Hierarchies of Planning and Reinforcement Learning for Robot Navigation

TL;DR

This paper introduces a hierarchical reinforcement learning framework with a trainable planning policy for robot navigation, improving adaptability and performance over traditional methods, especially in complex dynamic environments.

Contribution

It proposes a novel VI-RL planning policy that learns environment dynamics, enhancing hierarchical RL for navigation tasks with complex robot behaviors.

Findings

VI-RL outperforms vanilla RL in simulated navigation tasks.

VI-RL matches hierarchical RL on single layouts and generalizes better across multiple layouts.

Shows significant improvements in non-holonomic parking scenarios.

Abstract

Solving robotic navigation tasks via reinforcement learning (RL) is challenging due to their sparse reward and long decision horizon nature. However, in many navigation tasks, high-level (HL) task representations, like a rough floor plan, are available. Previous work has demonstrated efficient learning by hierarchal approaches consisting of path planning in the HL representation and using sub-goals derived from the plan to guide the RL policy in the source task. However, these approaches usually neglect the complex dynamics and sub-optimal sub-goal-reaching capabilities of the robot during planning. This work overcomes these limitations by proposing a novel hierarchical framework that utilizes a trainable planning policy for the HL representation. Thereby robot capabilities and environment conditions can be learned utilizing collected rollout data. We specifically introduce a planning policy based on value iteration with a learned transition model (VI-RL). In simulated robotic navigation tasks, VI-RL results in consistent strong improvement over vanilla RL, is on par with vanilla hierarchal RL on single layouts but more broadly applicable to multiple layouts, and is on par with trainable HL path planning baselines except for a parking task with difficult non-holonomic dynamics where it shows marked improvements.