Graph databases have grown in popularity in recent years as they are able to efficiently store and query complex relationships between data. Incidentally, navigation data and road networks can be processed, sampled or modified efficiently when stored as a graph. As a result, graph databases are a solution for solving route planning tasks that comes more and more to the attention of developers of autonomous vehicles. To achieve a computational performance that enables the realization of route planning on large road networks or for a great number of agents concurrently, several aspects need to be considered in the design of the solution. Based on a concrete use case for centralized route planning, we discuss the characteristics and properties of a use case that can significantly influence the computational effort or efficiency of the database management system. Subsequently we evaluate the performance of both Neo4j and ArangoDB depending on these properties. With these results, it is not only possible to choose the most suitable database system but also to improve the resulting performance by addressing relevant aspects in the design of the application.

翻译：近年来，图数据库因其能够高效存储和查询数据间的复杂关系而日益普及。导航数据与道路网络以图结构存储时，可被高效处理、采样或修改，因此图数据库作为解决路径规划任务的方案，正越来越受到自动驾驶汽车开发者的关注。为实现能够在大型道路网络或大量代理并发情况下完成路径规划的计算性能，解决方案的设计需综合考虑多个因素。基于集中式路径规划的具体用例，我们讨论了该用例中可能显著影响数据库管理系统计算效率或性能的特征与属性。随后，我们根据这些属性评估了Neo4j与ArangoDB的性能表现。这些结果不仅有助于选择最合适的数据库系统，还能通过在应用设计中针对相关因素进行优化来提升最终性能。