Mapless Humanoid Navigation Using Learned Latent Dynamics

TL;DR

This paper introduces a deep reinforcement learning method for humanoid robots to navigate without maps by planning in a learned latent space, integrating visual and non-visual data, and predicting terminal states to improve efficiency.

Contribution

It presents a novel approach combining learned latent dynamics, multi-modal observations, and termination prediction for mapless humanoid navigation.

Findings

Effective navigation in simulation and real-world scenarios.

Improved sample efficiency through termination likelihood prediction.

Successful collision avoidance with the NimbRo-OP2X robot.

Abstract

In this paper, we propose a novel Deep Reinforcement Learning approach to address the mapless navigation problem, in which the locomotion actions of a humanoid robot are taken online based on the knowledge encoded in learned models. Planning happens by generating open-loop trajectories in a learned latent space that captures the dynamics of the environment. Our planner considers visual (RGB images) and non-visual observations (e.g., attitude estimations). This confers the agent upon awareness not only of the scenario, but also of its own state. In addition, we incorporate a termination likelihood predictor model as an auxiliary loss function of the control policy, which enables the agent to anticipate terminal states of success and failure. In this manner, the sample efficiency of the approach for episodic tasks is increased. Our model is evaluated on the NimbRo-OP2X humanoid robot that navigates in scenes avoiding collisions efficiently in simulation and with the real hardware.