改进蚁群算法在无人艇路径规划中的应用

Application of Improved Ant Colony Optimization Algorithm in Path Planning of Unmanned Surface Vessels

为了更好地完成反潜作战、水面作战等军事任务,水面无人艇需要具备灵活自主的障碍规避能力、适应环境和任务变化的应变能力以及自主有效的路径规划能力。目前,蚁群优化算法经常被引入到无人艇路径规划问题中进行求解,但算法存在易陷入局部最优值、在迭代过程中不易收敛等弊端,此外,随着实际路径规划问题的维数不断增加,蚁群算法在迭代过程中可能陷入停滞状态,导致求解过程被迫中断。为了优化性能,提出一种改进的蚁群算法(BACO算法),将蚁群算法与贝叶斯网络相结合,并引入最大关联长度参数,根据构建的贝叶斯网络结构对传统的转移概率公式进行改进;同时为了提高算法的收敛性,改变了信息素浓度的更新策略。最后,在三种维度的栅格化地图中进行无人艇路径规划应用的仿真实验,实验结果表明,相较于传统蚁群优化算法,BACO算法具有更好的有效性和更稳定的收敛性。

In order to better complete anti-submarine operations,surface operations and other military tasks,unmanned surface vessels need to have flexible and autonomous obstacle avoidance ability,adaptability to changes in the environments and tasks,and independent and effective path planning ability.At present,ant colony optimization algorithm is often introduced to solve the path planning problem of unmanned surface vessels,but the algorithm is easy to fall into the local optimal value,and is not easy to converge in the iterative process.In addition,with the increasing dimension of the path planning problem in reality,ant colony optimization algorithm may fall into a stagnant state in the iterative process,resulting in the interruption of the solution process.In order to optimize the performance,an improved ant colony optimization algorithm(BACO algorithm)is proposed in this paper,which combines the ant colony optimization algorithm with Bayesian network,and introduces the maximum correlation length parameter.According to the Bayesian network structure,the traditional transition probability formula is improved.At the same time,in order to improve the convergence performance of the algorithm,the pheromone concentration update strategy is changed.Finally,simulation experiments are carried out in three rasterized maps with different dimensions.The experimental results show that BACO algorithm has higher effectiveness and more stable convergence than traditional ant colony optimization algorithm.