DeepNav: Learning to Navigate Large Cities
This work addresses urban navigation for autonomous systems or applications, but it is incremental as it builds on existing CNN and A* search techniques with a new dataset.
The authors tackled the problem of navigating large cities using street-view images by developing DeepNav, a CNN-based algorithm that learns to make navigation decisions at intersections, and showed it outperforms previous methods using hand-crafted features and SVR.
We present DeepNav, a Convolutional Neural Network (CNN) based algorithm for navigating large cities using locally visible street-view images. The DeepNav agent learns to reach its destination quickly by making the correct navigation decisions at intersections. We collect a large-scale dataset of street-view images organized in a graph where nodes are connected by roads. This dataset contains 10 city graphs and more than 1 million street-view images. We propose 3 supervised learning approaches for the navigation task and show how A* search in the city graph can be used to generate supervision for the learning. Our annotation process is fully automated using publicly available mapping services and requires no human input. We evaluate the proposed DeepNav models on 4 held-out cities for navigating to 5 different types of destinations. Our algorithms outperform previous work that uses hand-crafted features and Support Vector Regression (SVR)[19].