带动态障碍区的自由区域路径实时优化问题的混合算法

发布时间：2018-06-21 00:28

本文选题：交通工程 + 滑动地平线控制　；参考：《公路交通科技》2017年11期

【摘要】：自由区域路径优化问题可以在除障碍区之外的整个区域内自由规划路径,为了解决带动态障碍区的自由区域路径实时优化问题,提出了一种遗传算法(GA)加滑动地平线控制(RHC)的混合算法。首先,建立和讨论了带动态障碍区的自由区域路径实时优化问题的数学模型。然后,详细描述了提出的遗传算法加滑动地平线策略的混合算法,阐述了混合算法中滑动地平线控制与遗传算法结合的关键步骤之一:可变长度染色体的设计。全面探讨了混合算法中滑动地平线长度的选择对于混合算法的影响,说明了滑动地平线控制策略中不同终端加权设计的路径优化效果,并通过终端加权的设计,以保证路径规划的可行性和优化性能。仿真结果表明,遗传算法(GA)加滑动地平线控制(RHC)的混合算法非常有效,在确定性的环境条件下,获得与现有GA算法几乎相同的求解性能,而在动态和不确定的环境下,新算法则取得了更佳的求解效果。在这两种情况下,带RHC的混合算法的在线计算时间是单纯GA算法的一小部分。
[Abstract]:In order to solve the problem of free region path optimization with dynamic obstacles, the free region path optimization problem can be freely planned in the whole region except for the obstacle area, in order to solve the problem of real-time optimization of free region path with dynamic obstacle area. A hybrid genetic algorithm (GA) with sliding horizon control (RHC) is proposed. Firstly, the mathematical model of real time path optimization in free region with dynamic obstacle is established and discussed. Then, the hybrid algorithm of genetic algorithm and sliding horizon strategy is described in detail. One of the key steps of the combination of sliding horizon control and genetic algorithm in hybrid algorithm is described: the design of variable length chromosome. The influence of the choice of sliding horizon length in the hybrid algorithm on the hybrid algorithm is discussed. The path optimization effect of different terminal weighted design in sliding horizon control strategy is explained, and through the design of terminal weighting, To ensure the feasibility of path planning and optimal performance. The simulation results show that the hybrid genetic algorithm (GA) and sliding horizon control (RHC) algorithm are very effective. In deterministic environment, the solution performance of GA algorithm is almost the same as that of the existing GA algorithm, but in dynamic and uncertain environment. The new algorithm achieves better results. In both cases, the online computing time of the hybrid algorithm with RHC is a small part of the simple GA algorithm.
【作者单位】：北京师范大学地表过程与资源生态国家重点实验室;浙江省标准化研究院;交通运输部公路科学研究院;北京师范大学减灾与应急管理研究院;
【分类号】：U491

【相似文献】