Video of Performance Test 4
Preparations
For the fourth performance test, the robot needed to press and hold the correct fuel button for five seconds, navigate back to the lower level, and press the final button. Additionally, bonus was awarded for controlling the supplies.
In preparation for the fourth and final performance test, the biggest thing that needed to be perfected was the RPS X, Y, and heading codes. Currently, these X and Y codes worked, the heading code does not, and all three needed to be improved in an effort to reduce time and minimize error. To fix the heading code, the focus was shifted onto whether the robot should turn clockwise, counterclockwise, or not at all. This was done instead of trying to tell the robot to go to a specific heading. This ended up solving the issue after much testing. To streamline the RPS functions, the robot did things for a time (an approximation of when they would be done), then refined with time. Additionally, the robot would pulse on the X- and Y- motions when close to the range, moving continuously if further away. This minimized the time spent traveling and adjusting to get into desired range of values.
Once these algorithms and codes were sorted out, key RPS values on the course were taken so that we knew where we needed to send our robot. If the robot tried to move at too slow of speeds, it would drift off of its desired course, which was an issue. To account for this at the fuel buttons and supply pickup zone, the robot would then run back into the wall to fix its heading.
An unexpected issue arose during testing regarding the CdS cell. The blue light was showing up as 2.2 V, while our maximum threshold was 1.8 V. This was the first time that the blue light had ever shown up this high, and it was discovered that the robot would do nothing if the value was outside of the range by adding a print statement to the screen. To fix this, the blue threshold was increased so that the robot would press the blue fuel button if it did not make it to the light.
Next, one of the motors stopped working randomly during runs. It would work fine during some runs, not work at all during others, and work at random points during more. This was very concerning as it nearly damaged the budget substantially with the potential purchase of a $15 motor with $17 left and work to be done, but re-tightening the motor terminal screws fixed this issue.
Once all these issues were sorted out and a few more RPS heading checks were added to prevent the robot from going off course, the robot successfully completed the performance test with the stretch bonus. A full description of the tests made in preparation is given in the testing log. An image of the robot at this phase is given below.
Live Action Image of Performance Test 4 Passer (3/23/2016)
Path Taken
For this performance test, the robot first left the start platform on the start light. Then, it used RPS to drive over to and pick up the supplies with the magnet arm on a servo. Next, it drove up the temporary access ramp and read the fuel light. Then, it pressed the correct fuel button (in this case red) and held it for five seconds. Finally, it drove down the main ramp and pressed final button, still holding the supplies. An image of the path taken is given in the diagram below.
Performance Test 4 Path