本资料是关于寻找最佳路径蚁群优化算法Unreal Engine游戏素材资源，大小：33 MB，使用软件：Unreal Engine 4.26 – 4.27,语言：英语。
蚁群优化算法 (ACO) 是一种用于解决计算问题的概率技术，可以简化为通过图找到好的路径。人造蚂蚁代表受真实蚂蚁行为启发的多智能体方法。它有助于解决各种任务，包括车辆路由、互联网路由 [Wiki]。
另一方面，通过蛮力找到一条最佳路径需要大量的计算机能力：(n-1)!/2 个变体进行比较，其中 n 是代理的数量。对于 15 个点，需要比较 43,589,145,600 条路径。
使用此插件将能够在不通知的情况下找到相同数量的点的路线。
视频：
使用 Ant 算法在不断变化的环境中寻找最优路径链接；
变化的障碍物如何影响路径，使用蚂蚁算法链接计算；
移动平台上圆形物体的蚂蚁算法：链接；
蚂蚁算法表示由不同路径组成的埃菲尔铁塔：链接；
圆形对象链接的蚂蚁算法运行时路径计算；
使用 Unreal Engine 4 中的编辑器使用 Ant 算法绘制路径链接。
技术细节
特点：
为场景中的演员寻找最佳路径；
非常适合NPC的巡逻路径自动化；
使用所见即所得概念在编辑器中编辑路径；
适应移动目标的路径；
多线程计算不会冻结游戏线程；
包含的自动化测试将帮助您根据需要修改求解器。
蓝图数量：1
C++ 类数量：6
网络复制：不
支持平台：Win64
文档：教程
示例项目：4.26、4.27
注意：此插件对其可以解决的路径大小有限制。测试了15分，结果稳定。
The ant colony optimization algorithm (ACO) is a probabilistic technique for solving computational problems which can be reduced to finding good paths through graphs. Artificial ants stand for multi-agent methods inspired by the behavior of real ants. It helps to solve various tasks, including vehicle routing, internet routing [Wiki].
Finding an optimal path with brute force on the other hand would require a lot of computer power: (n-1)!/2 variant to compare, where n is the number of agents. For 15 points it would require to compare 43,589,145,600 paths.
With this plugin will be able to find the route for the same amount of points without notice.
Videos:
Finding an optimal path in changing environment using Ant Algorithm link;
How changing obstacles affects path, calculated using Ant Algorithm link;
Ant Algorithm in action for round objects on a moving platform: link;
Ant Algorithm represents Eiffel Tower made of different paths: link;
Ant Algorithm runtime path calculations for round objects link;
Drawing a path with Ant Algorithm using the editor in Unreal Engine 4 link.
Technical Details
Features:
Finding an optimal path for actors in the scene;
Great for NPC’s patrolling path automation;
Editing path in the editor with WYSIWYG concept;
Adapting path to moving targets;
Multithreaded calculations will not freeze the game thread;
Included Automation tests will help you to modify the solver for your needs if you want to.
Number of Blueprints: 1
Number of C++ Classes: 6
Network Replicated: No
Supported Platforms: Win64
Documentation: Tutorial
Example Project: 4.26, 4.27
Notes: This plugin has limits on the size of the path it can solve. It was tested for 15 points with stable results.
