This paper deals with a dynamical laser antimissile problem. The laser antimissile system can destroy many targets in a short time, the attacking sequence plays an important role in practical implement. This laser antimissile problem can be seen as a dynamical traveling salesman problem. This paper introduces an enhancing algorithm to solve this problem based on three exchange strategies - the near exchange, the symmetric exchange and the near block exchange. The near exchange strategy searches for a local optimal sequence amongst its neighbors, obtaining a local optimal solution fast. The symmetric exchange will search in a larger region so as to jump out from the local optimal solution. The near block exchange expands the search region. The combination of the three exchange strategies increases the probability of finding the global optimal solution. We consider a case where there are eight attacking targets. The results show that the convergence is very fast and the local optimal solution is, indeed, quite close to the global one.
展开▼
