2
$\begingroup$

I am looking to plan a solution for a workspace fault and not hardware faults.

Consider a task where a robot has to move balls from one place to another. In case it faces any condition which is outside the task for eg. someone snatches the ball from the robot while it is transferring or the robot drops the balls in between. These are some example faults that could occur, many other might be possible. I am trying to build a generalized algorithm that so that the robot can find a way to resolve unexpected changes itself.

I currently have an FSM for the whole task. Any fault that somehow changes any of the state machine variables should be considered. For instance, there are faults that deal with obstacles that may come in the way.

But there might be faults for example a cloth in front of the camera. This fault should be corrected by a human since the robot cannot manage that. All the faults like that are out of the scope of the robot.

Any suggestion or ideas related to the algorithm will be helpful.

$\endgroup$
  • $\begingroup$ The second one. In the first statement, I would need to know the faults which I might be not fully aware of. I am building an algorithm that is independent of the type of fault. Initially the robot knows what the correct state is, with the help of the a state machine. $\endgroup$ – Sam Sep 20 at 13:54
  • $\begingroup$ There are faults that deal with obstacles that come in the way. But there might be faults for example a cloth in front of the camera. This mentioned fault you stated should be corrected by a human since the robot cannot manage that. Any fault that somehow changes any of the state machine variable shall be considered. All the faults apart from that are out of the scope of the robot. $\endgroup$ – Sam Sep 20 at 14:26
  • $\begingroup$ Thanks for the clarifications. I have attempted to edit those into the question (this is standard on Stack Exchange, to clarify a question by using edit ). Feel free to adjust it further in case I have changed things to say something you did not really mean. $\endgroup$ – Neil Slater Sep 20 at 16:14
  • $\begingroup$ Thats very nice of you $\endgroup$ – Sam Sep 20 at 17:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.