YAY! Search algorithm issues have been resolved. After several weeks of tweaking this part of the code for the robot we've finally got it acting in a manner that we'd like. The issue before is that the robot would sweep past the target several times before noticing it there or it would notice the target and when approaching the target the angle wouldn't be good and the robot would then miss the target and go back to searching again.
The Fix:
A piece of code was inserted that had the robot sweep the the right until it noticed and object and then to continue sweeping until no more object was found. A beep would then sound to indicate that the robot found the right most edge of the object. The robot would then begin to sweep in the opposite direction until it past the object and would sound a beep indicating it has located the left most edge of the target.
Now the robot would take that distance, divide by 2 and move that number back to the right. This positioned the robot nearly centered to object every time and then the robot would approach the object without missing it, perform a function as desired depending on the type of target and then continue its way throughout the building.
Now that we've got this underway the robot is complete and ready for presentation!
Saturday, June 1, 2013
Week 8
New targets have been constructed to aid the robot in locating the targets better. A small target was taking to long to locate and the robot would sweep past it several times before noticing it was there.
The new targets are larger squares that were marked with different colors so the robot can determine which target is which and take the correct actions.
As mentioned last week our robot's search code is still not where we'd like it to be. While the robot can perform the functions intended inside the room it isn't performing as well as we'd like. The code for this area still needs some work and it is the only thing left for the completion of this project.
Over the next week we hope to have the rest of the bugs worked out of this part of the code and be ready for presentation!
The new targets are larger squares that were marked with different colors so the robot can determine which target is which and take the correct actions.
As mentioned last week our robot's search code is still not where we'd like it to be. While the robot can perform the functions intended inside the room it isn't performing as well as we'd like. The code for this area still needs some work and it is the only thing left for the completion of this project.
Over the next week we hope to have the rest of the bugs worked out of this part of the code and be ready for presentation!
Week 7
We've been running into issues with our code as of late and haven't quite been able to get the robot to react as intended once inside a room. From the testing we've done we have decided to make the target the robot is looking for bigger to help with this issue.
So far we've got the code pretty solid for navigating to the room and to begin the search algorithm. Once the robot gets in the room it takes far to long to locate and object, the search pattern is incredibly inefficient and we're working on bringing that up to speed.
The good new is that we have some ideas on how to fix it and will be trying out 2 different algorithms in the next coming weeks and see how that affects the robots performance.
So far we've got the code pretty solid for navigating to the room and to begin the search algorithm. Once the robot gets in the room it takes far to long to locate and object, the search pattern is incredibly inefficient and we're working on bringing that up to speed.
The good new is that we have some ideas on how to fix it and will be trying out 2 different algorithms in the next coming weeks and see how that affects the robots performance.
Subscribe to:
Posts (Atom)