PmWiki | Student / ChrisArndtLab1Write-up

Hypothesis:

When dealing with consecutive commands, I believe that the Create will react almost instantaneously and execute the following command as soon as it has finished executing the previous command (again, almost instantaneously).

Materials:

iRobot Create
1 Computer
1 Serial Cable
Realterm

Procedure:

1. I will write a script that had the Create drive at a slow speed, then drive at a faster speed. This will allow me to view how quickly it will move to the next command, by timing how long it stays at the slower speed.
2. This script is:
145 0 50 0 50 145 0 250 0 250
3. If this is instantaneous or something too fast to time, I will write a second script. This second script will light the advance, play, and power buttons one at a time, in order, creating a ripple effect.
4. This will show if the time between the commands is too small to measure accurately or if the Create just skips the first two commands.

Data:

Experiment 1:
The robot did not appear to change speed. The second command was executed quickly enough that there was one constant acceleration from 0-250 mm/sec, there was no distinguishable change in speed and, therefore the two consecutive drive commands were executed too quickly in sequence to be measured by a mere stopwatch and human reflexes.
Experiment 2:
Once the commands were submitted to the robot, the advance, play, and power buttons all flashed in a ripple effect, leaving the power LED on after the other two had flashed. This shows that the robot does not skip the first two consecutive commands. It executes them all, if only for a split second.

Observations and Conclusions:

The robot will execute the commands as quickly as it can. This is restricted by the processor and the speed of the serial cable transmitting information from the computer to the Create. The iRobot Open Interface Manual states that there must be at least 200 microseconds between each character (bit) at a baud rate of 115200 BPS, otherwise some characters may not be received. This is about 1,000 times faster than the blink of an eye. My hypothesis was both correct and incorrect. The robot executed the commands as quickly as it could, almost instantaneously. However, it did not execute the following command as soon as the previous command had finished. There is a 200 microsecond gap between the two commands. This is, however, immeasurable by human reflexes. A script that could demonstrate one's control over the speed at which consecutive commands are executed is:
152 12 145 0 75 0 75 155 40 145 0 0 0 0 153

Here's a Drive to Drive Direct command converter Δ

Back to my page->ChrisArndt