Abstract
This paper proposes a method to improve the robustness
of a robot arm control program generated by genetic
programming. The idea is to evolve the control program
with multiple targets in the uncertainty region. The
result shows that the robustness of a control program
is improved with the number of targets in the training
region. The analysis showed that a program trained with
more targets would have more paths that contribute to
its robustness.
Users
Please
log in to take part in the discussion (add own reviews or comments).