dsp_
Member level 3
- Joined
- Mar 2, 2002
- Messages
- 58
- Helped
- 3
- Reputation
- 6
- Reaction score
- 0
- Trophy points
- 1,286
- Location
- right here ... right now ...
- Activity points
- 525
inverse kinematics
A group of ours does a simple robot project which has to do with inverse kinematics.
since the arm is built that simple(2 joints) we could reduce calculations to 2d space with a closed form solution ..
now i'm looking for algorithms/advices which are optimal for an 8051 derivate..
i think there will be a problem with floating points because our rom is limited to 4k and an extension would take too much ports.
the 2 solutions I considered are:
taking the equation of the "rather big" closed form solution an puttin in the variables for the numeric solution.
iterate trought puttin an virtual force on a model of the arm until it is in a specified range.
next problem would be software pwm'ing through a shift register while receiving data on rs232(probably I will just clear the cts line)
any help is appreciated
TIA
dsp_
A group of ours does a simple robot project which has to do with inverse kinematics.
since the arm is built that simple(2 joints) we could reduce calculations to 2d space with a closed form solution ..
now i'm looking for algorithms/advices which are optimal for an 8051 derivate..
i think there will be a problem with floating points because our rom is limited to 4k and an extension would take too much ports.
the 2 solutions I considered are:
taking the equation of the "rather big" closed form solution an puttin in the variables for the numeric solution.
iterate trought puttin an virtual force on a model of the arm until it is in a specified range.
next problem would be software pwm'ing through a shift register while receiving data on rs232(probably I will just clear the cts line)
any help is appreciated
TIA
dsp_