Nearest Neighbor Migration in Parallel Genetic
Programming for Automatic Robot Programming
Proceedings of the Sixth International Conference on
Control, Automation, Robotics and Vision, Singapore, (декабря 2000)Аннотация
This work presents a study of parallelization of
genetic programming for automatically creating a robot
control program in a mobile robot navigation problem. A
nearest neighbor migration topology is proposed to
reduce the communication time. This study compares the
performance both in terms of the solution quality and
the gain in execution time. The timing analysis is
investigated to give insight into the behavior of
parallel implementations. The results show that the
parallel algorithm with asynchronous migration using 10
processors is 32 times faster than the serial
algorithm.