Research On The Path Optimization Technology In Labyrinth Using PCNN

As the development of biological neurological, the research on a new artificial neural network, which called Pulse Coupled Neural network (PCNN), is being warm up gradually, because it’s model has own comprehensive space-time characteristics. PCNN in optimum combination has been applied the most in solving the shortest path problem. Caulfield and Kinser first put forward to solve labyrinth problem with PCNN model. Based on that, some ways were put forward to solve labyrinth problem. But their different is that the labyrinth problem is abstracted into an image or adjacency matrix to solve the problem.Based on the former research on PCNN and combining A star heuristic search strategy, the thesis designs labyrinth intelligent optimization algorithms based on a modified PCNN model, and it is applied to solve labyrinth path, specific as follows:1. Through the study of the inner mechanism of the PCNN model and its own comprehensive space-time characteristics, the suitable PCNN’s parameters will be chosen to solve the shortest path problem of labyrinth. The experimental results show that the improved PCNN model don’t only retain the original PCNN model of structure and characteristics, but also be more suitable for solving the problem of optimal path labyrinth.2. Through the research on the labyrinth of model, the state space representation, search strategy, evaluation function and search efficiency, the thesis finally choose A star search algorithm which combined with the improved PCNN model, and demonstrates the superiority of the labyrinth of model based on PCNN intelligent optimization algorithm by the way of algorithm design, algorithm simulation and algorithm verification.3. The thesis also involves the based hardware design of the micro-mouse.