the highest priority, in order not to bump on obstacles.
  2. Distance from the point on trajectory $\sqrt{\delta x + \delta y} $
  3. $\Delta \theta$
  4. Velocity and acceleration are not so important.

First of all I would like to know if I am in the correct way about this assumptions, and if there is some literature that I could read on this topic.

Second how can I determine the limits for this errors? I was thinking for y taking the half of the robots' width.

Thanks in advance



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.