MetricNet: Recovering Metric Scale in Generative Navigation Policies
This addresses safety and efficiency issues for robots using generative navigation policies, though it is incremental as an add-on to existing methods.
The paper tackled the problem of generative navigation policies lacking metric scale and being short-sighted, resulting in unsafe actions; by proposing MetricNet to predict metric distances between waypoints, it improved navigation and exploration performance in simulation and real-world experiments.
Generative navigation policies have made rapid progress in improving end-to-end learned navigation. Despite their promising results, this paradigm has two structural problems. First, the sampled trajectories exist in an abstract, unscaled space without metric grounding. Second, the control strategy discards the full path, instead moving directly towards a single waypoint. This leads to short-sighted and unsafe actions, moving the robot towards obstacles that a complete and correctly scaled path would circumvent. To address these issues, we propose MetricNet, an effective add-on for generative navigation that predicts the metric distance between waypoints, grounding policy outputs in real-world coordinates. We evaluate our method in simulation with a new benchmarking framework and show that executing MetricNet-scaled waypoints significantly improves both navigation and exploration performance. Beyond simulation, we further validate our approach in real-world experiments. Finally, we propose MetricNav, which integrates MetricNet into a navigation policy to guide the robot away from obstacles while still moving towards the goal.