Generating Landmark Navigation Instructions from Maps as a Graph-to-Text Problem

TL;DR

This paper introduces a neural model that converts map graphs into human-like landmark-based navigation instructions, improving natural language guidance for navigation tasks.

Contribution

It presents a novel graph-to-text model trained on a new dataset, generating landmark-based instructions from map representations for navigation.

Findings

Generated instructions are similar to human instructions.

Instructions successfully guide humans in Street View navigation.

Model achieves effective landmark-based navigation guidance.

Abstract

Car-focused navigation services are based on turns and distances of named streets, whereas navigation instructions naturally used by humans are centered around physical objects called landmarks. We present a neural model that takes OpenStreetMap representations as input and learns to generate navigation instructions that contain visible and salient landmarks from human natural language instructions. Routes on the map are encoded in a location- and rotation-invariant graph representation that is decoded into natural language instructions. Our work is based on a novel dataset of 7,672 crowd-sourced instances that have been verified by human navigation in Street View. Our evaluation shows that the navigation instructions generated by our system have similar properties as human-generated instructions, and lead to successful human navigation in Street View.