Automated driving in urban scenarios requires efficient planning algorithms able to handle complex situations in real-time. A popular approach is to use graph-based planning methods in order to obtain a rough trajectory which is subsequently optimized. A key aspect is the generation of trajectories implementing comfortable and safe behavior already during graph-search while keeping computation times low. To capture this aspect, on the one hand, a branching strategy is presented in this work that leads to better performance in terms of quality of resulting trajectories and runtime. On the other hand, admissible heuristics are shown which guide the graph-search efficiently, where the solution remains optimal.
翻译:城市情景的自动驱动需要能够实时处理复杂情况的高效规划算法,一种流行的做法是使用基于图表的规划方法,以便获得一种粗略的轨迹,这种轨迹随后得到优化。一个关键方面是生成在图形搜索期间已经实施舒适和安全行为的轨迹,同时使计算时间保持低速。一方面,为了抓住这一方面,在这项工作中提出了分流战略,从而在结果的轨迹和运行时间的质量方面实现更好的表现。另一方面,展示了可接受的超常性,以高效地指导图形搜索,而解决办法仍然是最佳的。