安徽工程大学学报

针对传统路径规划A^*算法搜索效率低、路径拐点多以及所得路径不够平滑等问题，提出了一种改进双向A^*路径规划的算法。首先，调整A^*算法的整体搜索方向为双向搜索，初步提升搜索速度；其次，在规划过程中引入动态权重系数来调节启发式函数，通过减少搜索领域来提高搜索效率，平衡路径长短与规划速度之间的关系进一步提高搜索速度；最后，采用三阶贝塞尔曲线对所规划的路径进行平滑优化，解决A^*算法规划路径转角过多无法满足移动机器人实际运动控制的问题。仿真实验结果表明，相比传统的A^*算法，该算法在路径拐点个数方面有极大减少，搜索节点和规划时间方面分别平均减少74.29%和86.23%。

Addressing the issues of low search efficiency,numerous path inflections,and lack of smoothness in the paths generated by the traditional A^* path planning algorithm,an improved bidirectional A^* path planning algorithm is proposed.First,the overall search orientation of the A^* search algorithm is adjusted to a bidirectional search,initially increasing the search speed.Secondly,dynamic weight coefficients are introduced in the planning process to adjust the heuristic function and the search domain is reduced to improve search efficiency.By balancing the relationship between path length and planning speed,the search speed is further enhanced.Lastly,a cubic Bézier curve is employed to smooth the planned path,resolving the issue of excessive path turns which do not meet the actual movement control requirements of mobile robots.Simulation results demonstrate that compared to traditional A^* algorithms,this algorithm significantly reduces the number of path inflections,and the number of search nodes and planning time are reduced by an average of 74.29% and 86.23%,respectively.