In this paper, a novel knowledge-based genetic algorithm for path planning of a mobile robot in unstructured complex environments is proposed, where five problem-specific operators are developed for efficient robot path planning. The proposed genetic algorithm incorporates the domain knowledge of robot path planning into its specialized operators, some of which also combine a local search technique. A unique and simple representation of the robot path is proposed and a simple but effective path evaluation method is developed, where the collisions can be accurately detected and the quality of a robot path is well reflected. The proposed algorithm is capable of finding a near-optimal robot path in both static and dynamic complex environments. The effectiveness and efficiency of the proposed algorithm are demonstrated by simulation studies. The irreplaceable role of the specialized genetic operators in the proposed genetic algorithm for solving the robot path planning problem is demonstrated through a comparison study.

翻译：在本文中,为在非结构化复杂环境中对移动机器人进行路径规划提出了一种新的基于知识的遗传算法,其中为高效的机器人路径规划开发了5个针对具体问题的操作者。拟议的遗传算法将机器人路径规划的域知识纳入到其专门操作者中,其中一些还结合了当地搜索技术。提出了对机器人路径的独特而简单的描述,并开发了一个简单而有效的路径评估方法,可以准确探测碰撞,并很好地反映机器人路径的质量。拟议的算法能够找到在静态和动态复杂环境中的近乎最佳的机器人路径。拟议的算法的效力和效率通过模拟研究得到证明。通过比较研究可以证明专门遗传操作者在解决机器人路径规划问题的拟议遗传算法中不可替代的作用。