Minor moves-global results: robot trajectory planning
Heikki Hyötyniemi · [1990] Proceedings of the 2nd International IEEE Conference on Tools for Artificial Intelligence · 2002
Two simple algorithms are used to find the global optimum in a parameter space of high dimension. Both algorithms are based on local optimization-the first one uses the gradient method, the other one modified Kohonen nets. The main topic of interest is how the qualitative features of the global optimum emerge as a consequence of iterated minor parameter adjustments. Planning the trajectory for a spray-painting robot is considered as an application example.>