摘要
求解约束满足问题的修改算法从初始的有冲突的完整解出发，不断修改现有的变量赋值，从而得到无冲突的完整解．该法充分利用了当前解所提供的信息，在这方面优于不断扩大部分解最终形成完整解的生长法．但是生长法多年来发展了很多强有力的启发式方法以提高效率，这是修改法所不及的．本文将启发式方法应用于修改型算法。
The repair based method for solving constraint satisfaction problems starts from the initial solution with conflicts, keeps repairing the current assignments, and finally gets the solution without conflicts. This method has an advantage over the constructive one,with which the solution is obtained through incrementally extending consistent partial assignments in the usage of the information provided by the current assignments.But the constructive method can improve efficiency with various powerful heuristics developed in many years. In this paper, an efficient repair based algorithm for constraint satisfaction problem is developed through the application of heuristics.
出处
《计算机研究与发展》
EI
CSCD
北大核心
1997年第2期93-98,共6页
Journal of Computer Research and Development
关键词
人工智能
约束满足
搜索算法
artificial intelligence, constraint satisfaction, search algorithms
