Hierarchically Integrated Models: Learning to Navigate from Heterogeneous Robots
This work addresses the scalability issue of data collection for mobile navigation in robotics, offering a method to use diverse datasets from different platforms, though it is incremental as it builds on existing hierarchical and model-based approaches.
The paper tackles the problem of leveraging heterogeneous datasets from multiple robotic platforms for deep reinforcement learning in mobile navigation, proposing HInt, which learns separate perception and dynamics models and integrates them hierarchically, resulting in outperformance over conventional hierarchical policies and single-source approaches.
Deep reinforcement learning algorithms require large and diverse datasets in order to learn successful policies for perception-based mobile navigation. However, gathering such datasets with a single robot can be prohibitively expensive. Collecting data with multiple different robotic platforms with possibly different dynamics is a more scalable approach to large-scale data collection. But how can deep reinforcement learning algorithms leverage such heterogeneous datasets? In this work, we propose a deep reinforcement learning algorithm with hierarchically integrated models (HInt). At training time, HInt learns separate perception and dynamics models, and at test time, HInt integrates the two models in a hierarchical manner and plans actions with the integrated model. This method of planning with hierarchically integrated models allows the algorithm to train on datasets gathered by a variety of different platforms, while respecting the physical capabilities of the deployment robot at test time. Our mobile navigation experiments show that HInt outperforms conventional hierarchical policies and single-source approaches.