 | | Re: Short term memory ablut sensor input???
|
|
(...) never mind. I've done this once, to prove, that you can build a little robot, which won't get stuck in any kind of corner. Changing strategie, if it detects a corner, in which it was stuck. an programm it only with this kid programming (...) (24 years ago, 21-Sep-00, to lugnet.robotics.rcx)
|
| |
| | Re: Short term memory ablut sensor input???
|
|
(...) That's true. You can't remember something like a light-sensor reading, though. My thinking is: if you're at the point where you want to be doing stuff like this, you're probably ready to graduate to a better system! (24 years ago, 21-Sep-00, to lugnet.robotics.rcx)
|
| |
| | Re: Short term memory ablut sensor input???
|
|
Agreed, LegOS is the way to go. Here is a LegOS program I wrote for a robot that keeps track of its heading in one of 8, increments (45 degrees intervals, numbered 0 to 7 heading clockwise). It keeps track of the last 10 segments in it's travels (...) (24 years ago, 21-Sep-00, to lugnet.robotics.rcx)
|
| |
| | Re: Short term memory ablut sensor input???
|
|
Can you remember a touch sensor reading? This way by using only one light sensor as an irpd, I can detect which way to turn to avoid the obstacle. (...) */ (...) turning */ (...) */ (...) travel */ (...) */ (...) ------ (...) */ (...) ------ (...) (...) (24 years ago, 21-Sep-00, to lugnet.robotics.rcx)
|
| |
| | Re: Short term memory ablut sensor input???
|
|
(...) The program doesn't actually record in the Vectors array any information about where it saw an obstacle. It just records which direction the robot went and how far (according to the rotation sensor) during the last leg of the journey. Each (...) (24 years ago, 22-Sep-00, to lugnet.robotics.rcx)
|
