论文标题
Multion：使用多对象导航进行基准测试语义映射存储器
MultiON: Benchmarking Semantic Map Memory using Multi-Object Navigation
论文作者
论文摘要
光真逼真的3D环境中的导航任务具有挑战性，因为它们需要在部分可观察性下感知和有效的计划。最近的工作表明，类似地图的内存对于长途导航任务很有用。但是，尚未进行对地图对不同复杂性导航任务的影响的重点调查。我们提出了多个任务，该任务需要在现实环境中导航到特定于情节的对象序列。 Multion概括对象导航任务，并明确测试导航代理定位先前观察到的目标对象的能力。我们执行一组多和实验，以检查各种导航任务复杂性的各种代理模型的性能。我们的实验表明：i）导航性能会随着任务复杂性的不断升级； ii）一个简单的语义图代理相对于更复杂的神经图像特征映射代理表现出令人惊讶的表现； iii）即使是甲骨文地图代理的性能相对较低，这表明使用地图在训练体现的导航剂中有可能进行工作。视频摘要：https：//youtu.be/yqtlhnicgny
Navigation tasks in photorealistic 3D environments are challenging because they require perception and effective planning under partial observability. Recent work shows that map-like memory is useful for long-horizon navigation tasks. However, a focused investigation of the impact of maps on navigation tasks of varying complexity has not yet been performed. We propose the multiON task, which requires navigation to an episode-specific sequence of objects in a realistic environment. MultiON generalizes the ObjectGoal navigation task and explicitly tests the ability of navigation agents to locate previously observed goal objects. We perform a set of multiON experiments to examine how a variety of agent models perform across a spectrum of navigation task complexities. Our experiments show that: i) navigation performance degrades dramatically with escalating task complexity; ii) a simple semantic map agent performs surprisingly well relative to more complex neural image feature map agents; and iii) even oracle map agents achieve relatively low performance, indicating the potential for future work in training embodied navigation agents using maps. Video summary: https://youtu.be/yqTlHNIcgnY
